2. The case for providing tax concessions for philanthropy

While philanthropy plays an important role in most countries, this does not automatically mean that it justifies support through the tax system.1 This chapter examines the different rationales for and against providing tax incentives for philanthropy2 domestically and internationally. It considers both: (1) tax concessions for philanthropic entities; and (2) tax incentives for giving to philanthropic entities. The chapter highlights that there is no single generally accepted rationale for preferential tax treatment of philanthropic entities. Economic theory provides a limited rationale for providing tax concessions for philanthropy (potentially both for entities and giving) where there is under-provision of a public good or where there are positive externalities associated with philanthropic activity. In this regard, tax concessions will be justified if they result in a larger increase in social welfare than that which government could have otherwise achieved through direct spending. Legal scholars frequently refer to this as the subsidy rationale.

Another often articulated argument for exempting philanthropic entities from income tax is the “base defining” rationale which argues that the surplus of a philanthropic entity is different in nature to income and therefore beyond the scope of the income tax base. Additional arguments include that philanthropic giving, as well as the institutions it develops, strengthen civil society, and decentralise decision-making, and are thus an important feature of a democratic society and worth supporting.

A number of arguments have been raised against the provision of tax preferences for philanthropic entities and/or giving. The cost of providing concessions is often highlighted as a concern. By reducing government revenue, tax concessions for philanthropy require other taxpayers to bear an increased tax burden (or alternatively result in less government expenditure on other policy priorities). A concern regarding exemption of commercial income of philanthropic entities is that this may create an unfair competitive advantage for philanthropic entities over for-profit businesses.

Two related concerns that are raised regarding tax incentives for giving are that they may be regressive and undemocratic. Tax incentives may be regressive in that higher income taxpayers benefit from a larger tax incentive than lower income taxpayers. This can be the case in both aggregate terms, but also in proportionate terns as a tax deduction will provide a greater benefit to higher income taxpayers where they are subject to higher marginal tax rates than lower income taxpayers. The democratic argument highlights the concern that, as a tax incentive effectively reallocates tax revenue towards the favoured philanthropic entity, higher income taxpayers that make larger donations benefit from a disproportionate influence in the determination of how tax revenue is spent. This may be of particular concern where the priorities of donors are not consistent with those of society in general.

Irrespective of the arguments for and against tax concessions, most countries do provide tax incentives for giving, and in general provide exemptions from some taxes for philanthropic entities. The design of these tax concessions are examined in detail in the subsequent chapters of this report.

This chapter is organised as follows: Section 2.2 and 2.3 provide, respectively, an overview of the rationales for and against the provision of tax concessions for philanthropy in a domestic context. Section 2.4 then considers arguments for and against tax concessions for cross-border philanthropy.

A range of arguments can be made in favour of the provision of tax concessions for philanthropic entities and for tax incentives for giving to such entities. This section first considers arguments from economic theory that point to a potential market failure rationale for the subsidisation of both philanthropic entities and philanthropic giving. It then summarises a number of broader arguments drawing on legal, accounting and philosophical perspectives.

The section first outlines two economic theory-based rationales for government intervention to subsidise philanthropy: the under-provision of a public good; and the presence of positive externalities. It then considers whether such a subsidy, if warranted, should be provided via direct grants or via tax concessions (to philanthropic entities and/or philanthropic giving). Finally, it discusses the various trade-offs that must be made in determining the optimal level of a tax incentive for philanthropic giving.

The under-provision of public goods rationale requires three “failures” to occur to justify government subsidisation of philanthropy: “market failure”, “government failure”, and “voluntary failure” (Hansmann (1987[1]), Weisbrod (1975[2]) and Salamon (1987[3]) (2016[4]). A “market failure” case will exist for government to intervene and provide public goods that would be welfare improving, but that, due to their non-rival non-excludable nature, are not provided by the market. However, in some cases “government failure” may also occur where the government does not, or is unable to, provide (or unable to provide at a welfare maximising level) the public good. In such cases, philanthropic entities can play an important role in providing these public goods. However, “voluntary failure” may also occur in the sense that philanthropic entities provide an inefficiently low level of the public good, for example, due to insufficient resourcing.

In the presence of these three failures, there is a case for the government to subsidise the philanthropic activity in order to increase supply of the public good to the social optimal level. This subsidisation could occur via a tax incentive for giving, tax concessions to the philanthropic entities themselves, or direct grants to these entities.

While not providing a public good in the technical sense of a non-rival non-excludable good, a philanthropic entity may provide goods and services that produce positive externalities that are not fully captured by the entity itself or by those contributing to the entity. The presence of externalities may justify government intervention to correct the market failure. In the case of negative externalities, the intervention generally consists of a tax. In the case of positive externalities, on the other hand, the intervention may consist of a subsidy which could take the form of a tax incentive for giving, a tax concession for the entity itself, or a government grant.

It is often argued that philanthropic activity may be viewed as having consumption externalities. To the extent that the private marginal benefit of a gift to a philanthropic entity (i.e. the donor’s “warm glow” – see Box 2.1) is below the social marginal benefit of that gift, philanthropic giving has positive consumption externalities and may be ‘under-consumed.’ Although views in the literature differ to which extent this argument could justify tax subsidies for giving to philanthropic entities (in particular because the argument would provide a justification to subsidise giving to, for instance, other family members, but tax systems typically do not provide a tax subsidy for this type of giving).’

Therefore, to internalise the externality and correct the market failure, there may be a case for government to intervene. This intervention could occur via tax concessions or direct grants to the philanthropic entity. In the case of tax concessions to philanthropic entities, reducing the taxes borne will (directly or indirectly) lower the private marginal cost of producing the goods and services, which can increase the provision of these goods towards the social optimal level. In the case of philanthropic giving, a tax incentive lowers the price of giving so that the private marginal benefit of the donor increases towards the social marginal benefit, thereby increasing the level of giving towards the social optimal level.

The public good and positive externality arguments provide a case for government subsidisation of philanthropic entities and giving. However, as alluded to in the preceding discussion, an alternative to providing tax concessions to subsidise philanthropic activities, is for government to provide direct grants to a philanthropic entity. A grant may be preferable to a tax concession where the government wants greater control regarding the destination of the government support, where the level of “crowding out” of private contributions is low, or where a tax concession is not “treasury efficient” – that is, where it would result in a smaller increase in funding of the philanthropic entity than the tax revenue forgone. Equally, when government grants largely crowd out philanthropic giving, tax concessions may be preferable to government grants, even when tax incentives for philanthropic giving are not treasury efficient. More generally, a tax incentive may still be welfare increasing even if it is not treasury efficient if the benefit to society of the activity funded by the giving is sufficiently large.

The economic literature has focused on two key factors that may influence whether government grants or tax concessions are preferable: crowding out of private contributions and the treasury efficiency of a tax incentive. For the crowding out effect, the hypothesis is that since government grants are financed through taxes, taxpayers will be less inclined to donate to a philanthropic entity that has already received their tax dollars (Andreoni and Payne, 2003[8]). Research suggests that although government support for a non-profit entity might influence private donations, it is unlikely to fully ‘crowd-out’ private giving (references). There is, in fact, some support for the opposite conclusion, namely that government support for a philanthropic entity may be a signal of the entity’s quality, resulting in a crowding-in effect. A variation on the notion of crowding-out is that government grants may discourage an entity from fundraising and that this might then lead to a decline in private support (Andreoni and Payne, 2003[8]).

Whether a tax incentive for philanthropic giving is “treasury efficient” is typically examined by empirical estimation of the price elasticity of philanthropic giving – with an elasticity greater than one indicating the tax incentive is treasury efficient.3 More generally, the issue of whether philanthropic giving is responsive to tax incentives lowering the price of giving, has prompted significant debate in the econometric literature, mostly based on data from the United States, but more recently also on European data.

A major review by Clotfelter in 1985 found a notable consistency in the findings, with the consensus being that the price elasticity for the population of taxpayers was probably greater than -1, with a range of -0.9 to -1.4. As well, it was observed that the price elasticity appeared to rise with income; there are substantial lags in giving behaviour; and there is little effect of ‘crowding out’ individual contributions through government contributions (Clotfelter, 1985[9]). While there were tax effects on corporate giving, this appeared to be less than for individual contributions, and there was also evidence that corporations time-shift their donations (Clotfelter, 1985[9]).

The literature has examined both the ‘price effects’ (including tax rates), which influence the cost of giving, and the ‘income effects’, such as inflation or economic growth, that affect the income available for philanthropic giving (Clotfelter and Salamon, 1982[10]). Different methodologies used in other studies reported much lower price elasticities (e.g. Steinberg (1990[11]) and Randolph (1995[12])). Nevertheless, an analysis in 2005 of 40 years of research in this field concluded that tax deductions were treasury efficient, and (surprisingly) that the price elasticity was not significantly higher for high-income earners (Peloza and Steel, 2005[13]).

A more recent paper by Backus and Grant (Backus and Grant, 2019[14]) noted that results varied depending on whether studies were based on tax return data of individuals who itemise their deductions (a group substantially wealthier than the average taxpayer), or were based on general population survey data. Backus and Grant concluded that the top 10% of income earners had an elasticity of at least -1, but middle-income taxpayers were less sensitive (see also Fack and Landais (2010[15]), Bönke et. al. (2013[16]), and Bönke and Werdt (2015[17])).4

When there is a rationale for government intervention in the form of a tax concession for philanthropic entities or a tax incentive for philanthropic giving to these entities, determining the optimal subsidy level for philanthropic giving is complicated and involves various trade-offs. On the one hand, there are the welfare gains from increasing the provision of the public good or the externality generating activity. On the other, there is the opportunity cost in terms of what the tax revenue that would have otherwise been collected could have been used for (acknowledging the distortionary impact of taxation). Consideration must also be given to the distributional impact of the tax concession. In particular, if the benefit (e.g. the additional warm glow, or the reduced cost of generating the same warm glow) of the tax concession is primarily enjoyed by individuals at the top of the income distribution, this may conflict with the underlying redistributive preferences of government. That said, if the resulting increase in the philanthropic activity primarily benefits lower-income households this will aid redistribution goals. Moreover, tax concessions can only be enjoyed if individuals give away part of their income or wealth to a philanthropic activity, and the mere fact of giving by the rich will reduce income and wealth inequality, irrespective of the design of the tax concession.

The optimal tax literature has attempted to incorporate these trade-offs into a single welfare-maximising framework (see Box 2.2 for more detail on optimal taxation of philanthropic giving). In this regard, Saez (2004[18]) models the optimal tax rate for private contributions to a public good in the presence of warm-glow effects, externalities, crowding out, and the redistributive preferences of government. Subject to a number of strong assumptions, the model suggests that the optimal tax subsidy for philanthropic giving:

  • increases with the size of the external effect of a marginal increase in the level of contributions;

  • increases with the responsiveness of the donor to the subsidy;

  • increases with the level at which public contributions crowd out private contributions;

  • and decreases with the proportion of giving made by high-income individuals (assuming government values redistribution).

The optimal tax subsidy itself will depend on the interaction of all these factors. In the model, the optimal subsidy rate is found to decrease with the proportion of giving made by high-income individuals because government is placing a lower weight on the utility that higher income individuals derive from contributing as compared to lower income individuals. The model also suggests that the optimal tax subsidy for philanthropic activity does not necessarily have to be linked to the personal income tax rate schedule, which is the case in countries where contributions are deductible from the personal income tax base.

Base-defining theories aim to identify what is properly taxable e.g. as income or profit. This approach recognises that some revenue of philanthropic entities may not be appropriately included in the tax base. These theories assert that income tax can only logically be levied on activities undertaken for profit (see for example, Bittker and Rahdert (1976[23])). For example, for many philanthropic entities, a not insignificant portion of revenue will comprise contributions or membership payments which may not fall within notions of ‘income’. Similarly, many expenses incurred in operating the philanthropic entity e.g. the Red Cross providing relief after a disaster, may not fit within notions of deductible expenditure as ordinary and necessary business expenses. There will, of course be some philanthropic entities that derive most of their income from ‘business’ activities, e.g. hospitals or universities, and it would be relatively easy to calculate taxable income.

Brody (1999[24]) proposes what she describes as a ‘sovereignty view’ as a variation on the base defining approach. She argues that charities go untaxed because ‘Caesar should not tax god’ (or its modern secular equivalent). Brody acknowledges that although this might have something to do with the role of religion in early charity, its continued existence can be justified based on the independence of the sector. She argues that a sovereignty view also explains why a subsidy would take the form of tax exemption rather than more logical form of direct grants: for all its imperfections, tax exemption keeps governments out of charities day to day business and keeps charities out of the business of petitioning government for subvention.

Another variation on the tax base theory is that taxation of corporations is sometimes viewed as a proxy for taxation of shareholders, and the philanthropic entity will generally not have shareholders (Rushton, 2007[25]) (Buckles, 2005[26]). Rather those who are beneficiaries of the activities of the philanthropic entity, would not be viewed as appropriate subjects of taxation.

Fleischer (2018[27]) has put forward an alternative to the traditional theories. She argues that support for philanthropy can be justified based on what she terms ‘two bedrock principles of Western liberal democracies’ namely: limited government and equal opportunity. She argues that the charitable tax subsidies reflect these principles, as expressed in the two theories of distributive justice respectively associated with them, libertarianism and resource egalitarianism. However, she acknowledges problems that may need to be tempered. In her view, the tax subsidies may undermine the principle of limited government by coercing taxpayers to subsidise activities that are not the legitimate purview of government. The subsidies’ relation to resource egalitarianism is more complex: she argues that tax subsidies may undermine basic equality of opportunity notions both by subsidising activities that increase the head-start of the wealthy and by giving wealthy taxpayers more say over government resources than poorer taxpayers.

Although there were earlier critics of the tax concessions,5 the case against the tax concessions was put most powerfully in the 1960s and 1970s by US scholars, most notably Kahn (1960[28]), and Rabin (1966[29]), in the context of the charitable contribution deduction. The arguments of the critics are considered below. Some, but not all of these arguments can be used to critique all tax concessions for philanthropic entities.

The starting point is that tax concessions have a ‘cost’, that is, they reduce government revenue, and therefore shift the tax burden to other taxpayers. This is relatively uncontroversial. More controversial, however, is a second related argument that the loss in revenue amounts to a ‘tax expenditure’. Tax expenditure analysis distinguishes between tax measures which seek to achieve the primary goal of income taxation and those (‘tax expenditures’) which reduce tax liability to support social or economic objectives. Tax expenditure analysis treats tax exemptions and concessions as government subsidies and evaluates them in the same way as direct expenditures. Tax expenditure analysis has its modern genesis in a seminal US Treasury analysis in 1968 (1968[30]) and subsequent explanations of its implications by Assistant Secretary of the United States Treasury, Stanley Surrey (1970[31])

It is necessary to consider whether tax concessions for philanthropy are, in fact, tax expenditures. This is likely to be the case for property tax exemptions. It is also relatively, although not universally, accepted in the case of the charitable contribution deduction or credit, since the outgoing does not fall into either the recognised category of expenses in the production of income, or expenses that are in a legal or moral sense necessary or involuntary. There has been significantly more contest over whether the income tax exemption (or other relief) is a tax expenditure or could be justified by principles of income taxation (‘base-defining theories’) (Brody, 1999[24]) as discussed above.

There are a number of issues relating to the notion of tax expenditure analysis in relation to the philanthropic sector:

  • Tax expenditure analysis compares the current or prospective tax treatment of taxpayers who receive the concession to a ‘benchmark’ treatment. (Andrews, 1972[32]) For example, in relation to the gift concession it assumes that the same amount would be donated even without the concession. That is, it assumes that taxpayer behaviour is unchanged and for that reason may not accurately reflect revenue foregone.

  • There is some debate about whether the tax exemptions for non-profits are, in fact, tax expenditures. For example, much of the revenue received by philanthropic entities would not be income in most countries, such as donations and government grants. In that sense the exemption may not be a concession in relation to that revenue (Krever, 1991[33]) . There is also an argument, discussed above, that philanthropic entities may be ‘outside’ the system for taxing corporations.

  • There are also concerns about the reliability and accuracy of the tax expenditure statements. For example, tax expenditure estimates are only concerned with statutory provisions and do not take into account situations where income is not taxed for some other reason e.g. because of the common law principle of ‘mutuality’ or because of the exercise of an administrative discretion (Burton and Sadiq, 2013[34]).

  • Perhaps most significantly, there is simply insufficient data available to quantify the tax expenditures in relation to the philanthropic sector across jurisdictions. There are a number of reasons for this: many countries do not calculate tax expenditures – of the countries responding to the survey, less than half were able to provide estimates of revenue foregone in relation to either gift concessions or exemptions from other taxes. In addition, information to estimate the amount of tax that might be payable is often not available e.g. if the philanthropic entity is not required to lodge a tax return, it will not be possible to quantify the amount of revenue foregone. Several countries that did provide responses to the questions relating to the cost of concessions, were only able to provide estimates, suggesting either that it is too soon to provide the data or that the data cannot be accurately identified.

What does appear from the responses provided to our survey is that there is an amount of tax revenue foregone, and hence the tax treatment of philanthropy is an important topic that needs to be considered in detail to ensure that the concessions are justified and well designed.

The pluralism argument considered as part of the subsidy rationale carries with it the importance of a heterogeneous and large number of donors. However, the design of tax-subsidies may lead to incentivising large donations from a small number of wealthy donors rather than smaller donations from a large number of donors.

In this regard, it has been argued that one of the main objections to the philanthropic gift concessions is the inequity that results from its regressive nature (Rabin, 1966[29]). This arises because the deduction is tied to progressive tax brackets. A progressive income tax system results in all deductions ‘benefiting’ higher income taxpayers more, and the philanthropic contribution deduction has a similar effect of reducing the ‘price of giving’ more for higher income earners. This is said to conflict with the basic premise of a progressive income tax. It is also said to be inequitable because the evidence is that higher income taxpayers favour different types of charities, typically higher education and arts and culture, than lower income taxpayers which tend to favour religion and welfare (Rabin, 1966[29]) (Atkinson, 1997[35]). This may have implications for the potentially undemocratic nature of tax concessions for giving (discussed below). Incentivising the wealthy more may increase the treasury efficiency of the tax subsidies, that is, if those with higher income levels are more responsive to tax incentives for donations, a tax-subsidy would be more efficient if it focused on big donors6

Some authors have observed that this is the effect of all tax deductions and not peculiar to the charitable contribution (Bittker, 1972[36]) Most of the proposals that seek to redesign the philanthropic gift concession attempt to minimise this regressive effect by proposing tax credits or matching schemes (Duff, 2014[37]).

It is often claimed that commercial operations run by non-profit entities have an unfair advantage when competing with for-profit organisations offering the same or similar goods and services (Brody and Cordes (2001[38]); Sharpe (1996[39])). A competitive advantage may result from tax concessions that apply to the income, inputs, or outputs of philanthropic entities, including when they operate businesses. In this context it is argued that philanthropic entities can undercut the competition. The notion of unfair competition underpinned the introduction of the unrelated business income tax (‘UBIT’) in the United States in the 1950s. Before that time, the tax system followed a ’destination of income’ approach under which income, whatever the source, could be earned tax-free if profits were dedicated to a charitable or philanthropic purpose. The introduction of the UBIT was also said to be about preserving the corporate income tax base, and to have been the result of the infamous ownership of Mueller Macaroni by New York University Law School (Brody and Cordes, 2001[38]). The UBIT operates so that to the extent that an activity is ‘substantially related’ to the entity’s tax-exempt purpose, the income is tax-free (and the associated expenses are, essentially, not deductible). By contrast, net income from ‘unrelated business activities’, is subject to the UBIT, which generally taxes such income at ordinary corporate (or trust) tax rates. Congress, however, has exempted dividends, interest, rents, and royalties from the UBIT.

It has also been argued that an income tax exemption (on all income or on income from related activities) does not provide an unfair advantage to philanthropic entities (Henry et al. (2009[40]); Steuerle (1998[41])). An income-tax exemption is not a subsidy on the cost of inputs; it does not reduce the charities cost of purchasing goods. One commentator has argued that:

the zero rate for charity is no more ‘unfair’ to a [fully] taxed competitor than are the progressive income-tax rates on individuals who conduct business activities in a sole proprietorship or through a partnership, [or a corporation]….. Nor is a non-profit organisation likely to under-price its for-profit competitor (the ‘unfair’ part of the competition), just as it would not accept a lower return on an (untaxed) passive investment (Weisbrod, 1988[42]).

The UBIT in the United States, and similar arrangements in other countries to impose tax on ‘unrelated’ commercial profits, suffer from the difficulty of trying to identify what is related and what is unrelated. It has been noted that very little revenue is in fact collected in the United States from the UBIT. However, the income tax concessions available to philanthropic entities may provide them with some advantages over for-profit firms, such as in relation to cash flow.

In response to these concerns, many countries do tax income or profits derived from commercial operations. A variety of terms are used to signify the types of income being taxed e.g., commercial, business or trading income; and a distinction is often made between commercial activities that are part of the philanthropic activities of the PBO, such as operating a school or hospital (commonly referred to as ‘related commercial income’) and activities that are not part of the philanthropic activities, other than as providing revenue to undertake those philanthropic activities (commonly referred to as ‘unrelated commercial income’). These distinctions are often difficult to make and complex to administer. Some countries either prohibit, or at least tax, commercial activities that are undertaken by for-profit competitors (see Chapter 3).

It has also been argued that concessions related to the cost of inputs, e.g., employee-related tax concessions, do provide a competitive advantage for the commercial activities of philanthropic entities compared with for-profits and that they could be distortionary because they provided an incentive for non-profits to favour the use of the inputs that attracted the concessional taxation treatment.

Distortions from VAT concessions for philanthropic entities typically arise from the exemption from VAT of the output of these entities. The distortion can result in either a competitive advantage or a competitive disadvantage to the philanthropic entity, depending on who is the recipient of the supply and what the recipient uses it for.

A VAT exemption can provide a competitive advantage to a philanthropic supplier if the recipient is a consumer or an entity that uses it as an input to the production of its own exempt supplies. This is because the total price paid by the recipient is lower than the VAT-inclusive price they would pay to a for-profit supplier.

A VAT exemption can create a competitive disadvantage for a philanthropic supplier if the recipient is an entity, such as a for-profit business, that uses the good or service as an input to a taxable supply. This is because input tax credits would allow the purchaser to fully recover any VAT paid on inputs purchased on a non-exempt basis (e.g., from another for-profit business), and the suppliers of those inputs are entitled to input tax credits on their own inputs. In contrast, the philanthropic supplier of exempt goods and services cannot recover the VAT it pays on its inputs. This VAT gets embedded into the cost of the good or service itself and is not recoverable by the purchaser.

When the exemption of the entity’s outputs causes a competitive disadvantage, input concessions may reduce the distortion. Either way, philanthropic entities that make VAT exempt supplies will tend to favour the use of inputs with VAT concessions. The VAT treatment of philanthropic entities in countries is discussed in more detail in Chapter 3.

A somewhat related argument is that tax subsidies are not subject to the same periodic review that spending programs receive. As a result, once enacted, there is no need for the recipients to justify the concession. Further, the operation of a tax concession can result in unexpected budgetary outcomes. Tax incentives are usually ‘open-ended’ – that is, they do not limit the tax benefits a taxpayer can receive. In the case of direct expenditures, if the legislator considers that certain programmed costs in a given year are too high, it can cap them in advance. However, it is often impossible to apply such restrictions to existing tax incentives: they do not require annual approval from the legislator and remain valid as long as the tax law remains unchanged (Lideikyte-Huber, 2020[43]).

Perhaps surprisingly it has been suggested that tax expenditure analysis, which aimed to highlight the ‘cost’ of providing various concessions has resulted in more ‘tax expenditures’ being included in tax legislation in many jurisdictions. One commentator has noted that the effect of producing tax expenditure statements, ‘unintended by the advocates of tax expenditure analysis, has been to legitimate and expand tax expenditures (Zelinsky, 2012[44]).

Reich (2019[45]) argues that much philanthropy is undemocratic and unaccountable. His arguments are really concerned with ‘big philanthropy’ or large private foundations, and is focussed on the United States.

Reich notes that a fundamental commitment of a democratic society is that individuals have an equal opportunity to influence politics or public policy. This is enshrined in the constitutional protections for ‘one person, one vote’. However, by operating through a private foundation, wealthy people are able to uniquely influence public policy. In other words, the power to spend money gives the philanthropist significant political power. This is not restricted to tax-preferred philanthropy e.g., Mark Zuckerberg and Priscilla Chan have established a limited liability company rather than a private foundation to undertake philanthropic activities. This means they will not be subject to IRS rules about disbursements or any reporting requirements. Furthermore, there is no guarantee that big philanthropy will relieve poverty or direct monies to reducing inequality.

Reich (2019[45]) also notes that large foundations are largely unaccountable – they do not have to account to customers or competitors and, unlike politicians, cannot be voted out of office. Reich notes that this was a matter of real concern in the 1880s when the first private foundations were mooted, and refers to the fact that Rockefeller had considerable trouble in setting up his private foundation – and that there were no tax concessions for doing so at the time. Moreover, he argues that instead of moderating the behaviour of big philanthropy, governments encourage it with tax concessions, and the charitable deduction in the US provides benefits that reinforce inequality. However, Reich does not argue that there should be no tax support, but rather that the contribution concession should be a credit rather than a deduction, so that the value of the concession does not give greater support to wealthy donors.

While not accountable to ‘customers’, foundations are accountable to the tax authorities in staying compliant with the tax rules including using funds for personal benefit and disposing of donated stock within a specified period. In the United States, for example, IRS audits periodically result in imposition of penalties, court cases or even the closing down of a foundation.

The discussion above has assumed that the donor and the recipient entity are located in the domestic or home jurisdiction and that the entity pursues its objectives domestically. The notion of cross-border philanthropy raises distinct issues for various stakeholders. Cross-border philanthropy has been defined as ‘voluntary contributions from private donors in one country to a recipient in another country’ (Moore and Rutzen, 2011[46]).

There is little comprehensive or comparable data on the extent of cross-border philanthropy, although the data that is available suggests that such giving is growing. In many cases, the discussion around cross-border philanthropy revolves around assistance for developing countries or in conflict situations. But potentially, cross-border philanthropy could relate to any of the worthy purposes identified previously. A philanthropic intermediary in Europe has reported that for the period 2010 to 2016 donations channelled related to education accounted for 42%, social matters 18%, heritage and culture 14%, health 11%, international development 11%, environment 3% and religion 1%. (European Foundation Centre, 2017[47]) There is some data on private philanthropy related to development. An OECD survey on private foundation giving for development found that the 147 foundations surveyed provided approximately USD7.96 billion per year to developing countries from 2013 to 2015, representing an average annual increase of 19% (OECD, 2018[48]). The OECD survey also noted that the sources of philanthropic giving for development are highly concentrated. Of the 143 foundations included in the data survey sample, the Bill & Melinda Gates Foundation was by far the most significant philanthropic donor, and 81% of the total philanthropic giving during 2013-15 was provided by only 20 foundations.

The increase in cross-border philanthropy has given rise to concerns in both donor and recipient countries. Donor countries are concerned that the money donated may be diverted for the purposes of terrorism and money-laundering. This was the view taken by the Financial Action Task Force (‘FATF’) – an intergovernmental body that promotes implementation of anti-money laundering and anti-terrorist financing measures through its recommendations and country evaluations – in its initial assessment of the terrorist financing vulnerabilities and threats faced by the non-profit sector (FATF, 2012[49]). Since 2012, FATF’s Recommendation 8 has served as an international policy standard influencing the domestic regulation of cross-border philanthropy. FATF subsequently acknowledged that the non-profit sector’s vulnerability to terrorist abuse in Recommendation 8 may have been overstated given that ‘not all non-profit organisations are inherently high risk (and some may represent little or no risk at all)’ (FATF, 2016[50]).

The initial view about the vulnerability of the non-profit sector has resulted in additional administrative and due diligence requirements being imposed on donors and philanthropic entities seeking to work abroad in many countries. It has also been put forward as one reason for imposing restrictions on tax deductibility of donations as revenue authorities refer to the lack of control over the recipient entity (Charities Aid Foundation, 2016[51]).

Some recipient countries have also viewed foreign funding or foreign activities as a threat to national sovereignty and have imposed restrictions that might include prohibiting foreign funding or requiring all funding to be channelled through the government or approved by government (Indiana University Lilly Family School of Philanthropy, 2018[52]). Concerns about cross-border philanthropy raise some of the same issues as official aid programs, although as noted, not all cross-border aid is concerned with humanitarian or development activities.

There are various ways that cross-border philanthropy can occur where:

  • a donor in one country makes a donation directly to a philanthropic entity in another country (‘direct cross-border philanthropy’); or

  • the donation is made to a domestic philanthropic entity that pursues its programs abroad; or the donation is made to a domestic philanthropic entity that channels the funds to the foreign philanthropic entity (‘indirect cross-border philanthropy’).

Most countries do not permit tax relief for donations to foreign philanthropic entities (direct cross-border philanthropy), subject to some exceptions that allow tax relief for donations within a geographic region.

The position in the European Union (‘EU’) is based on the fundamental freedoms of the Treaty on the Functioning of the EU (‘TFEU’) mandating the non-discrimination of philanthropic entities and their donors. This has been confirmed by the case law of the European Court of Justice (‘ECJ’), which ensures that the tax autonomy of the Member States is exercised in accordance with the fundamental freedoms as enshrined in the TFEU. In the landmark judgement in Stauffer (Case C-386/04), the ECJ held that non-resident philanthropic entities should not be treated differently for tax purposes simply because they are resident in another Member State. Thus, if a Member State grants an income tax exemption to domestic philanthropic entities, it should extend such advantageous tax treatment to entities in other Member States which meet the same conditions as domestic philanthropic entities. In a subsequent judgement in Persche (Case C-318/07), which complements the Stauffer case with regard to the tax treatment of donors in respect of cross-border giving, the ECJ ruled that limiting the preferential tax treatment for donations to domestic philanthropic entities while excluding donations to comparable foreign entities is not compatible with the free movement of capital as guaranteed by the TFEU. This means that tax relief should be provided where the foreign charities in a Member State can be shown to be ‘comparable’ to domestic organisations holding charitable tax status. Practical difficulties remain in demonstrating comparability and the process for seeking tax incentives is complex and burdensome (European Foundation Centre, 2017[47]). The end result is that donors are more likely to use indirect channels (discussed below).

Another regional grouping that permits some cross-border donations to qualify for tax relief concerns the United States. The US permits deductions for philanthropic contributions in its treaties with Canada (USA-Canada DTA, Article XXI), Mexico (USA-Mexico DTA Article 22) and Israel (USA-Israel DTA Article 15A). Like the EU, the basis for the concession is ‘comparability’ with eligible domestic entities. The rationale for the philanthropic contribution provisions in the tax treaties with Canada and Mexico are close geographic and economic ties. In the case of Israel, the reasons for the preferential tax treatment appear to be the close political ties and the extensive funding of philanthropic activities in Israel by US citizens. There is however no equivalent provision in the Canada-Mexico DTA.

Some countries do permit domestic philanthropic entities to transfer funds or to operate overseas. There are various models that are used to facilitate this. Many international non-government organisations (NGOs) such as the Red Cross, Amnesty International, Greenpeace and World Vision, establish domestic entities in a large number of countries but essentially undertake all their activities offshore. There may also be a process for other domestic philanthropic entities to be approved so that they can undertake activities overseas. The approval processes for such entities tend to be quite onerous and may be restricted to entities that undertake humanitarian and/or development type of activities. In addition to an approval process, there may be a code of conduct imposed to ensure that entities operating overseas, that are eligible for tax preferred donations, meet certain standards. Some countries permit a wider group of domestic entities to be approved. One example is the proliferation of entities designated by the prefix ‘friends of’ which allows tax relief for donations that are to be used offshore. The need to go through an onerous approval process tends to penalise smaller philanthropic entities. However, a number of countries permit a ‘work around’ that is, entities can request an approved entity to act as a conduit and pass on donations to intended recipients overseas. The entity acting as a conduit will typically charge a fee of between 5% and 10% and will be responsible for the due diligence associated with the funds being passed on.

An alternative model has developed in Europe where, despite the rulings by the ECJ, cross-border giving is still not easy. A private initiative, the Transnational Giving Europe (‘TGE’) network, is a partnership of leading European foundations and associations that facilitates tax-efficient cross-border giving within Europe (Transnational Giving Europe[53]).The TGE network covers 19 countries and enables donors, both corporations and individuals, resident in one of the participating countries, to financially support non-profit organisations in other Member States, while benefiting directly from the tax advantages provided for in the legislation of their country of residence. (Transnational Giving Europe[53])

As the world becomes more interconnected, the argument that countries should treat cross-border philanthropy in much the same way as domestic philanthropy becomes harder to ignore. If governments accept that they should subsidise domestic philanthropy, arguments can also be made as to why they should subsidise cross-border philanthropy. The arguments in favour of tax incentives fall into two categories: arguments that rely on what might be described as the ‘moral imperative’ to assist others, especially those less well off, and arguments that are based on the ‘self-interest’ of the country providing the tax relief.

The global nature of many of the challenges facing the world require global responses. If the relief of poverty, advancement of health and education and preservation of heritage are worthy purposes domestically, they should be seen as ‘deserving’ globally and so private contributions should be encouraged (Buijze, 2016[54]). Many issues also transcend borders e.g., environmental concerns, or medical research or public health issues such as fighting pandemics, can only be dealt with by countries cooperating. There is already considerable cooperation in the area of development and humanitarian aid, as well as in responding to international disasters. The global public benefit these causes relate to could be a possible reason for some governments to stimulate private contributions to these causes through tax incentives.

The provision of tax incentives for philanthropy directed abroad can also be justified using the subsidy rationale by refocussing on the notion of what constitutes the public benefit. There may be a number of benefits to a country from engaging with global causes by supporting cross-border giving. Just as aid programs enable a country to develop ‘soft power’ through cultural and economic influence, so does the provision of support for philanthropy (Jenkins, 2007[55]). Certainly, there appears to have been a level of acceptance of this argument by many countries, given the number of domestic philanthropic entities that already have this support. A more focussed notion of self-interest might be present in allowing tax relief in a regional context, where wellbeing of a region, increased sense of solidarity and strengthening of community ties, may generate benefits for the country in which the donors are present.

There are, of course, a number of arguments that can be raised against providing support for cross-border giving. Those arguments tend to focus on the lack of benefit for the ‘donor’ country and the lack of oversight of the funds once they leave the country. There may also be concerns for the donors or intermediaries about navigating the legal and cultural terrain of the recipient country.

One possible argument is that the granting of tax relief is a cost to the donor country, and there is no matching benefit in terms of spending within the country (Buijze, 2016[54]). This is in effect confining the public in the public benefit requirement to the domestic sphere, although it could be argued that a country derives a benefit from the provision of assistance to less fortunate countries. In some cases, this is recognised as where there is already tax relief for donations to recognised disasters.

Lack of oversight of the actual spending of the private contributions is often raised as a concern (Buijze, 2016[54]). This has been exacerbated by the FATF Report in 2012 describing the vulnerability of the non-profit sector. Although the FATF has since moderated its view, the uneasiness remains. There are also issues concerning compliance of domestic entities operating abroad, often in partnership with local entities, to ensure they comply with any codes of conduct as well as the laws of the country in which they are operating e.g., Oxfam in Haiti. Philanthropic activities abroad are also more difficult to administer, which in some cases may raise accountability and transparency issues.

Entities that operate across borders may also encounter additional costs of complying with a different legal regime, having to navigate supply arrangements in countries that may have an element of endemic corruption as well as translation and other costs associated with engaging with local populations (Charities Aid Foundation, 2016[51]). All of these costs may mean that less money is actually being spent on the worthy purpose and may discourage countries from supporting philanthropy across borders.


[7] Andreoni, J. (1990), “Impure Altruism and Donations to Public Goods: A Theory of Warm-Glow Giving”, The Economic Journal, Vol. 100/401, pp. 464-477.

[8] Andreoni, J. and A. Payne (2003), “Do Government Grants to Private Charities Crowd Out Giving or Fund-raising?”, American Economic Review, Vol. 93/3, pp. 792-812.

[33] Andrews, W. (1972), “Personal Deductions in an Ideal Income Tax”, Harvard Law Review, Vol. 89, pp. 309-385.

[21] Atkinson, A. and N. Stern (1974), “Pigou, Taxation and Public Goods”, The Review of Economic Studies, Vol. 41/1, pp. 119-128.

[23] Atkinson, R. (1997), “Theories of the Federal Income Tax Exemption for Charities: Thesis, Antithesis and Synthesis”, Stetson Law Review, Vol. 27, pp. 395-431.

[56] Auten, G., C. Clotfelter and R. Schmalbeck (2000), "Taxes and philanthropy among the wealthy", in J. Slemrod (Ed.), Does atlas shrug: The economics consequences of taxing the rich, New York: Russell Sage.

[14] Backus, P. and N. Grant (2019), “How sensitive is the average taxpayer to changes in the tax-price of giving?”, International Tax and Public Finance, Vol. 26/2, pp. 317-356.

[5] Bekkers, R. and W. Pamala (2011), “A Literature Review of Empirical Studies of Philanthropy: Eight Mechanisms That Drive Charitable Giving”, Nonprofit and Voluntary Sector Quarterly, Vol. 40/5, pp. 924–973.

[36] Bittker, B. (1972), “Charitable Contributions: Tax Deductions or Matching Grants?”, Tax Law Review, Vol. 28, pp. 37-63.

[24] Bittker, B. and G. Rahdert (1976), “The Exemption of Non-profit Organisations from Federal Income Tax”, Yale Law Journal, Vol. 85/3, pp. 299-358.

[22] Boadway, R. and K. Michael (1993), “Public Goods, Self-Selection and Optimal Income Taxation”, International Economic Review, Vol. 34/3, pp. 463-478.

[16] Bönke et. al. (2013), “Charitable giving in the German welfare state: fiscal incentives and crowding out”, Public Choice, Vol. 154, pp. 39-58.

[17] Bönke, T. and C. Werdt (2015), “Charitable giving and its persistent and transitory reactions to changes in tax incentives: Evidence from the German taxpayer panel”, School of Business and Economics Discussion Papers, No. 2015/2, Free University Berlin.

[25] Brody, E. (1999), “Charities in Tax Reform: Threats to Subsidies Overt and Covert”, Tennessee Law Review, Vol. 66/3, pp. 687-763.

[38] Brody, E. and J. Cordes (2001), The Unrelated Business Income Tax: All Bark and No Bite?, Urban Institue, Washington, D.C.

[27] Buckles, J. (2005), “Community Income Theory of the Charitable Contributions Deduction”, Indiana Law Journal, Vol. 80, pp. 947-988.

[54] Buijze, R. (2016), “Approaches towards the Application of Tax Incentives for Cross-Border Philanthropy”, Intertax, Vol. 44/1, pp. 14-28.

[35] Burton, M. and K. Sadiq (2013), The Tax Expenditure Concept in Tax Expenditure Management: a Critical Assessment, Cambridge University Press, Cambridge.

[51] Charities Aid Foundation (2016), Donation States: An international comparison of the tax treatment of donations, https://www.cafonline.org/docs/default-source/about-us-publications/fwg4-donation-states.

[6] Charities Aid Foundation (2014), Why we give, https://www.cafonline.org/docs/default-source/about-us-publications/caf-why-we-give-jan14.pdf.

[9] Clotfelter, C. (1985), Federal Tax Policy and Charitable Giving, University of Chicago Press, Chicago.

[10] Clotfelter, C. and L. Salamon (1982), “The Impact of the 1981 Tax Act”, National Tax Journal, Vol. 35/2, pp. 171-187.

[19] Diamond, P. and J. Mirrlees (1971), “Optimal Taxation and Public Production I: Production Efficiency, II Tax Rules”, American Economic Review, Vol. 61/1, pp. 8-27 and 261-78.

[37] Duff, D. (2014), “The Tax Treatment of Charitable Contributions in a Personal Income Tax: Lessons from Theory and the Canadian Experience” in M. Harding, A. O’Connell and M. Stewart (eds.) Not-for-Profit Law: Theoretical and Comparative Perspectives, Cambridge University Press, Cambridge.

[47] European Foundation Centre (2017), Boosting Cross-Border Philanthropy in Europe: Towards a Tax-Effective Environment, https://www.efc.be/uploads/2019/03/Boosting-Cross-Border-Philanthropy-in-Europe-Towards-a-Tax-Effective-Environment.pdf

[15] Fack, G. and C. Landais (2010), “Are Tax Incentives for Charitable Giving Efficient? Evidence from France”, American Economic Journal: Economic Policy, Vol. 2/2, pp. 117-141.

[50] FATF (2016), Outcomes of the Plenary meeting of the FATF, Paris.

[49] FATF (2012), International Standards on Combating Money Laundering and the Financing of Terrorism & Proliferation - the FATF Recommendations, Paris.

[28] Fleischer, M. (2018), Research Handbook on Not-For-Profit Law: Subsidizing Charity Liberally, Edward Elgar, Cheltenham.

[1] Hansmann, H. (1987), Economic theories of non-profit organizations, Yale University Press, London.

[40] Henry, K. et al. (2009), Australia’s Future Tax System, https://treasury.gov.au/review/the-australias-future-tax-system-review.

[52] Indiana University Lilly Family School of Philanthropy (2018), The Global Philanthropy Environment Index, https://scholarworks.iupui.edu/handle/1805/15958.

[55] Jenkins, G. (2007), “Soft Power, Strategic Security and International Philanthropy”, North Carolina Law Review, Vol. 85/3, pp. 773-846.

[29] Kahn, H. (1960), Personal Deductions and the Federal Income Tax, Princeton University Press, Princeton.

[34] Krever, R. (1991), Tax Deductions for Charitable Donations: A Tax Expenditure Analysis, Australian Tax Research Foundation, Sydney.

[43] Lideikyte-Huber, G. (2020), “Tax Incentives for Charitable Giving as a Policy Instrument: Theoretical Discussion and Latest Economic Research”, World Tax Journal, Vol. 12/3.

[46] Moore, D. and D. Rutzen (2011), “Legal framework for global philanthropy: Barriers and opportunities”, International Journal of Not-for-Profit Law, Vol. 13/1-2, pp. 5-42.

[48] OECD (2018), Private Philanthropy for Development, OECD Publishing, Paris.

[4] Ott, J. and L. Dicke (2016), The nature of the non-profit sector, Westview Press, Boulder.

[13] Peloza, J. and P. Steel (2005), “The Price Elasticities of Charitable Contributions: A Meta-Analysis”, Journal of Public Policy and Marketing, Vol. 24/2, pp. 260-272.

[30] Rabin, E. (1966), “Charitable Trusts and Charitable Deductions”, New York University Law Review, Vol. 41, pp. 912-925.

[12] Randolph, W. (1995), “Dynamic Income, Progressive Taxes, and the Timing of Charitable Contributions”, Journal of Political Economy, Vol. 91/1, pp. 371-382.

[45] Reich, R. (2019), Just Giving: Why Philanthropy is Failing Democracy and How it Can do Better, Princeton University Press, Princeton.

[26] Rushton, M. (2007), “Why are Non-profits Exempt from the corporate Income tax?”, Non-profit and Voluntary Sector Quarterly, Vol. 36/4, pp. 662-675.

[18] Saez, E. (2004), “The optimal treatment of tax expenditures”, Journal of Public Economics, Vol. 88/12, pp. 2657–2684.

[3] Salamon, L. (1987), “Of Market Failure, Voluntary Failure and Third Party Government: Towards the a Theory of Government-Non-profit Relations in the Modern Welfare State”, Non-profit and Voluntary Sector Quarterly, Vol. 16, pp. 29-49.

[20] Sandmo, A. (1975), “Optimal Taxation in the Presence of Externalities”, The Swedish Journal of Economics, Vol. 77/1, pp. 86-98.

[39] Sharpe, D. (1996), “Unfair Business Competition and the Tax on Income Destined for Charity: Forty-Six Years Later”, Forida Tax Review, Vol. 3, p. 367.

[11] Steinberg, R. (1990), “Taxes and Giving: New Findings”, International Journal of Voluntary and Non-profit Organizations, Vol. 1/2, pp. 61-79.

[41] Steuerle, E. (1998), The Issue of Unfair Competition, Center for the Study of Philanthopy and Voluntarism, Institute of Policy Sciences and Public Affairs, Duje University, Durham NC.

[32] Surrey, S. (1970), “Tax Incentives as a Device for implementing Government Policy: A Comparison of Direct Government Expenditures”, Harvard Law Review, Vol. 83/4, pp. 705-738.

[53] Transnational Giving Europe (n.d.), How Does it Work? https://www.transnationalgiving.eu/how-does-it-work.

[31] Treasury Department United States (1968), Annual Report of the Secretary of the Treasury on the State of Finances for the Fiscal Year Ended June 30, Washington, D.C.

[42] Weisbrod, B. (1988), To Profit or Not to Profit: The Commercial Transformation of the Non-Profit Sector, Oxford University Press.

[2] Weisbrod, B. (1975), Toward a Theory of the Voluntary Non-Profit Sector in a Three-Sector Economy, Oxford University Press, Oxford.

[44] Zelinsky, E. (2012), “The Counterproductive Nature of Tax Expenditure Budgets”, Tax Notes, Vol. 137/12, pp. 1-9.


← 1. The focus of this chapter is on the case for (and against) providing tax concessions for philanthropy. A broader discussion of the benefits to society of philanthropy and the philanthropic sector is beyond the scope of this chapter.

← 2. The meaning of the term ‘philanthropy’ has been discussed in Chapter 1. The term ‘philanthropic giving’ is used in this report to refer to the act of giving to entities, rather than to individuals, as this is the type of giving that may qualify for tax relief. This report uses the term philanthropic entities to refer to not-for-profit entities with a ‘worthy’ purpose that typically provide public benefits, and the philanthropic sector to refer to the sector covering such entities.

← 3. The price elasticity of philanthropic giving is generally estimated by analysing the effect a change in the price of giving (which, in a country where private contributions are tax deductible, is equal to (1-t), where t is an individual’s marginal tax rate) has on the level of philanthropic giving.

← 4. There is also evidence that high income taxpayers are more likely to donate large amounts periodically rather than regular amounts every year (Auten, Clotfelter and Schmalbeck, 2000[56]).

← 5. See for example, Chancellor of the Exchequer in UK, William Gladstone put forward a bill in 1863 to abolish the exempt status of charities on the grounds that it was an undiscriminating public subsidy for a large group of organisations not subject to adequate public scrutiny, including elite schools, but was defeated.

← 6. See the discussion relating to ‘efficiency’.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2020

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at http://www.oecd.org/termsandconditions.