This chapter focuses on the drivers of tax morale of individuals. Using micro-econometric analysis of public opinion surveys at the global and regional levels it identifies a number of socio-economic and institutional factors that influence tax morale. The chapter also discusses a number of policy responses to help build tax morale, including taxpayer education programmes, behavioural economics approaches, and hypothecated taxes. While more detailed country level analysis is required before embarking on concrete policy changes, the overview provided in this chapter provides a starting point for discussion and further research on the role of tax morale in developing countries.
Tax Morale
What Drives People and Businesses to Pay Tax?
1. Tax morale in individuals
Abstract
There is growing research that demonstrates a significant and positive relationship between tax morale and tax compliance in developed and developing economies (Ali, Fjeldstad and Sjursen, 2014[1]; Cummings et al., 2009[2]). Recent growth in the use of behavioural economics in tax compliance illustrates how tax administrations are seeking to use an understanding of tax morale to increase compliance (OECD, 2017a[3]).
This chapter uses recent public opinion surveys to revisit previous findings (OECD, 2013[4]) and deepen understanding of the determinants of tax morale, especially in Africa, Asia and Latin America. At the global level, the survey questions focussed on the justification for tax avoidance or evasion, which necessarily limits the definition of tax morale, but at the regional level more proxies are used. The research looks at socio-economic and institutional factors, both at the global and regional levels. Using these elements, it also seeks to test for the robustness of the fiscal contract as being a key aspect of tax morale (OECD/CAF/UN ECLAC, 2018[5]).
The findings, combined with lessons from other research, highlight the potential for greater consideration of and research on tax morale in the design and administration of tax policy. For instance, to improve taxpayer profiles and taxpayer education programmes, as well as indicating the need for further research, especially on gender differences in tax morale.
A focus on tax morale should complement, not replace, other efforts to improve compliance. Tax morale forms a part of an individual’s compliance decisions, and so needs to be considered alongside other factors (e.g. enforcement). While this work provides a useful starting point for discussion and consideration of tax morale, more detailed country-level research is needed to identify how best to practically incorporate tax morale considerations into broader tax policy and administration in countries.
1.1. Tax morale in the world today: global analysis
This report confirms that previous findings on institutional and socio-economic factors associated with tax morale have held over the past decade. Previous research (OECD, 2013[4]) used the World Values Survey to identify institutional and socio-economic factors associated with tax morale. This exercise has been repeated with new data from a more recent round of the survey. Repeating this exercise (see Annex A and Table A.9 for methodology) confirms previous findings, providing confidence in the identification of long-term trends and possible policy recommendations.
OECD member and Latin American economies have the highest levels of tax morale, with Africa and Eastern Europe reporting lower levels. Comparing individual attitudes towards tax avoidance and evasion1, it is possible to measure and compare tax morale in different regions. The results are shown in Figure 1.1. By this measure, OECD2 and Latin American economies show the highest levels of tax morale (see Section 1.3 on recent changes in Latin America), with more than 70% of the population reporting that they would never justify cheating on taxes. This drops to just over 50% in Africa and Eastern Europe. There are, however, variations within each region, most dramatically in Africa, where the range for respondents never justifying cheating stretches from 82% in the highest-ranked country to 32% in the lowest-ranked country.
Figure 1.1. Tax morale in the world by regions, 2010-2014
Note: Tax morale across regions is measured as the percentage of respondents that report that they would never justify cheating on taxes. See Annex A for detailed country classification, Table A.1.
Source: OECD Development Centre / Centre for Tax Policy and Administration calculations based on World Values Survey (2010-2014).
Countries with higher levels of taxation as a percentage of GDP appear to have higher levels of tax morale. Figure 1.2 shows a positive correlation between the level of tax morale and the tax-to-GDP ratio in a country, where each dot represents a country in the World Values Survey. This may help explain the variation between the regions as the OECD member countries and Latin America tend to have higher tax-to-GDP ratios than other regions. The causality of this correlation is not clear. One possibility is that the correlation is indicative of a virtuous circle of public service provision, voluntary tax compliance, and fiscal stability issues that many developing countries face challenges in addressing. Other relationships could be of interest, such as the impact of informality on tax morale for instance, but data limitations impede this type of exercise (at least in a multi-country analysis) for the moment. The rest of this chapter, including the regional analysis, seeks to explore some of these dynamics further.
Figure 1.2. Tax morale and tax-to-GDP ratios
Source: For tax morale, the data is from the World Values Surveys (2010-14). For tax-to-GDP ratios, the data is from the OECD Global Revenue Statistics Database and the World Development Indicators.
Institutional and socio-economic factors are strongly linked with tax morale, a result that confirms previous findings. Figure 1.3 shows the marginal effects (see Table A.2 for full results) of each variable on the probability of reporting high tax morale. In other words, what effect an additional unit of – say, increased trust in government, being a year older or having an extra level of education (college degree vs. high school) has on tax morale. Figure 1.3 includes the results from previous OECD research (based on World Values Survey 2005 data), where available, for comparison (OECD, 2013[4]). The analysis did test for the role of different perceived levels of income and (un)employment status, but no significant association was identified (see Annex A, Table A.3). The consultation responses highlighted that further research, using alternative data sources, would be useful to understand the tax morale of HNWIs. Such research could look both at how HNWIs respond to policies such as tax amnesties, where consultation responses identified both potential benefits and risks, as well as how the tax behaviour of HNWIs affects the tax morale of other taxpayers (OECD, 2019[6]).
Figure 1.3. Institutional and socio-economic factors associated with tax morale
Note: * no data from previous analysis. Marginal effects on the probability of reporting high tax morale. Results based on regression coefficients of significant variables in column 7, Table A.3 in Annex A.
Source: OECD Development Centre / Centre for Tax Policy and Administration estimations based on World Values Survey (2010-2014), and OECD, 2013.
The analysis of global results emphasises that socio-economic factors help explain individuals’ tax morale:
More educated individuals have more positive attitudes towards paying taxes.
Women have higher tax morale than men.
Older people are less likely to justify cheating on taxes than younger people.
Those who are citizens of the country they live in have higher tax morale than non-citizens.
Individuals who claim a faith or religious identity have more positive attitudes towards paying taxes.
Some of these results are perhaps expected – those with higher education, for example, are more likely to have a deeper understanding of the role of taxation in the economy. Nevertheless, confirming these links does have practical value, giving confidence to use such factors in constructing taxpayer profiles and enabling better targeting of efforts to increase compliance. It is necessary, however, to undertake such analysis at the national (or even sub-national) level to ensure that such analysis is valid for the local context. Regional analysis shows (see Sections 1.2 - 1.4 below) that the significance of these factors can vary.
Institutional factors show that government policies and performance are also likely to have a significant impact on tax morale:
Individuals who believe they are living in a meritocratic society have significantly higher tax morale.
Those who trust their national government display higher tax morale than those who do not.
People who identify fiscal redistribution to be essential (i.e. governments should tax the rich to enable support for the poor) show higher tax morale.
Those who perceive democracy to be the best system of government for their country tend to think that cheating on taxes is unjustifiable.
A common thread of all these factors is that government performance and its responsiveness to the needs of citizens drive tax morale. Trust in government is affected by how effective the government is in delivering on its promises, while perceived meritocracy relies upon governments helping create a society where hard work results in success and where opportunities are available to all. As highlighted in responses to the consultation on this report, it would also be interesting to fill a current research gap by testing how taxpayers’ perception of the fairness of the tax system affects tax morale (OECD, 2019[7]).
While this global analysis is useful up to a point, it is only regional or country-level data that can help explore specific aspects of government performance which drive tax morale. The following sections investigate this issue in Africa, Asia and Latin America.
1.2. Tax morale by region - Africa
Tax morale appears to be improving in Africa. Figure 1.4 below shows that between 2005 and 2015 the proportion of Africans who strongly believed that the tax authorities had the right to make people pay taxes increased from 22% to 30%. In addition, more Africans expressed the view that not paying taxes was wrong and punishable in 2015, than in the previous surveys (Afrobarometer, 2011-2013). There is, however, significant variation between countries and although most countries have seen increased tax morale, some have seen it decline (Afrobarometer, 2011-2015).
Figure 1.4. Tax Morale in Africa
Note: Panel A shows responses to the question “For each of the following statements, please tell me whether you disagree or agree: The tax authorities always have the right to make people pay taxes.” In Panel B, tax morale is measured as the average of all available observations for the question “How wrong is not paying the taxes” on a scale from 1 (not wrong at all) to 3 (wrong and punishable). The black bar represents a simple average of the countries included in the Afrobarometer.
Source: OECD Development Centre / Centre for Tax Policy and Administration calculations based on Afrobarometer (2015).
There are several possible explanations for this increase, including improvements in the tax-to-GDP ratio and in the tax administration. Figure 1.2 highlighted the correlation between tax morale and tax-to-GDP ratios and this has been gradually increasing in Africa for some time. Notably, the average of the 21 countries included in Revenue Statistics in Africa 2018, passed the 15% mark in 2004 (OECD/ATAF/AUC, 2018[8]), the threshold generally accepted as necessary for minimum effective state functioning. The achievement of this minimum threshold of state capacity may therefore reflect the development of a virtuous circle between tax and service delivery described earlier. In addition, Moore et al. emphasize the steady improvements made by revenue authorities in Africa (Moore, Prichard and Fjeldstad, 2018[9]). The more detailed survey data in Africa enables factors on tax morale, such as service delivery and tax administration, to be tested.
The perceived improvement in the quality of tax administration impacts significantly tax morale. The Afrobarometer survey is the most detailed of the regional surveys and includes a focus on tax administration. Figure 1.5 below shows how important the perceptions and quality of the tax authority are to taxpayers. Gains in the perceived legitimacy of the tax authority have the largest impact in boosting tax morale, and linking taxes to public spending, particularly health, could improve the willingness to increase tax payments. The recent improvements in tax administrations may therefore have been a key driver of increasing tax morale in Africa.
Making taxes simpler to pay appears to generate more willingness to pay. Another constant observed is that a higher perceived difficulty in finding out how to pay taxes has a negative impact on tax morale. There are several potential policy responses to this challenge. Firstly, taxpayer education programmes, which engage with taxpayers on why and how to pay tax, can be used to make it easier for taxpayers to navigate the tax system. In addition, increased use of technology offers potential for improvements in tax collection. Used effectively, technology can both simplify the process of paying taxes and help increase the integrity of the tax system by reducing opportunities for corruption (Bird and Zolt, 2008[10]). Many developing countries have been making significant progress when it comes to corruption (one of the factors likely to be driving the improved perceived legitimacy of tax authorities in Africa), yet there remains significant further potential. The OECD’s Forum on Tax Administration (FTA) is responding to this challenge by identifying best practices in the use of technology in tax administration. Lastly, addressing the high prevalence of ‘small taxes’ in Africa, which while insignificant in statistical terms, are the dominant taxpaying experience for many individuals (especially women) in Africa, may offer a route to improve tax morale in Africa (Moore, Prichard and Fjeldstad, 2018[9])3.
Figure 1.5. Factors associated with tax morale in Africa, 2015
Note: Marginal effects on the probability of reporting high tax morale and a high willingness to increase taxes. Panel A, B and C results are based on regression coefficients of significant variables in the last column of Table A.6, Table A.7 and Table A.8, respectively (Annex A). In Panel A, tax morale is measured through the question “Do you justify cheating on taxes if you have the chance?”
Source: OECD Development Centre / Centre for Tax Policy and Administration estimations based on Afrobarometer (2015).
Improved public service delivery appears to be a driver of tax morale in Africa. Effective public services are a means to demonstrate how well governments turn tax revenues into beneficial expenditures, so these can produce a double dividend comprising both the intrinsic benefit of the service provided and the spillover benefits from public satisfaction generated by its provision. The results from Africa (Figure 1.5) suggest that this relationship exists with respect to tax morale, with the satisfaction with roads, education and health all showing a positive impact on tax morale.
Hypothecated or earmarked taxes seek to use the link between service delivery and tax morale to create the political space for tax reforms, but can be problematic. By linking certain taxes to spending on specific services, hypothecated taxes have been suggested as a tool to build the fiscal contract, especially in cases where trust in government is especially low (Prichard, 2015[11]). While earmarked taxes may offer an attractive option in the short-term, there are a number of challenges, not least in how they are designed (McCleary, McCleary and William, 1991[12]) 4 and the impact that they can have over the longer term – see Box 1.1.
Box 1.1. Hypothecated taxes in Ghana
Ghana has made repeated use of earmarked taxes to provide funding in several key areas. Three of the 14 earmarked funds are:
The Ghana Education Trust Fund (GETFund) was introduced in 2000 to fund education and is now funded by an additional flat 2.5% VAT.
The National Health Insurance Scheme was established in 2005 to support healthcare and is now funded by both an additional 2.5% VAT flat rate and a 2.5% Social Security and National Insurance Trust contributions.
The Ghana Infrastructure Investment Fund was established in 2014 and is funded by a 2.5% VAT rate.
The growth of earmarked taxes in Ghana, and the funds generated by them, resulted in nearly 40% of tax revenue becoming allocated to earmarked funds. This reduced the fiscal space for the government and in response an Earmarked Funds Capping and Realignment Act was introduced in 2017 to cap earmarked funds tax revenues to 25% of tax revenue, with the additional revenues being made available to the general budget. While this has increased the fiscal space for the government, it has significantly diminished funding for the earmarked funds, reducing its ability to deliver services. There are concerns that this may result in decreased tax morale – if service delivery is scaled down while the earmarked tax rates remain.
Source: Cassiel Forson MP (Ghana).
Behavioural economics approaches provide another, less rigid, way to use the link between service delivery and tax morale. A number of countries have used ‘nudge’ approaches in enforcement (see Box 1.2). Based on behavioural economics principles, these methods generally use the introduction of new messages in communications from tax authority to taxpayers. In Rwanda, this approach included a focus on service delivery. In a taxpayer letter experiment conducted by the Rwanda Revenue Authority in 2016, letters to taxpayers highlighting the links between taxes and public services were more successful than deterrence letters (Mascagni, Nell and Monkam, 2017[13]). The literature on these exercises highlights their benefits, but also their limits; for example, if behavioural economics approaches are not accompanied by substantial improvements in governance and public services, then the positive results can fade over time (Castro and Scartascini, 2015[14]; Coleman, 2007[15]; Cummings et al., 2009[2])). Some of the consultation responses also highlighted the role for other academic disciplines beyond economics and psychology in tax compliance, including political science, anthropology and sociology (OECD, 2019[7]).
Box 1.2. Behavioural economics approaches to tax compliance
The Behavioural Insights Team (BIT)5 have been supporting the use of behavioural economics by a number of countries, examples include:
The United Kingdom: With the help of the BIT, the UK tax department sent letters to taxpayers who had not paid their taxes on time highlighting social norms to improve tax payments. Examples included: “Nine out of ten people with a debt like yours, in your area, pay their tax on time”, “The great majority of people in your local area pay their tax on time” and “Most people with a debt like yours have paid it by now”. The latter example increased tax payments from the people receiving the letters from 34% to 39%.
Guatemala: In collaboration with the Guatemalan tax authority, the BIT and the World Bank carried out a randomised controlled trial in Guatemala. The trial consisted of modifying reminder letters sent to taxpayers who had failed to declare their income tax on time. The letter included the following statement: “According to our records, 64.5% of Guatemalans declared their income tax for the year 2013 on time. You are part of the minority of Guatemalans who are yet to declare for this tax.” Tax payment among letter recipients increased by 5.5 percentage points. (The Behavioural Insights Team, 2014[16]). It is estimated that if the letter was sent to all taxpayers it would generate USD 760 000 additional revenue, which is 36 times the cost of sending the letters (Kettle et al., 2016[17]).
Mexico: The Mexican authorities along with the Inter-American Development Bank and the BIT team sought to encourage businesses to declare revenues via SMS reminders. The messages were sent to 748 499 taxpaying businesses that had not submitted their declaration on time. The most effective reminder, which highlighted the potential fines for non-compliance, increased declaration rates from 24% to 33% (Kettle et al. forthcoming).
Costa Rica: The BIT, in conjunction with the Costa Rican authorities, sent emails to 12 515 firms to encourage them to submit their 2014 income tax declarations. The firms were randomly selected to receive either no email, a behaviourally informed email (including a deterrence message), or the same email but including additional examples of transactions made by the recipient firm, highlighting that these were known to the tax authority. Results show that the firms that received the behavioural email increased tax declaration from 11.5% up to 32.5% (simple email) and up to 34.2% (email with transactions).
There are several lessons to bear in mind with such behavioural approaches however.
Communication interventions are most effective when taxpayers have few other sources of information, hence the communication is more likely to update the taxpayers’ prior beliefs.
It is important to target interventions at sub-groups in the population, e.g. those with specific past behaviour, the wealthiest taxpayers etc.
Behavioural approaches do not exist in isolation, they need to work together with other approaches, especially enforcement and policies that aim to rebuild trust and legitimacy and thus increase tax morale.
Although behavioural approaches are relatively cost effective in the short term, they are not a replacement for a well-designed tax system, but can help build morale effectively when the fundamentals of the system are in place.
Source: OECD and www.bi.team/blogs.
In general, the factors influencing tax morale in Africa match those at the global level. Support for democracy, faith, educational attainment and trust in government, which were all identified as factors influencing tax morale globally, also appear as factors in Africa, though with some variations. Of these, trust in government was the most consistent factor, further reinforcing the relationship between tax morale and government performance.
An important divergence from the global analysis is that women appear to have lower tax morale than men in Africa. Tax systems can affect men and women in different ways. For instance, recent research in Nigeria has identified implicit tax biases that affect women negatively (Akpan and Sempere, 2019[18]). This sometimes takes the form of explicit references to “women” in the tax code, but mainly arises from the interaction of the tax system with other patterns of economic behaviour, including differences in labour participation, entrepreneurship, wage rates, education decisions, savings, property ownership, and consumption. When tax policies fail to consider how gender roles and inequalities affect how women and men earn, control, save and spend their income and wealth, they can entrench gender inequalities and deepen women’s poverty and/or decrease their willingness to participate in the fiscal contract.
There is a need to increase both the research and availability of data on gender and taxation, especially in Africa. The findings in this report on the tax morale of women in Africa emphasise the need to research gender and taxation. With currently available research, it is not possible to identify specific reasons for the gender variation in tax morale and why women have lower tax morale in Africa. Moore, Prichard and Fjeldstad (2018[9]) highlight some possible reasons, notably market taxes for small traders, as well as biases in administration, but reach no clear conclusions. In Sierra Leone, Van Den Boogaard finds that female-headed households pay fewer formal taxes than men, but more informal taxes; however, the total tax burden is higher for women, both in terms of formal and informal taxation (Van Den Boogaard, 2018[19]). The study of the gender dimensions of tax (and wider fiscal policy) is further complicated by a lack of sex-disaggregated data. Additional research should seek to explore gender dimensions, especially in areas such as: differences in tax regimes for predominately female-/male-headed sectors, how tax systems recognise unpaid care, how tax systems deal with different family structures, differences in taxation between rural/urban and informal/formal sectors. Tax morale surveys themselves could provide useful input into this work, as significant gender discrepancies in tax morale may be indicative of a gender bias in the fiscal system.
1.3. Tax morale by region - Latin America
Tax morale appears to be falling across Latin America, from a relatively high base. Figure 1.6 below shows the evolution of tax morale in Latin America. Since 2011, tax morale has weakened: the percentage of Latin Americans who would never justify evading paying taxes decreased from 54% to 48% in 20156. In contrast, the proportion of those who find evading taxes ‘justifiable’ increased to 27% in 2015 from 17% in 2011 (Panel A, Figure 1.6). A reduction in tax morale has been universal across all economies in the region covered by the Latinobarómetro, although to different degrees (Panel B, Figure 1.6).
Figure 1.6. Tax Morale in Latin America
Note: In panel A, tax morale is measured through the question “Do you justify cheating on taxes if you have the chance?” on a scale of 1 (not at all justifiable) to 10 (totally justifiable). “Never justifiable” is the share of respondents that answered 10; “slightly justifiable’’ corresponds to the share of respondents who answered between 9 and 6; and “justifiable’’ is the share of respondents that answered between 1 and 5.
In panel B, tax morale is represented as the average of all available observations for the question “How justifiable do you believe it is to evade paying taxes?’’ on a scale of 1 (totally justifiable) to 10 (not at all justifiable) and normalised between 1 and 3 for illustration purposes. The black bar is a simple regional average that includes all economies in the Latinobarómetro.
Source: OECD Development Centre / Centre for Tax Policy and Administration calculations based on Latinobarómetro (2016).
Decreasing tax morale appears linked to recent economic and social changes in Latin America. The fall in tax morale in Latin America coincides with an economic slowdown that also began in 2011 and may thus explain it. The Latin American Economic Outlook 2018 notes the economic slowdown brought an end to a period of economic and social progress in Latin America, where poverty had declined sharply and created an increasingly demanding middle class. Recent reversals in poverty reduction and inequality have combined with global trends such as technological change, aging populations and a number of corruption scandals across Latin America. These factors have created a climate of uncertainty regarding the ability of public institutions to respond effectively to the needs of citizens in Latin America (OECD/CAF/UN ECLAC, 2018[5]).
While tax morale overall is declining, it is possible to identify a number of factors that positively affect tax morale. Figure 1.7 shows factors that were associated with individual’s tax morale in Latin America. There were several similarities with the global-level analysis, as higher age, level of education, and faith are all associated with higher tax morale. Lower tax morale among younger individuals highlights the need for differentiated strategies. For instance, potential taxpayer education programmes targeted at Latin America’s large cohort of younger taxpayers (or future taxpayers) including integration with the education system (see Box 1.3 for an example); in 2017, 26% of the Latin American population was between 15 and 29 years old (UN, 2017[20]). As an aside, it is interesting to note that those reporting democracy as their preferred system of government are significantly more likely to never justify cheating on taxes (Figure 1.7, Panel A), but are also less likely to argue for increased taxes to pay for development (Figure 1.7, Panel B). This calls attention to the fact that high tax morale in one area (never justifying cheating on taxes) does not automatically correspond to higher taxes in another area (willingness to pay more taxes).
Figure 1.7. Factors associated with tax morale and taxation in Latin America, 2016
Note: Marginal effects on the probability of reporting high tax morale and high willingness to pay more taxes. Panel A and B results are based on regression coefficients of significant variables in the last column in Table A.4 and Table A.5, respectively (Annex A). In Panel A, tax morale is measured through the question “Do you justify cheating on taxes if you have the chance?”. In Panel B, willingness to pay more taxes is measured through the question “How much are you willing to increase taxes to finance infrastructure works that promote the integration of your country with the world (bridges, highways, airports, ports)?’’.
Source: OECD Development Centre / Centre for Tax Policy and Administration estimations based on Latinobarómetro (2016).
Box 1.3. Educating the next generation – the EuroSocial experience
EuroSocial is a co-operation programme between Latin America and the European Union that has been supporting taxpayer education to increase social cohesion through its Tax and Accounting Assistance Hub (NAF, Spanish initials).
The hub trains university students in the social benefits of taxes and basic tax- and customs-related topics, helping them to co-operate in providing advice on taxes to entrepreneurs and owners of micro and small companies. In 2017, NAF included 571 universities in 10 Latin American countries, resulting in 6,000 students supporting more than 80,000 taxpayers.
The programme provides benefits for all stakeholders. It enables students to strengthen their sense of social responsibility, deepen their general tax knowledge and apply it practically. For the tax administrations, they benefit from the programme as it encourages responsible behaviour by accountants, improves tax morale (including supporting voluntary compliance), and it facilitates a better relationship between the administration and accountants and citizens. The support that students contribute to local communities helps low-income taxpayers on basic issues at no cost to them, as well as supporting small businesses to be formalised and compliant.
The success of NAF programmes requires a joint commitment from both the tax administration and education system. This ensures that programmes have a clear pedagogical approach and allows the tax content to be integrated effectively into the lessons. A successful programme incorporates both an understanding of the role and function of taxes and the importance of the fiscal contract between state and citizens.
Source: EuroSocial.
The relationship between tax morale and public services is less clear for Latin America, with mixed results between previous studies and recent data. Previous research using 2008-2009 Latinobarómetro data finds a positive relationship between tax morale and satisfaction with education and health services (Daude and Melguizo, 2010[21]). Other research has also found a positive relationship with public services (Carrillo, Castro and Scartascini, 2018[22]). The current analysis, however, finds a mixed relationship between public services and tax morale: individuals who are more satisfied with public health provision have a lower level of tax morale, while satisfaction with education shows no link with tax morale. The inconsistency of results for healthcare may be linked to the way in which this is financed in Latin America, with many countries having mixed financing models including social security payments, general taxation and significant out-of-pocket expenses which result in many individuals in Latin America paying significantly for healthcare individually, rather than through taxation (Izquierdo, Pessino and Vuletin, 2018[23]). While the relationship between specific services and tax morale was not clear for Latin America, trust in government and the perceived level of corruption both showed an impact on tax morale, highlighting how government performance overall appears to influence tax morale in Latin America.
1.4. Tax morale by region - Asia
Data is more limited for Asia, but it is possible to identify a number of factors that affect tax morale. The Asiabarometer has not been updated since 2005-2007 and cannot be used to track tax morale over time. It is, however, possible to use World Values Survey data from Asia to identify the factors that appear to drive tax morale, albeit with fewer observations than with the regional surveys. From this data, it can be seen that taxpayers having preferences for more equal and meritocratic societies have higher tax morale (Figure 1.8). Individuals who are older, more educated and female also show higher tax morale - once again highlighting the potential for taxpayer education programmes for younger, less educated citizens. The limited data in the World Values Survey means it is not possible to test for the strength of the fiscal contract in Asia.
Figure 1.8. Factors associated with tax morale in Asia
1.5. Conclusions and next steps
These findings support the existence of a fiscal contract, but suggest its strength may vary between regions and over time. The positive findings from Africa and Asia point clearly to the existence of the fiscal contract as an influence on tax morale, supporting previous research (Torgler, 2005[24]). While the link with public services is strong in Africa, it appears weaker in Latin America (and weaker today than previously reported), suggesting that the strength of the fiscal contract is variable across regions and over time.
There are a broad range of tools and approaches that can help build confidence in the fiscal contract and trust in government on taxation. Many include more explicitly linking the revenue and expenditure sides of the fiscal framework: transparent budgeting; space for parliaments, courts, civil society and media to hold governments accountable, both in addressing corruption and for collecting taxes and spending them effectively. This is a well-known general governance agenda, but there are some specific actions where additional progress could be made:
Taxpayer education programmes can be informed by tax morale data. The findings in this report highlight how tax morale differs across different sections of society, suggesting diverse needs for taxpayer education. For example, lower tax morale among young people suggests a role for taxpayer education programmes to be integrated into school curricula. The 2015 publication Building Tax Culture, Compliance and Citizenship (OECD/FIIAPP, 2015[25]) showcases a number of taxpayer education initiatives, including several (e.g. Kenya. Malaysia, and Morocco) that bring taxpayer education to children. Others (e.g. Burundi, Costa Rica, and Guatemala) seek to combine understanding of the fiscal contract with improving the legitimacy of the tax authority and the ease of paying taxes. And others aim also to recognise and reward exemplary taxpayers (e.g. the Korean National Tax Service).
Careful consideration is needed before introducing hypothecated taxes. While hypothecated taxes may provide a useful tool to create political space for tax reform in the short-run, there are long-term challenges. Pros and cons need to be carefully analysed and reflected in the design and duration of hypothecated taxes, if adopted. Tax morale dynamics should be included in the design of hypothecated taxes as taxpayers might be more willing to pay more taxes for certain public services and less for others.
Build taxpayer profiles. Analysing the attitudes towards and perceptions of taxation (including through the use of survey data as in this report), tax administrations in all countries can develop comprehensive profiles of their taxpayers. Such profiles could provide information on groups which are most resistant to paying tax, as well as those which are most easily convinced of the benefits of taxation, and help develop approaches to engage all taxpayers more effectively in the tax system. For instance, a better understanding of the drivers of tax morale for the wealthy (including responses to amnesty programmes), and their influence on the rest of the taxpayers could provide valuable insights.
Support the development of tax administrations, especially in improving the ease of paying taxes. There is already significant work being undertaken in this area, for example the International Monetary Fund (IMF) and World Bank Group, which both have several decades of experience of providing technical assistance in tax administration, while the more recently established African Tax Administration Forum (ATAF) has driven pan-African co-operation in tax administration development. More can be done. For example, the 50 tax administrations of the OECD’s FTA have extensive experience in this area to be shared. Through the development of the FTA’s Capacity Building Network there are increased opportunities for this experience to be made available to developing countries. Improved use of technology is likely to be at the heart of efforts to improve the ease of paying taxes (Bird and Zolt, 2008[10]), as well as strengthening the integrity of tax administrations. As such, helping developing countries design and manage the use of technology in tax administrations should be a key part of support. More broadly there may be further applications of technology to increase citizen engagement in both how taxes are raised and spent, that could impact tax morale.
Deepen the understanding of the drivers of trust in government. Trust in government is a broad concept, covering trust in the political system, rule of law etc.; as such the relationship between trust in government and tax morale needs to be further explored, and disaggregated. The OECD TrustLab project7 has undertaken some work looking at the components of trust (though not in relation to tax morale) and has run behavioural experiments across OECD member countries. Results show that institutional characteristics (i.e. competences and values, perceptions about public service delivery, and integrity) account for about 40% of differences in trust in government, but there are significant variations between countries (Murtin et al., 2018[26]). Expanding the TrustLab experiments to developing countries would help identify the determinants of trust in developing countries, enabling policies to be developed which specifically target building trust.
Increase the research and data available on developing countries, especially on tax and gender. While there has been a range of research on tax morale spanning over 40 years, there has been relatively little research focussing on developing countries. Addressing this gap will require greater cooperation and sharing of data between developing countries and researchers, from a range of disciplines (including economics, psychology, political science, anthropology and sociology). The OECD may be well placed to help facilitate such links (and has already done so in some cases following the January 2019 tax morale conference8). Different levels of tax morale between men and women are likely to be indicative of a deeper issue in how the fiscal system treats men and women differently. The 2018 Platform for Collaboration on Tax conference on Taxation and the Sustainable Development Goals identified the need to address both implicit and explicit gender biases in taxation, reiterating that gender disaggregated data is a vital starting point for further research (Platform for Collaboration on Tax, 2018[27]).
A universal consideration is that in-depth country (or sub-national) level analysis is a vital precondition to inform policy changes. As highlighted by the World Bank approach (see Figure 1), tax morale sits within complex inter-relationships between enforcement, facilitation and trust which affect compliance; countries (or sub-national authorities) should therefore look to take a holistic approach that identifies the context of tax morale locally. More in-depth country-level surveys can be part of this process, used to identify tax morale factors at the local level and, where relevant, at sub-national level. Country-level surveys could also include surveying which taxes individuals are actually paying (and how much) to increase the understanding of how the various taxes and other payments (including to non-state actors) are perceived by individuals in developing countries. Randomised control trials have also been used in some countries, including Rwanda and a number of Latin American countries (Castro and Scartascini, 2013[28]; Kettle et al., 2016[17]; Mascagni et al., 2016[29]), to measure the effects of different methods of tax enforcement. Results have been promising, suggesting other countries could benefit from such studies.
References
[18] Akpan, I. and K. Sempere (2019), “Hidden Inequalities: Tax Challenges of Market Women in Enugu and Kaduna States, Nigeria”, https://opendocs.ids.ac.uk/opendocs/handle/123456789/14500 (accessed on 2 July 2019).
[1] Ali, M., O. Fjeldstad and I. Sjursen (2014), “To Pay or Not to Pay? Citizens’ Attitudes Toward Taxation in Kenya, Tanzania, Uganda, and South Africa”, World Development, Vol. 64, pp. 828-842, http://dx.doi.org/10.1016/J.WORLDDEV.2014.07.006.
[10] Bird, R. and E. Zolt (2008), “Technology and Taxation in Developing Countries: From Hand to Mouse”, Law-Econ Research Paper No. 08-07, http://dx.doi.org/10.2139/ssrn.1086853.
[22] Carrillo, P., E. Castro and C. Scartascini (2018), “Do Rewards Work?: Evidence from the Randomization of Public Works”, SSRN, http://dx.doi.org/10.2139/ssrn.3103797.
[14] Castro, L. and C. Scartascini (2015), “Tax compliance and enforcement in the pampas evidence from a field experiment”, Journal of Economic Behavior & Organization, Vol. 116, pp. 65-82, http://dx.doi.org/10.1016/j.jebo.2015.04.002.
[28] Castro, L. and C. Scartascini (2013), “Tax Compliance and Enforcement in the Pampas: Evidence from a Field Experiment”, Working Paper Series, No. 472, IDB, http://www.iadb.org (accessed on 27 November 2018).
[15] Coleman, S. (2007), “The Minnesota Income Tax Compliance Experiment: Replication of the Social Norms Experiment”, SSRN Electronic Journal, http://dx.doi.org/10.2139/ssrn.1393292.
[2] Cummings, R. et al. (2009), “Tax morale affects tax compliance: Evidence from surveys and an artefactual field experiment”, Journal of Economic Behavior & Organization, Vol. 70/3, pp. 447-457, http://dx.doi.org/10.1016/j.jebo.2008.02.010.
[21] Daude, C. and Á. Melguizo (2010), “Taxation and More Representation?: On Fiscal Policy, Social Mobility and Democracy in Latin America”, OECD Development Centre Working Paper series, No. 294, OECD, https://www.oecd-ilibrary.org/docserver/5km5zrrs9bbt-en.pdf?expires=1544198553&id=id&accname=ocid84004878&checksum=F5E4B51BCF133B26DE8D8EED55EEC708.
[23] Izquierdo, A., C. Pessino and G. Vuletin (2018), Better Spending for Better Lives: How Latin America and the Caribbean Can Do More with Less, IDB, https://webimages.iadb.org/publications/english/document/Better-Spending-for-Better-Lives-How-Latin-America-and-the-Caribbean-Can-Do-More-with-Less.pdf.
[17] Kettle, S. et al. (2016), Behavioral Interventions in Tax Compliance Evidence from Guatemala, http://econ.worldbank.org.
[13] Mascagni, G., C. Nell and N. Monkam (2017), “One Size Does Not Fit All: A Field Experiment on the Drivers of Tax Compliance and Delivery Methods in Rwanda”, SSRN Electronic Journal, No. 48, ICTD, http://dx.doi.org/10.2139/ssrn.3120363.
[29] Mascagni, G. et al. (2016), “The Carrot and the Stick: Evidence on Voluntary Tax Compliance from a Pilot Field Experiment in Rwanda”, SSRN Electronic Journal, http://dx.doi.org/10.2139/ssrn.3120358.
[12] McCleary, W., McCleary and William (1991), “The Earmarking of Government Revenue: A Review of Some World Bank Experience”, World Bank Research Observer, Vol. 6/1, pp. 81-104, https://econpapers.repec.org/article/oupwbrobs/v_3a6_3ay_3a1991_3ai_3a1_3ap_3a81-104.htm (accessed on 13 December 2018).
[9] Moore, M., W. Prichard and O. Fjeldstad (2018), Taxing Africa: Coercion, Reform and and Development, Zed Books.
[26] Murtin, F. et al. (2018), “Trust and its determinants: Evidence from the Trustlab experiment”, OECD Statistics Working Papers, No. 2018/2, OECD Publishing, Paris, https://dx.doi.org/10.1787/869ef2ec-en.
[6] OECD (2019), OECD Secretary-General Report to the G20 finance ministers and central bank governors, http://www.oecd.org/tax/oecd-secretary-general-tax-report-g20-finance-ministers-june-2019.pdf (accessed on 16 July 2019).
[7] OECD (2019), Public comments on the discussion draft on “What is driving tax morale?”, https://www.oecd.org/tax/tax-global/compilation-public-comments-what-is-driving-tax-morale.pdf (accessed on 12 July 2019).
[4] OECD (2013), What drives tax morale?, http://www.oecd.org/ctp/tax-global/what-drives-tax-morale.pdf.
[3] OECD (2017a), Behavioural Insights and Public Policy: Lessons from Around the World, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264270480-en.
[8] OECD/ATAF/AUC (2018), Revenue Statistics in Africa 2018, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264305885-en-fr.
[5] OECD/CAF/UN ECLAC (2018), Latin American Economic Outlook 2018: Rethinking Institutions for Development, OECD Publishing, Paris, https://dx.doi.org/10.1787/leo-2018-en.
[25] OECD/FIIAPP (2015), Building Tax Culture, Compliance and Citizenship: A Global Source Book on Taxpayer Education, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264205154-en.
[27] Platform for Collaboration on Tax (2018), Taxation and SDGs, https://www.oecd.org/tax/tax-global/first-global-conference-of-the-platform-for-collaboration-on-tax-february-2018.pdf (accessed on 18 July 2019).
[11] Prichard, W. (2015), Taxation, Responsiveness and Accountability in Sub-Saharan Africa, Cambridge University Pres, Cambridge, http://www.cambridge.org.
[16] The Behavioural Insights Team (2014), Results from BIT tax trial in Guatemala | The Behavioural Insights Team, https://www.bi.team/blogs/results-from-bit-tax-trial-in-guatemala/ (accessed on 14 March 2019).
[24] Torgler, B. (2005), “Tax morale and direct democracy”, European Journal of Political Economy, Vol. 21/2, pp. 525-531, http://dx.doi.org/10.1016/j.ejpoleco.2004.08.002.
[20] UN (2017), World Population Prospects 2017, https://population.un.org/wpp/ (accessed on 8 January 2019).
[19] Van Den Boogaard, V. (2018), “Gender and the Formal and Informal Systems of Local Public Finance in Sierra Leone”, ICTD Working Paper, No. 87, Institute of Development Studies, https://opendocs.ids.ac.uk/opendocs/bitstream/handle/123456789/14127/ICTD_WP87.pdf?sequence=1&isAllowed=y (accessed on 5 December 2018).
Notes
← 1. Based on the response to the question “Do you justify cheating on taxes if you have the chance?” For further detail, see Annex A.
← 2. Excluding Chile and Mexico, which are included in Latin America.
← 3. ‘Small taxes’ cover a range of taxes, in three broad categories; Formal subnational taxes and charges, formal taxes and payments on small and micro-businesses, informal taxes – comprising both illicit formal taxes and revenues collected by non-state agents and organisation.
← 4. Earmarking can have a negative impact on budget flexibility.
← 5. The Behavioural Insights Team, originated within the UK government and now an international company, seeks to generate and apply behavioural insights to inform policy and improve public services – www.bi.team.
← 6. Results differs from section 1.1 as they are from a different data sources (WVS), different periods of analysis and different question (see table A.2).
← 8. The Task Force on Tax and Development organised this conference on the role of tax morale in development to discuss the preliminary findings in this report. The event brought together over 125 delegates including government representatives, leading academics and researchers, regional tax organisations, business and civil society, in a multi-stakeholder environment to share experiences and thinking on tax morale. For more information see: http://www.oecd.org/tax/role-of-tax-morale-in-development-conference-january-2019.htm.