WPS8391 Policy Research Working Paper 8391 Taxation and the Shadow Economy How the Tax System Can Stimulate and Enforce the Formalization of Business Activities Rajul Awasthi Michael Engelschalk Governance Global Practice March 2018 Policy Research Working Paper 8391 Abstract Cash transactions for goods and services in which no receipts It argues that the many new and sometimes innovative are issued greatly increase the risk of tax evasion. Despite the approaches developed to support the formalization of cash availability of banking services and alternative payment, key transactions will have little impact on the shadow econ- sectors of the economy remain largely cash-based in almost omy if applied in isolation. A successful strategy to tax cash all developing countries. This paper shows the apparent economy businesses and transactions requires a holistic strong negative correlation between the use of electronic approach to compliance management in which traditional or formal payments and the size of the shadow economy monitoring and enforcement tools, such as enabling tax and reviews the approaches used by tax policy makers and administrations to access taxpayer data and match informa- administrators to achieve better control of cash transactions. tion from various public and private sources, play a key role. This paper is a product of the Governance Global Practice. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted at rawasthi@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Taxation and the Shadow Economy How the Tax System Can Stimulate and Enforce the Formalization of Business Activities Rajul Awasthi1 Michael Engelschalk2 Keywords: Tax policy; Tax administration; Tax compliance; Shadow economy; Tax evasion; Informal sector JEL codes: H26, H31, H32, K34                                                                    1  Senior Public Sector Specialist, Global Tax Team, Global Domestic Resource Mobilization Unit (GGOGT), Governance Global  Practice, Equitable Growth, Finance, and Institutions Vice Presidency, The World Bank.  2  Senior Tax Consultant, Former Head of Revenue Administration and Policy Group, PREM Network, The World Bank.    1 Section 1   Impact of the Shadow Economy and Other Factors on Taxation  Tax administrations increasingly worry about the complex problem of the shadow economy; in particular, they realize that valuable tax revenues are lost from noncompliant members of the shadow economy. On average, one-third of the world economy is in the shadows.3 The highest incidence is in Sub-Saharan Africa, where the weighted average size of the shadow economy (as a percentage of GDP) is 37.6 percent. In Europe and Central Asia (mostly transition countries), the weighted average is 36.4 percent, and in high income OECD countries it is 13.4 percent.4 How do we define shadow economy?5 For this paper, we propose to follow the “narrow” definition used by Schneider, Buen, and Montenegro: “The shadow economy includes all market-based legal production of goods and services that are deliberately concealed from public authorities to avoid payment of income, value added or other taxes; to avoid payment of social security contributions; having to meet certain legal labor market standards, such as minimum wages, maximum working hours, safety standards, etc.; and complying with certain administrative procedures, such as completing statistical questionnaires or administrative forms.”6 This definition suits the main objective of this paper, which is to identify strategies tax administrations can adopt to tackle the shadow economy with the goal of plugging tax leaks and augmenting revenues. Various papers and studies have shown that high levels of tax complexity, weak or ineffective tax enforcement, and lack of incentives to carry out business transactions and make payments through formal banking channels are among the factors affecting the level of informality. Another important factor is the level of trust taxpayers have in government.7 These factors can all be influenced by tax policies and tax administration measures. Other factors likely to impact the level of the shadow economy include the state of the official economy, labor market participation rates, quality of public-sector services, and a “culture of tolerance,” but as these factors are beyond the control of tax administrations they are not discussed in this paper. In general, a strategy to address the factors that can help reduce the size of the shadow economy needs to be comprehensive, involving all stakeholders, for e.g., the central bank, the banking sector, the ministry of finance or economy, and the tax administration. Other governance-related factors considered include the six Worldwide Governance Indicators (WGI).8 Of these, two, Regulatory Quality and Control of Corruption, appear to have a measurable impact on the shadow economy and the level of taxation. We also explored the impact of the competitiveness factors analyzed by the World Economic Forum report; of the 12 factors considered there, one, Technological Readiness, has significant bearing on the shadow economy and taxation. First, we analyzed the relationship between the ratios of tax collection to gross domestic product (GDP) and the ratios of the shadow economy to GDP, to confirm the presence of a negative relationship: high levels of the shadow economy correlate with low levels of tax collection. This comparison used the tax- GDP ratios for the year 2007 (from the IMF Government Financial Statistics data set) and compared them                                                              3. F. Schneider, Andreas Buen, Claudio Montenegro 2010. 4 See Appendix B for a brief description of the model used by Schneider et al (2010) to estimate the size of shadow economies in their paper. 5 This paper is not going into broader definitions such as “informality”, in order to keep the focus strictly to the shadow economy. The informal economy may in fact be larger than the shadow economy, particularly in low income countries. Not all participants in the informal economy are evading taxes – they may be below tax or registration thresholds, and they may not have access to identification or financial services. 6. Ibid. 7  Forthcoming: “Innovations in Tax Compliance”, Verhoeven, M, et al, World Bank Working Paper, 2018. 8. These factors are Voice and Accountability, Political Stability and Absence of Violence/Terrorism, Government Effectiveness, Rule of Law, Regulatory Quality, and Control of Corruption.   2 with the shadow economy–GDP ratios for the same year (as computed by Schneider, Buehn, and Montenegro 2010). The findings were straightforward and pointed clearly to a negative relationship, as seen in Table A.1 in the Appendix.   The fundamental regression is the relationship between the ratios of tax collection to Gross Domestic Product (GDP) and the ratios of the shadow economy to GDP. From this regression, we see that the coefficient of shadow economy to GDP has a low p-value (< 0.05) and is statistically significant, meaning that changes in ratios of the shadow economy to GDP are significantly correlated to changes in ratios of tax collection to GDP. The value of the coefficient is -0.278; that is, for every incremental percent of ratio of shadow economy to GDP, we can observe that the ratio of tax collection to GDP decreases by an average of -0.278 percent.9 Figure 1.1 Scatter Tax to GDP Ratio: Shadow to GDP Ratio 50 40 30 20 10 0 10 20 30 40 50 60 Shadow-GDP Ratio Tax-GDP Ratio Fitted values Source: The authors. We also examined the relation between tax-GDP and shadow-GDP, controlling for per capita GDP. We find that the negative relation between the two key ratios sustains, and is significant at the 5 percent level. We get a coefficient of –0.12 which indicates that a 1 percent reduction in the                                                              9  The authors would like to clarify that the relationships presented here do not imply causality, rather they merely  serve to indicate that shadow economy ratios tend to be correlated with several indicators which are statistically  significant.     3 shadow economy ratio is correlated with an estimated 0.12% increase in the tax ratio’s numerical value.10 Next, we examined tax regime complexity and its correlation with shadow economies. The hypothesis studied was that greater tax complexity, by imposing heavier compliance burdens on taxpayers, disincentivizes tax compliance, encouraging taxpayers to slip into the shadows. We measure tax complexity by determining the number of hours needed to comply with a tax regime, as defined in the Doing Business report of the World Bank Group. The Doing Business report is published with a two-year time lag; Doing Business 2009 thus presents data from the year 2007. Using that year’s data to compare tax regime complexity with the shadow economy ratios computed by Schneider, Buehn, and Montenegro (2010), we get a result in line with our hypothesis: Figure 1.2 Positive Correlation in Complexity and Informality 1200 1000 Time to comply in hours 800 600 400 200 0 0 10 20 30 40 50 60 70 Shadow Economy as a percent of GDP Sources: World Bank Group 2009; Schneider, Buen, and Montenegro 2010 A third factor found to be significant is a country’s level of electronic payments. Studies show that the higher the level of electronic payments—and thus with correspondingly lower levels of cash payments—                                                              10       4 the lower the level of the shadow economy. A study by A. T. Kearney and F. Schneider for the European economy confirms this hypothesis.11 Figure 1.3 Share of Shadow Economy Our study also considered six dimensions of governance, measured by the Worldwide Governance Indicators (WGI),12 in 143 countries for the year 2007, to examine the data and run regressions between these governance factors and both the shadow economy–GDP ratios and the tax-GDP ratios. Two factors— Regulatory Quality and Control of Corruption—were found to have the expected correlation with the ratios of tax collection to GDP and shadow economy to GDP. The values of the correlation coefficients between shadow economy–GDP and Regulatory Quality (RQ) and Control of Corruption (CC) are -0.6435 and -0.6763, respectively, showing strong negative correlations between shadow economy–GDP and Regulatory Quality (RQ) and between shadow economy–GDP and Control of Corruption (CC). (See Table A.2 in the Appendix.) The regression of shadow economy–GDP with Control of Corruption (CC) showed a p-value at 0, significantly related to changes in ratios of the shadow economy to GDP. A coefficient of -5.94 means a unit increase of Control of Corruption (CC) will be accompanied by an observation of a lower shadow economy by 5.94 units. (See Appendix Table A.3.) The following scatter charts demonstrate that good quality regulation and a higher ability to control corruption are correlated with lower measures of the shadow economy.                                                              11. A. T. Kearney and F. Schneider 2009.   12  Please see the WGI site, http://info.worldbank.org/governance/wgi/#home; for a detailed description of the  methodology and analytical issues, please go to:  https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1682130    5 Figure 1.4 Scatter Shadow to GDP Ratio: Regulatory Quality Figure 1.5 Scatter Shadow Economy to GDP Ratio: Control of Corruption 60 60 50 50 40 40 30 30 20 20 10 10 -2 -1 0 1 2 -2 -1 0 1 2 3 Regulatory Quality Control of Corruption Shadow-GDP Ratio Fitted values Shadow-GDP Ratio Fitted values Source: The authors. Source: The authors. When we analyze the relationships between Regulatory Quality (RQ) and levels of tax collection, as well as between Control of Corruption (CC) and levels of tax collection, we find a positive relationship between them. This can be observed visually in the scatter plots below: Figure 1.6 Scatter Tax to GDP Ratio: Regulatory Quality Figure 1.7 Scatter Tax to GDP Ratio: Control of Corruption 50 50 40 40 30 30 20 20 10 10 0 0 -2 -1 0 1 2 -2 -1 0 1 2 3 Regulatory Quality Control of Corruption Tax-GDP Ratio Fitted values Tax-GDP Ratio Fitted values Source: The authors. Source: The authors. (See also Table A.4 in the Appendix.) By running a regression of tax-to-GDP with Regulatory Quality (RQ) and Control of Corruption (CC) we found that Regulatory Quality (RQ) and Control of Corruption (CC) also are correlated with levels of tax   6 collection. The coefficients are 1.473 and 2.8, meaning that a unit increase of Regulatory Quality (RQ) and Control of Corruption (CC) can go hand in hand with an increase in the observed tax-to-GDP ratio by 1.473 and 2.8 units, respectively. Also, the p-value of Control of Corruption (CC) is low at 0.012, which means changes in the Control of Corruption (CC) score are significantly correlated to changes in ratios of tax collection to GDP.   Exploring the impact of competitiveness factors analyzed by the World Economic Forum report revealed that of the 12 pillars (factors) considered there, one, Technological Readiness, is significantly correlated to the shadow economy and to taxation.13 The correlation coefficients and regression between shadow-to- GDP and Technological Readiness are demonstrated in Table A.5 in the Appendix. Technological Readiness is an important factor likely to have a major impact on the shadow economy: the coefficient of –3.88 means that a unit improvement of Technological Readiness is correlated with a reduction of the shadow economy by an average of -3.88 units. In addition, Technological Readiness shows a p-value at 0, indicating the high statistical significance of this factor. Technological Readiness also shows a strong positive correlation with the tax-to-GDP ratio, as indicated in Table A.6 of the Appendix. The coefficient is 1.284, meaning that for a unit improvement of Technological Readiness the observed tax-to-GDP Ratio increases by 1.284 units. The low p-value at 0.001 also indicates high statistical significance of Technological Readiness. Tax‐GDP Ratio vs. Technological Readiness 50.0 45.0 40.0 35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 Source: IMF Data; World Economic Forum Data; Author’s calculations                                                              13. The 12 pillars or factors are Institutions, Infrastructure, Macroeconomic Environment, Health and Primary Education, Higher Education and Training, Goods Market Efficiency, Labor Market Efficiency, Financial Market Development, Technological Readiness, Market Size, Business Sophistication, and Innovation. Score ranges from 1 (worst) to 7 (best).       7 This finding is very interesting. In the modern, digital world, tax administrations are increasingly run on ICT platforms and use modern technology to enforce tax compliance. Modern ICT systems allow tax administrations to access data on taxpayers’ financial transactions with banks and other institutions on a regular and automatic basis. These data can be analyzed automatically by IT systems to generate taxpayer risk profiles and to aid in risk management. Tax administrators use technological means as well, such as electronic invoicing systems, to ensure that sales and profits do not go underreported or unreported. These measures can significantly increase the amount of tax that can be forced out of the shadow economy and into tax compliance. Clearly, the higher an economy’s Technological Readiness, the better and more effective such measures can be. The analysis in this section suggests that the problems of the shadow economy are best addressed through a two-pronged approach to tax administration: (a) Strategies to improve tax enforcement against shadow economy transactions, particularly measures to plug tax evasion directly related to transactions with the shadow economy, such as underreporting or non-reporting of sales or inflating expenses through false invoices. The use of ICT-based technological solutions is most useful for effective implementation of this strategy. (b) Tax measures—both policy and administrative—to incentivize formal payments for business transactions through banking channels, including electronic payments and the use of “plastic money.” The following sections outline some experiences of various countries that implemented these strategies.   Section  2  Strategies  to  Improve  Tax  Enforcement  against  Shadow  Economy  Transactions  The terms shadow economy and cash economy are largely used synonymously. Activities in the shadow economy are paid in cash to prevent the tracking of money flows. At least in theory, therefore, an obstacle to shadow economy activities can be created by banning the use of cash as a payment option.14 Several countries in the European Union have established quantitative limits on the permissible amount that can be spent in cash purchases of goods and services. Such limits are meant to force individuals to use electronic means of payment—such as credit, debit, and ATM cards—which necessarily go through the banking system and hence are far easier to track. Table 2.1 Cash Payment Restrictions in the European Union Belgium €3,000 January 1, 2014 Bulgaria BGN 10,000 (≈€5,112) July 1, 2011 Czech CZK 350,000 (≈€12,763) January1, 2013 Republic Denmark DKK 10,000 (≈€1,340) July 1, 2012                                                              14In a recent book, “The Curse of Cash”, Kenneth Rogoff has made a case for getting rid of most paper money. According to him, cash is feeding tax evasion, corruption, terrorism, the drug trade, human trafficking, and the global underground economy. His book offers a plan for phasing out most paper money—while leaving small-denomination bills and coins in circulation indefinitely—and addresses the issues the transition will pose.   8 France €3,000 (residents and €15,000 (nonresident consumers) January 1, 2002 nonresident traders Greece €1,500 January 1, 2011 Hungary HUF 1.5 million (≈€5,000) (legal January 1, 2013 persons) Italy €999.99 December 6, 2012 Portugal €1,000 May 14, 2012 Slovak €5,000 €15,000 (natural persons who are January 1, 2013 Republic not entrepreneurs) Spain €2,500 (residents) €15,000 (nonresidents) November 19, 2012 Source: European Consumer Center France 2014, Veber and Brosche 2013. Similar provisions exist in other regions as well. In Peru, for example, pursuant to Law No. 28194,15 obligations exceeding S/. 3,500,16 approximately US$1,000, fulfilled through cash payments must be made via bank account deposits, wire transfers, payment orders, credit cards, non-negotiable checks, or other means of payment provided by entities of the Peruvian financial system. If obligations fulfilled through cash payments are not made using such means, debtors may not deduct them as expenses or use them as credits for tax purposes. Turkey, as part of an initiative aimed at reducing the use of cash transactions, has also established new payment requirements and mechanisms. Payments of amounts over TRY 8,000 (US$ 2,300) and any rental payments over TRY 500 must be made through the banking system or post offices. Mexico has taken an alternative approach to restricting the use of cash in the financial sector as part of its National Anti-Money Laundering (AML) and Countering the Financing of Terrorism (CFT) strategies. Limits are established on the amounts of US dollars in cash that banking institutions may accept from clients or occasional customers through deposits or other transactions, and a reporting obligation for banking institutions was established for cash transactions involving US dollars that exceed the predetermined limits established in the regulations. Almost identical rules have also been imposed on currency exchange houses and securities brokerage firms. New regulations have been put in place to identify clients performing transactions in smaller amounts (starting at US$500), to counter structuring transactions to avoid detection under the reporting requirements. The most straightforward approach to imposing additional costs on access to cash is to tax cash withdrawals from bank accounts. A withholding tax on cash withdrawals, however, may be ineffective from the point of view of both generating revenue and of discouraging cash transactions. In fact, such a tax is highly unpopular with the business community, as traders are concerned that the tax reduces cash available for purchases and thus has a negative impact on trade. Business communities therefore tend to oppose the introduction of such taxes; examples abound, including India, Pakistan, Jamaica, and Greece. Far more important, however, is the risk that the tax on withdrawals contradicts attempts to increase the use of banks                                                              15. Decreto Supremo N.150-2007-EF. 16. S/. stands for nuevo sol, the Peruvian currency.   9 for day-to-day business and supports the rather widespread perception in many developing countries that using a bank account has more disadvantages than advantages. Thus, a salaried employee may prefer to receive his salary in cash at the end of the month if a payment into his bank account risks facing an additional tax burden. Rather than making cash transactions less popular, such plans make the use of the banking system less popular overall. On the other hand, because the tax burden imposed is limited, it is highly questionable that such taxes promote change. Compared to other charges, such as banking fees or credit card charges, withdrawal taxes tend to be negligible. This can be illustrated using the case of Ireland, where the new charge on cash withdrawals remains much lower than the stamp duty on credit cards, which is six times the maximum of the cash withdrawal amount. More effective in the longer term may be initiatives such as a recent EU regulation halving interchange charges on credit cards to reduce obstacles preventing shop owners from accepting card payments; a Danish plan to abolish the legal obligation for certain businesses, such as clothing retailers, gas stations, and restaurants to accept payment in cash; or the Mexican effort to simplify the opening of small-value bank accounts and to increase the interest among banks in serving small business clients and employees. Country practices show that neither cash-in nor cash-out taxation through financial institutions achieves the results expected. On the contrary, the potential detrimental effects of such taxation and the considerable opposition of businesses and consumers to such cash-use taxes make this approach a highly risky exercise. International experience suggests that a reliance on reporting instead of withholding requirements has a greater potential for achieving some control over cash economy transactions. This change in approach can be seen in India’s reform process. Similar to Pakistan, India had introduced a Bank Cash Transaction Tax (BCTT) in 2005, under which all cash withdrawals by individuals amounting to more than 50,000 INR (US$ 780) per day were taxed at 0.1 percent. The tax was withdrawn in 2009; the finance minister stated that “the information is also being gathered through other instruments introduced in the last few years.”17 (See the discussion of reporting requirements below.) Mexico’s cash deposit tax represents another interesting experience. In July 2008, Mexico introduced the Cash Deposits Tax (Impuesto de Depositos en Efectivo, or IDE), which was withheld by the financial institution on the amount in excess of 25,000 MXN (US$ 1,400) deposited in cash in a given month. Mexican authorities found the tax useful for containing informality, but following the 2014 Tax Reform bill package,18 the Mexican flat tax and tax on cash deposits were repealed, because they had not met expectations. The assumption was that the IDE discouraged small business operators from putting cash earnings into bank accounts and actually increased rather than reducing undeclared transactions. Revenue collection from IDE decreased from 5.4 billion pesos in the first seven months of 2011 to just 1.9 billion pesos in the same period of 2012.19 The IDE was replaced by new reporting requirements; now, institutions that are part of the financial system must submit an annual report to the tax administration of all cumulative monthly deposits in excess of MXN15,000 made to the accounts of individual taxpayers and legal entities. The ability of tax administrations to track shadow economy activities increases substantially when any transaction made leaves a paper trail. Such trails may consist of either a regular invoice issued by the business for a good or service or of a recorded payment for the good or service. The starting point for developing a paper trail for cash transactions is the requirement to issue a tax invoice for any sale with a price above a certain petty amount. Many countries have established such legal requirements under their tax laws; however, such requirements are subject to the risk that both businesses and customers will have little interest in or inclination toward compliance. Physical control to ensure compliance with invoice                                                              17. Shome 2011. 18. These reforms have been in place since January 1, 2014. The Tax on Cash Deposits Law was approved by the Mexican Congress on October 31, 2013, and was published in the Official Gazette on December 11, 2013. 19. Burgess et al. 2013.   10 issuance obligations is burdensome and costly. Nevertheless, some tax administrations have substantially increased their resources for performing spot-checks of businesses, including making anonymous test purchases. Spain, for example, has almost doubled its spot-checks at cash businesses from 8,700 in 2011 to 15,700 in 2012 to better detect irregularities.20 Tanzania is implementing an extensive physical control system of businesses as part of its Block Management System (BMS). This effort can even include prolonged supervision of a specific business by a tax inspector. Italy and some Latin American countries, such as Argentina and Ecuador, have tried for many years to extend the control mechanism to customers. Customers are required to carry invoices with them for a certain time after leaving a shop, and the tax administration (or in Italy the tax police or Guardia di Finanza), can stop customers in the street and ask for the invoice. Such an approach, of course, is resource-intense and risks inefficiency.21 It also tends to be highly unpopular among customers, which is why it was discontinued in Italy, where sanctions on customers for being unable to show an invoice were abolished in 2003.22 Nevertheless, some countries have introduced obligations for private customers to retain receipts from purchases; in Malta, for example, customers must retain a receipt for period of 24 hours from the time of the purchase. With no effective control mechanism in place, however, the obligation remains unenforceable. Overall, therefore, experience shows that compliance control using invoice issuance obligations generally works better when performed by the business rather than the customer. Use of point-of-sale (POS) electronic tracking of invoices issued by businesses has become widespread in efforts to improve compliance control. Tremendous progress has been made in recent years in the design of such systems, from the first generation of simple stand-alone cash machines with a built-in register to sophisticated electronic fiscal device (EFD) systems that directly transfer data to a tax administration database. An overview of the different types of EFDs is provided in an IMF working paper.23 Modern POS systems ensure the real-time transfer of sales data to the tax administration database and allow ongoing monitoring of individual business transactions and fluctuations in the sales volume. They replace the previously cumbersome paper-based process of allocating tax invoice numbers to businesses through the automatic generation of a unique tax number for each invoice issued. Figure 2.1 Automatic Allocation of a Unique Invoice Number in Croatia                                                               20. Brat 2013.   21. See, however, Berhan and Jenkins 2005, who describe the approach as quite effective. 22. Fabbri and Hemels 2013, 430. 23. Casey and Castro 2015.   11 Source: Kudeljan 2015. POS System Introduction Mandatory use of POS systems partly shifts the costs of invoice compliance monitoring from the tax administration to the taxpayer, as the installation and maintenance of such systems is an additional cost to the business. As these costs can be a considerable burden on small businesses, mandatory use of EFDs has in many countries been limited to medium and large businesses above the VAT threshold. Nevertheless, some countries, such as the Republic of Korea, have expanded mandatory system use to small traders operating below the VAT threshold. Expansion to smaller businesses is often implemented in phases. Tanzania, for example, first introduced a mandatory EFD use for businesses with turnover above the VAT threshold of T Sh 40 million24 (US$18,500), and in a second stage they were made compulsory for traders with annual turnover above T Sh 14 million (US$6,300), traders operating in prime areas, and traders operating in certain segments, such as petrol stations, bars and restaurants, and mobile phone shops. Croatia introduced compulsory cash register use in three stages: the first stage targeted large businesses and businesses in the accommodation and catering sector; the second stage extended the requirement to wholesale and retail traders and self-employed services; and the third stage covered all other businesses, with a few exceptions, such as passenger transport or sale of homegrown fruits and vegetables. Some countries, including Greece, Romania, Italy, Poland, and Sweden, have made the use of certified systems mandatory only in cash-intensive sectors. In Sweden about 40 percent of all cash registers are operated in the restaurant and hair salon segments. Belgium and the Canadian province of Quebec each introduced mandatory EFD use in one business sector only (in Quebec, restaurants, catering services, and bars; in Belgium, the hotel and catering industry).                                                              24. T Sh stands for the Tanzanian shilling.   12 The introduction of mandatory POS system use may actually partly increase the compliance management challenge for a tax administration. Resistance to EFD introduction tends to be high in the targeted business segments. Reasons for this resistance are generally very similar worldwide: (i) the high costs of EFDs, an expense borne in most countries by the business owner; (ii) problems with EFD supply and installation; and (iii) to a lesser extent, problems with EFD operation, for example, due to frequent power outages. In addition to overcoming user resistance, an EFD usage control mechanism must be put in place. Finally, experience in OECD countries has shown a high risk of EFD manipulation through specialized zapper software; control strategies must therefore extend beyond simple reviews of the level of device use to checking for possible system manipulation. Introducing EFD use therefore requires a major change management program addressing resistance in the business community. Business operators may perceive the requirement to operate EFDs as an additional tax compliance burden. This response is reflected in a survey of Ethiopian businesses (although limited in size, with only 363 EFD users participating). There, all users stated that EFD use had resulted in additional costs to their business, including EFD installation and maintenance costs, payments for stationery, and costs for the annual machine checkup.25 Some countries have addressed the problem of the high costs of EFD use by co-financing or even fully financing the devices. The Kenya Revenue Authority (KRA) fully refunded taxpayers for the costs incurred in buying EFDs. In Mexico, the government subsidizes the purchase and installation of electronic payment terminals in retail shops, which has significantly increased their use. This is an exceptional approach, however. At a minimum, active consultation with the business community prior to EFD introduction should be undertaken. In the Dominican Republic, a major contribution to smooth EFD introduction was the early outreach to business associations, with the government negotiating for their support during the process of adapting their members’ existing sales systems to the required EFD standard. In Sweden, the tax administration launched a dialogue with both the retail business segment and the cash register industry, and a Cash Register Council (Kassaregisterrådet, KRR) was established as a collaborative organization for cash register manufacturers and suppliers to ensure an ongoing dialogue with the tax administration. In Korea, a variety of tools were used to promote the introduction of mandatory POS system use, ranging from advertisements in both mass media (such as television, radio, newspapers, homepages, community papers, subway commercials, leaflets, and magazines) and interactive media (such as naming, logo, and marketing team contests and amateur marketers’ blogs). In addition, numerous one-time events were held in parallel to draw the public’s attention to the change, such as a two-day street campaigning event in which about 17,000 people participated, including NTS staff, tax accountants, celebrities, and various consumer and business association members.26 In Kenya, in response to business resistance to the POS systems, the tax administration launched a major information, education, and communication initiative. Survey evidence from Kenya shows that despite initial resistance to the introduction of POS systems, businesses that have started to use them recognize their advantages, particularly in terms of easier transaction processing and invoice issuance.27 POS System Impact Despite the widespread use of POS systems, reliable data on their impact on tax compliance, particularly on VAT performance, are scarce. An IMF survey of 19 tax administrations worldwide showed that most administrations that have measured the impact on compliance of POS system introduction reported an improvement in accurate reporting compliance.28                                                              25. Mohammed and Gela 2014. 26. H. J. Lee 20136.    27. Mativo 2015.    28. Casey and Castro 2015.     13 Figure 2.2 IMF Survey: Reported Benefits of Fiscal Devices Source: IMF 2015. More difficult to assess is the revenue impact of POS operation. The only recent thorough impact analysis was done by the Swedish tax administration (Skatteverket) in 2013. According to Skatteverket calculations, reported turnover from larger businesses (subject to monthly VAT filing) increased on average 5 percent in the first month after introducing EFDs; for smaller businesses with quarterly VAT filing, the impact was even higher, leading to estimates that an average 7 percent increase in turnover reporting for VAT could be achieved. However, the impact of EFD introduction varied substantially by business segment sector. Table 2.2 Sweden: Immediate Effect of Cash Registers Being Reported on the Company’s Reported Turnover for Companies with Monthly VAT Filing Source: Skatteverket 2013 These results were confirmed by audit findings showing that the average amendment of turnover declared in the audit process was 17 percent lower after EFD introduction. However, the Swedish analysis focuses   14 on business attitudes immediately after EFD introduction. Tax authorities are concerned that this short-term impact will diminish over time, but no reliable analysis is available to assess such longer-term impact. In the Dominican Republic, as well, visible impact of POS system use on VAT (in the Dominican Republic called ITBIS) performance could be observed. A comparison of VAT performance of businesses equipped with EFDs with that of businesses not yet using such devices showed that the increase in VAT collection from the taxpayer group using EFDs exceeded that of the group not using them. Additionally, VAT noncompliance decreased by 14.7 percentage points between 2004, when the EFDs were introduced, and 2008. Figure 2.3 Comparison of VAT (ITBIS) Increase: Taxpayers With and Without Fiscal Printers Source: Cardoza 2012.   Finally, data analysis in Ethiopia and Rwanda showed that the average amount of VAT paid by firms increased substantially after EFD introduction, and that, similar to the Swedish case above, the impact of EFD varied by business sector, greater impact in segments with higher levels of tax evasion.29 In Rwanda, the average increase in VAT payments after EFD introduction was 6.5 percent, with the highest impact in retail, construction, and computing and printing.30 Because POS system introduction is frequently part of a broader program to improve compliance and reduce tax evasion, it can be difficult to link increases in revenue performance and compliance levels clearly to POS system operation. The IMF analysis of VAT performance in eight countries and one state in Brazil before and after POS system introduction concludes that, except for Chile, the introduction of POS systems has not been associated with noticeable increases in VAT revenues as a percentage of GDP; in fact, it is difficult to determine any positive impact at all from the introduction of such systems. Certainly, even with some indication from the country cases cited above usefulness of POS systems as a tool to address cash business tax evasion, they are not stand-alone remedies to cash economy compliance problems, nor are they the new silver bullet for solving these problems. They have the potential to serve as a building block in a cash economy compliance management program; however, this is only the case if proper POS system use itself is ensured by special compliance management activities.                                                              29. Ali, Shifa, Shimeles, and Woldeyes 2015. 30. Eissa et al.  2014.      15 Monitoring System Use Ensuring the proper use of POS systems is the central problem of system management. In the first years after POS introduction particularly an intensive supervision mechanism will be required. Different control approaches should be combined to ensure proper system use; in addition to advisory visits highlighting the importance of correct EFD use, communicating the risk of penalties in case of abuse, and exploring any problems the business may experience in using the EFD, other essential management tools will include regular targeted audits focusing on EFD use and control purchases to identify violations of EFD usage requirements. The approach followed by the Swedish tax administration in the first three years after the introduction of mandatory EFD use provides an example of such a control and inspection mechanism. Table 2.3 Sweden: Supervision, Inspection Visits and Audits Within the Cash Trading Operation in the First Three Years of Compulsory Cash Register Use 2010 2011 2012 Total Supervisory visits 50,353 20,782 10,308 81,443 Inspection visits 3,100 7,198 11,900 22,198 Audits 319 257 306 882 Source: Skatteverket 2013. POS System Manipulation The use of electronic sales suppression (ESS) techniques has been identified as a major compliance management challenge, even for OECD countries.31 For instance, Revenue Québec, the agency responsible for the administration and collection of the province’s income and consumption taxes, has estimated that its tax losses from such techniques were Can $417 million for 2007–08. The Canadian Restaurant and Food Services Association estimated that in 2009 “phantom” cash sales could add up to nearly Can $2.4 billion (OECD 2013a, 6). The German Federal Court of Auditors raised concerns in its Annual Report 2003 on Federal Financial Management regarding losses from ESS and concluded that “with cash transactions running into tens of billions of euros, the risk of tax evasion should not be underestimated.” Detection of sales suppression is an even bigger problem in developing countries, particularly due to the generally limited capacity for e-auditing. The two critical elements in confronting ESS are (a) the need for an approach to detecting POS system manipulation, and (b) legal provisions to penalize such manipulation appropriately. Detecting POS Manipulation Detecting sales suppression is a major challenge even for OECD countries. Many revenue authorities have trained specialized e-auditors in the use of computer-assisted audit tools and techniques (CAATTs) to import large data sets and carry out a wide range of analysis with the use of specialized audit software, such as IDEA, ACL, or SESAM. Computer forensic investigations can be used for testing and analyzing systems and tools seized in criminal investigations. While the primary focus for seizure of data will be the POS system, businesses under investigation offer other sources of digital information. Computers with back office systems and external storage media could reveal the use of zappers and phantom-ware. Once the seizure and review of such sources is legally authorized, the challenge is to access the information so that it can be copied. If the digital analysis emphasizes detection of ESS software, the analysis will largely focus on program files and entries in the operating system. The analysis procedure can vary, requiring a range of expertise. One common approach used with good results is to obtain and                                                              31. For an extensive discussion of issues and approaches see OECD 2013a.         16 review application software for the secured materials. This can be done by running the program on another physical computer or on a virtual machine. Finally, a full criminal investigation can come into play with the use of traditional investigation techniques, or even undercover operations managed jointly by tax authorities and other law enforcement agencies, to target POS manufacturers. When evidence is sufficient to indicate guilt beyond a reasonable doubt, a referral is made to the prosecutor for tax and any other criminal charges. The goal of any criminal prosecution, in addition to punishing the offender, is to deter others from committing similar offenses and to enhance compliance by communicating the message that tax evasion is an offence that will be prosecuted and publicized. Introducing a Penalty Regime POS system manipulation is a serious offence that facilitates tax evasion. Therefore, the use of sales suppression techniques could be prosecuted by applying the general penalty regime for tax evasion cases. Given the severe consequences of system manipulation, many countries have introduced higher penalties for manipulation than for a simple irregular use of POS systems (see the example of Tanzania below). However, this does not capture the risk for tax compliance stemming from the existence of a malware market. The introduction of a broader penalty regime, penalizing not just the use of system manipulation techniques, but also their production, distribution, and acquisition, is advisable. With such an approach, it is not required to prove that malware has actually been used or to determine the scope of taxes evaded. While POS systems offer particular value added for monitoring B2C cash transactions, a more comprehensive monitoring of supply chains, particularly B2B transactions, can be achieved through electronic invoicing. Combining an e-invoicing system for transactions between businesses with a POS system for sales to consumers can provide the necessary data for extensive input-output monitoring; it also facilitates comparison of invoice data with financial and accounting data. A notable benefit provided is easier tracking of discrepancies between levels of recorded supplies and unrecorded sales. Extensive cross-checking programs involving paper invoices have been tried in the past but were difficult to implement. An e-based invoice monitoring system can extend the cross-checking capacity substantially, subject to sufficient IT capacity and a reliable taxpayer identification system. Pioneers of e-invoicing were Korea in Asia and Chile in Latin America, but the use of e-invoices is increasingly compulsory elsewhere, especially for VAT-registered businesses. Electronic invoicing may require major changes in businesses’ processes and invoicing systems. On the tax administration side, it requires investment in IT capacity to process and match data from e-invoices. Gradual introduction of comprehensive e-invoicing systems is therefore recommended. As an EY survey32 shows, many countries introduced e-invoicing as an optional rather than a compulsory system, with a primary focus on giving businesses opportunities to reduce tax compliance costs.                                                              32. EY 2014.   17 Figure 2.4 Country Approaches to Electronic Invoicing Source: EY 2014. Many countries have initiated mandatory e-invoicing for specific business segments or transactions. Often a first area targeted for mandatory e-invoicing is business to government (B2G) transactions, with suppliers obligated to send invoices electronically to public-sector clients (as in Denmark, Norway, Finland, Italy, Austria, and Singapore, for example), while some countries made e-invoicing compulsory for specific business sectors (for example, financial institutions and exporters in Ecuador, the telecom sector in Turkey, and large businesses in Chile and Uruguay). Although comprehensive mandatory e-invoicing remains an exception, it has been or will be applied in Korea (since 2011), Guatemala (since 2013), Indonesia (from 2016), and Chile (from 2017). In 2010, the European Union introduced a new rule (Directive 2010/45/EU) on e-invoicing, liberalizing e-invoicing requirements and laying down the principle of equal treatment for paper and electronic invoices in terms of integrity and authenticity of contents in a bid to encourage e- invoicing. This initiative, especially the allowance of PDF files as eligible invoices, has led to expectations that the share of e-invoicing will reach 24 percent in 2014, up from 8 percent in 2008, of some 17 billion invoices received by European businesses and governments (Koch-Billentis 2014, 32). Some countries have introduced measures to reduce the financial burden and provide incentives for the move to e-invoicing. An example is Korea, where in the initial phase of e-invoicing a tax credit of KRW 100 (US$0.09)33 was given for each tax invoice issued electronically, with a maximum of KRW 1 million (US$850) per year. As distinct from POS systems, e-invoicing is seen by businesses as a compliance simplification tool that can reduce the VAT filing compliance burden. Resistance on the business side to the introduction of e- invoicing tends therefore to focus on technical and capacity aspects. For tax administrations, specifically regarding nonregistered cash transactions, e-invoicing may facilitate the identification of businesses that purchase a lot of inputs but produce only a small number of sales invoices. In Korea, it also helped identify professional sellers of fictitious invoices. Box 2.1 Korea: Invoice Matching before and after E-Invoice Introduction34 Korea had initially started a comprehensive program to cross-check invoices following the introduction of VAT in 1977. Customers receiving VAT invoices with a value above KRW 300,000 (around US$450) were obliged to send these invoices to their district tax offices. Invoices were then digitalized and processed by the tax administration IT center and aggregated by the seller. The aggregate was subsequently compared to                                                              33. W stands for the won, the currency of the Republic of Korea. 34. Based on IMF 1996 and H. J. Lee 2013.   18 the volume of sales reported on the seller’s VAT return. The system was successful insofar as it contributed to increasing sellers’ reporting compliance. However, invoice matching was time-consuming, and it could take up to two and half years to identify discrepancies between sales and purchase declarations. A large staff was required to operate the system, and the results of the matching exercise often were unreliable, with many discrepancies between sales and purchase invoices resulting not from hiding transactions, but from data entry mistakes. E-invoicing allows real-time VAT invoice monitoring, and data entry mistakes have been eliminated, reducing the data matching period to three months. The system includes an early warning system, which immediately highlights suspicious cases, such as imbalances between sales and purchases, a low ratio of tax invoice issuance, or issuance of many high amount invoices over a short period without filing VAT returns. According to the Korean tax administration, the early warning system contributed to detection of KRW 5.04 trillion worth of fictitious transactions and the levy of an additional tax of KRW 374 billion in 2012. China has had in place a mixed electronic/paper invoice control system for a considerable time. While the Chinese35 government has not yet rolled out electronic invoicing for combined business and tax purposes, it has acknowledged the benefits of using information technologies for tax control. This led to the creation of the Golden Taxation Project (“Golden Tax System”) in 1994, which is being rolled out gradually. Where the system is in place, its use is mandatory for all invoices VAT-able under Chinese law. The Golden Tax system is an online invoice checking network based on paper invoices. It now links some 4,000 tax authorities at and above the county level. It has significantly decreased tax fraud and thus is viewed as a major success. Currently, the system is in its third rollout phase. Special Cash Economy Reporting Requirements Several countries have introduced special reporting or withholding requirements under income tax for payments to businesses in high cash economy sectors and areas in which compliance control is difficult. This is based on the conclusion that compliance increases substantially when a transaction is subject to either withholding of income tax or reporting to the tax authority. The basic assumption behind the approach is that in a B2B relationship the payer will have less incentive to hide the payment than will the provider of the goods or services, particularly when the noncompliance with reporting or withholding requirements is subject to substantial penalties. As the impact of information reporting on compliance, although not fully reaching compliance levels in case of withholding, is comparatively high,36 reporting is a suitable and effective alternative to tax withholding, and it reduces compliance costs for the provider of the good or service. Tax reporting requirements either can be industry based or can extend to any cash transaction above a certain threshold. Transaction reporting by large, formal businesses facilitates information cross-checking and supervision of medium and small businesses, which present higher risks of informality. A typical example is the Canadian construction-sector contract reporting system. The Canadian tax administration identified compliance problems in the construction and home renovation sectors and developed the Contract Payment Reporting System. It launched the system on a voluntary basis in 1996. When it deemed the                                                              35. China’s Law on Administration of Taxation of the People’s Republic of China (2001) regulates the tax compliance process in mainland China. The implantation ordinance of the law No. 362 Decree of State Council of the People’s Republic of China— Regulation of Implementation of Tax Collection and Administration of the People’s Republic of China (2002) also addresses tax compliance. 36. IRS analysis has shown that the net misreporting percentage is only 8% in case of substantial information reporting, while it is 56% in case of little or no reporting; see IRS 2012, “Tax gap for the Year 2006”.   19 voluntary participation rate too low, the Canada Revenue Agency (CRA) made participation in the system mandatory, as of 1 January 1999. Under the Contract Payment Reporting System, individuals, partnerships, and corporations whose primary activity is construction must report their payments to subcontractors to the agency on an annual basis. The agency then matches this information against the data it maintains to identify taxpayers who have not filed tax returns or have underreported their income. The information reported by the construction and home renovation sectors allows the agency to better target noncompliance. Box 2.2 Canada: Contract Payment Reporting System (CPRS) Any construction business making payments to subcontractors must report to the tax administration any payment exceeding Can $500 (US$370). The report must name the subcontractor, provide the business address and business registration number, and specify the amounts paid in the reporting period. The CPRS has proved highly successful. The number of income tax returns filed by construction businesses went up by 182,000 between 2001, when the system was introduced, and 2007; 4,200 construction businesses registered for VAT for the first time, and unreported income of US$203 million was detected. Interestingly, a public opinion survey conducted by the CRA in 2004 revealed considerable opposition to the CPRS on the grounds that, although it was effective in counteracting the underground economy, businesses incurred costs in administering it and some participants were unaware of how the information collected was used.37 This demonstrates the importance of active outreach to businesses and of effective communication to promote system acceptance and proactively address business concerns and resistance. The United States has also introduced a requirement to report payments made to independent contractors. Any business making payments to an independent contractor (primarily operated by individuals, but in certain cases also operated in the form of a legal entity) for business-related services in the amount of more than US$600 per year must report such payments to the IRS. An example of a general major cash payment reporting system is the United States’ Form 8300 system. Any person engaged in a trade or business who receives more than $10,000 in one transaction or several related transactions must file a cash payment report to the IRS. The cash payment report includes detailed information on the payer of the amount, including its tax identification number. The system was developed jointly by the IRS and the Financial Crimes Enforcement Network (FinCEN) and is used by the US government to track tax evaders and individuals profiting from criminal activities. Use of Information Collected by Financial Intelligence Units Many countries introduced general cash reporting systems less for tax compliance management purposes than for the primary goal of combatting money laundering and hence tend to benefit financial control authorities more than tax administrations. Nonetheless, information collected through these channels is also extremely useful for tracking major cash economy activities. It is therefore vital to establish a regular data exchange between tax and anti–money laundering authorities. The benefits of such access have recently been highlighted in an OECD analysis (which, although focusing primarily on access to information from suspicious transaction reports, also values information from cash transaction reports).38 This analysis has shown that key obstacles for such data access are either legislative barriers restricting access to reports or conceptual/operational barriers, such as the perception that counteracting tax evasion is secondary to fighting money laundering. The most far-reaching approach is to provide direct, unlimited access to cash transaction reports for compliance management purposes; however, this might not be feasible in all                                                              37. OECD 2009. 38. OECD 2015.     20 countries. Alternatives to consider include agreements on which reports should be used by tax as well as anti–money laundering authorities, joint review of those reports, or, at a minimum, unilateral review of the reports by the anti–money laundering body, which then passes on information it considers as relevant for taxation purposes to the tax administration. As noted, whereas the US Form 8300 system addresses both tax and money laundering concerns, many countries use cash reporting requirement primarily to fight money laundering. In Korea, financial institutions must report daily cash transactions by traders totaling W 20 million (US$17,000) or more. These currency transaction reports must be filed with the Korea Finance Intelligence Unit within 30 days of the transaction date. In Canada, certain institutions, such as banks, real estate agents, insurance companies, and money services businesses, must report cash payments of more than Can $10,000 (US$7,200) to FINTRAC, the Financial Transactions and Reports Analysis Center of Canada. Similarly, banks in Australia are required to collect information and record and report the identity of anyone performing a cash transaction of AUD10,000 (US$7,200) or more to the Australian Transaction Reports and Analysis Centre (AUSTRAC), and financial institutions in Malaysia must report any cash transaction above MYR 50,00039 (US$ 11,600) per day to the financial intelligence system. For tax compliance management purposes, therefore, it is extremely important that data be exchanged between tax administrations and anti–money laundering authorities. India provides an example of this cooperation in action. India’s Prevention of Money Laundering Act of 2002 (PMLA) and its Financial Intelligence Unit (FIU) operating rules form the core of the legal framework put in place to combat money laundering. The directors of FIU-IND and of Enforcement have been exclusive and concurrent powers under relevant sections of the Act to implement its provisions. These regulations also require reporting of cash transactions to the tax authorities, thus allowing them to investigate to determine if the cash represents unaccounted income subject to tax. The PMLA and FIU rules impose the following obligations:  Banking companies, financial institutions, and intermediaries must verify identity of clients.  All Reporting Entities are required to furnish to Director, FIU-IND the report on all purchases and sales by any person of immovable property valued at INR 5 million or more that is registered by the reporting entity, as the case may be.  All Reporting Entities are required to furnish to Director, FIU-IND the report of all cross-border wire transfers of the value of more than INR 500,000 or its equivalent in foreign currency where either the origin or destination of the fund is in India.  Every banking company, financial institution, and intermediary is required to furnish to FIU-IND information related to all cash transactions where forged or counterfeit currency notes or bank notes have been used as genuine or where any forgery of a valuable security or a document has taken place facilitating the transactions.  Every Reporting Entity (a banking company, financial institution, intermediary, or person carrying on a designated business or profession) is required to maintain records and furnish information40 to FIU-IND related to, e.g., all cash transactions of the value of more than INR 1 million or its equivalent in foreign currency; all series of cash transactions integrally connected to each other that have been valued below INR 1 million or its equivalent in foreign currency, where such series of transactions have taken place within a month; and all cross-border wire transfers of the value of                                                              39.RM stands for the Malaysian ringgit. 40.In India, provisions in the Income Tax Act also require third parties, such as banks, companies, and property registries to provide information on financial transactions regularly undertaken by taxpayers (Section 285BA).     21 more than INR 500,000 or its equivalent in foreign currency where either the origin or the destination of fund is in India. Considering the difficulties of ensuring the correct reporting of cash income, tax administrations aim to trace simultaneously the deposit and use of cash revenues. Cash income eventually will be deposited in a private bank account, spent, or transferred abroad. Matching data on the wealth and spending behavior of a cash economy business operator with data in his income tax return can reveal major discrepancies and hidden sources of income. Analyzing bank account data is the first obvious step in such an exercise. Substantial progress has been made in recent years to broaden tax administration access to banking information. For developed countries, the OECD could already assert more than 15 years ago that a large majority of OECD member countries can obtain information about a specific taxpayer’s accounts by requesting the information from the bank directly or indirectly through a judicial or administrative process. The tax administrations of some member countries have the authority, under certain circumstances, to enter the bank premises and obtain directly the necessary bank account information. The tax administrations of other member countries have direct access to bank information through centralized databases. Other tax administrations may have less direct access, requiring a formal process (for example, an administrative summons, requirement, or court order) to obtain such information. Many tax administrations also receive certain types of information from banks (for example, the amount of interest payments) on an automatic basis.41 In non-OECD countries, access to banking information may be far more restricted. The most frequent restrictions require tax authorities to obtain a court order requesting information from banks, the agreement of the central bank, or the restriction of information access to business accounts, while private accounts are protected by the bank secrecy law. Some countries also limit access to banking information to cases of tax fraud investigation and do not permit routine inquiries. Restrictions on bank information frequently are included in banking laws and not in tax administration legislation, indicating the conflict between the banking sector and the tax administration. A typical example is the banking law of Azerbaijan, which guarantees the secrecy of client information received by banks, account information, and details of transactions. Box 2.3 Ensuring Tax Administration Access to Banking Information France Pursuant to Article L85 of France’s Tax Procedures Code, tax officials may require taxpayers, including banks and financial institutions, to disclose all their accounting records, as well as the statements of private accounts of individuals or companies and copies of the face and reverse sides of checks. Pursuant to Article L83 of the Code, the administration can also require banks to disclose internal documents that go beyond accounting records, such as proxy forms and specimen signatures (persons with powers of attorney over an account) or contracts for opening an account, any guarantees that may have been constituted (bonds or cash) in the context of setting up a loan or overdraft privileges, or vault visit records. The bank is not bound by any obligation of discretion, and it may inform its customer of the tax authorities’ request for disclosure; the bank itself is not, however, advised of the reason for the request. Lastly, pursuant to Article L96A of the Tax Procedures Code and Article L152.3 of the Monetary and Financial Code, the administration may require banks to disclose information on capital transfers by French residents to a foreign destination or to nonresident accounts for which they are the depositories. This information includes the date and amount of the sums transferred, the identity of the initiator of the transfer and of the beneficiary, and references for the accounts concerned in France and abroad.                                                              41. OECD 2000.     22 Estonia In Estonia, banks are required to maintain general confidentiality of client information under subsection 88(1) of the Credit Institutions Act. The disclosure of such information to the tax authority is governed by subsection 88(5)4, which allows a credit institution to disclose information subject to banking secrecy to “a tax administrator pursuant to the provisions of the Taxation Act.” In 2012, Estonia amended the Credit Institutions Act to introduce a new section 88(6). This allows banks, pursuant to the provisions of the Taxation Act, to disclose confidential information in response to an inquiry from the tax authority if such inquiry sets out: (i) the name or business name of the client together with the personal identification code, date of birth, or registry code; or (ii) the account number of the client; or (iii) another unique identifier that enables identification of the client with respect to whom the inquiry has been lodged. Source: OECD 2011; OECD 2013b. In addition to the authority to request information from banks, it can be extremely useful to establish certain specific automatic reporting requirements. At the most basic level, banks can be required to report automatically to tax authorities the opening of any new bank account. Such a requirement exists, for example, in Kazakhstan, the Slovak Republic, and Georgia, although it is limited to accounts opened by legal entities or business accounts held by individual entrepreneurs, advocates, and notaries. An example of a much broader automatic reporting obligation is included in Article 623 of the Tax Statute of Colombia. The provision obliges financial institutions to annually remit to the tax administration the names and TIN of clients receiving amounts above COP25 million (US$24,300) as well as the amounts of all transactions they carried out during the fiscal year and the balance of their accounts. In India, Section 285BA of the Income Tax Act requires third parties, such as banks, companies, and property registers, to regularly provide information on financial transactions of taxpayers to the tax administration. Apart from banking information, spending by recipients of cash income can be rather easily monitored by accessing credit card information. However, credit card payments do not automatically facilitate compliance monitoring. The move to increased payment by credit card benefits tax authorities only if they get automatic and full access to payment data from banks and credit card companies allowing comparison of payments and recorded sales. This is typically done by requiring credit card companies to report all transactions to the tax department on a regular basis. Box 2.4 Australia’s Credit and Debit Card Data Matching Program The ATO in Australia has been operating a data matching program since 2008. Details of all credit and debit card payments received by merchants are collected annually from 11 financial institutions (banks and credit card companies). The ATO matches these data against its own internal data with the objective of detecting unreported income through discrepancy matching; identifying businesses that have failed to meet their registration, filing, or payment obligations; identifying deregistered businesses continuing to operate; and identifying “cash only” businesses. The total number of merchant account records obtained is around 900,000, with approximately 90,000 individuals linked to those accounts. An overall automatic matching rate of over 85 percent is generally achieved across the credit and debit card collection. This rate is improved further through targeted manual matching. Therefore, a promotion of credit card transactions must be combined with an awareness campaign informing transaction parties that compliance control will be enforced. In 2008, the US Congress enacted a law intended to improve business tax compliance. Starting in 2011, banks processing credit card transactions and proprietors of third-party payment mechanisms, such as PayPal, were required to report to the IRS total receipts of businesses accepting these forms of payment from customers using these forms of payment. These reports are made on a new report form, Form 1099-K, which provides to the IRS, for each business, information on the gross value of transactions paid by credit card, the value of transactions for each month, and data on any income tax withheld. Data analysis showed that the introduction of the information reporting system increased business reporting of receipts by up to 24 percent; however, firms   23 largely offset this change with increased reported expenses (an area not subject to information reporting), so that the overall effect on reported net taxable income was significantly smaller than would be expected otherwise without the increase in expenses.42 For EU countries, Madzharova,43 analyzing EU country-level data, concludes that the visibility of electronic payments does not appear to influence VAT’s collection efficacy in a significant manner; however, the analysis shows a more clear-cut picture with respect to cash, which has an unambiguously negative effect on VAT performance, at least in countries with well-established card payments. Tax authorities have a wide range of additional information collection possibilities that could offer useful information on a taxpayer’s spending behavior, such as vehicle registries, land registries, and other asset registries. Some countries also have introduced automatic reporting requirements by chambers of commerce allowing them to compare chamber membership with the official business registry. Such extensive information reporting requirements for public and private organizations and registries risk information overload on the tax administration side. Information received is of little value if it cannot be attributed clearly and easily to a specific taxpayer. This requires the use of a uniform identifier. In practice, many countries’ automatic analysis and matching of information received is hampered by the different identification codes and systems used by their various registries. Sometimes this can be addressed by negotiating agreements between the tax administrations and the other government registries to include the registered person’s or property owner’s TIN in the registration data set. A more efficient method provides each person a single national identity number to be used in all registries and when opening bank accounts or requesting credit cards. When third-party information can be clearly and easily attributed to a specific taxpayer, and the necessary IT infrastructure and capacity are available for large-scale data matching, third-party data matching can be an effective tool to detect noncompliance, as the example of India shows. Following an intensive analysis conducted some years ago, the Direct Taxes Board of India has been building its efforts to counter the underground economy. Among a broader range of actions taken, its focus has been strengthening the regime of third-party reporting and developing the administrative capacity required to capture and process large volumes of financial data to identify noncompliers based on risk and revenue potential. Annual mandatory reporting requirements for high-value transactions were introduced for certain agencies, such as banks and credit card companies, companies and institutions issuing bonds or shares, and trustees of mutual funds. Also, obligatory use of the tax identification number (in India, the PAN or permanent account number) when conducting certain transactions has been introduced. Box 2.5 India’s New Income Tax Reporting Norms as of April 2016 To curb the menace of black money, the Income Tax department of India has established new requirements for banks, property registrars, and other agencies to report high-value transactions beyond a certain threshold for cash receipts or withdrawals; purchase of shares, mutual funds, immovable property, and term deposits; and sale of foreign currency. The new reporting norms became effective April 1, 2016. Here are the highlights: 1) The new norms require property registrars to report to income tax authorities any sale or purchase of any immovable property of value exceeding INR 3 million. 2) Banks must report cash deposits of INR 1 million or more in a financial year. The same limit will apply for term deposits (excluding renewal deposits) with a bank. For current accounts, the limit is INR                                                              42. Slemrod et al. 2014. 43. Madzharova 2014.   24 5 million for a financial year. 3) Credit card issuers must report to tax authorities any credit card payment of INR 100,000 or more in cash or INR 1 million or more in any other mode in a financial year. The new rules also laid down the reporting norms for cash payment of INR 1 million or more in a financial year for purchase of bank drafts or pre-paid instruments issued by the Reserve Bank of India (the Central Bank). 4) A receipt from any person for sale of foreign currency, including any credit of such currency to a foreign exchange card or expense in such currency through a debit or credit card or through traveler’s checks or drafts, of an amount aggregating to INR 1 million or more during a financial year must also be reported to the tax authorities. 5) A company or institution issuing bonds, debentures, or shares must comply with the new norm if the aggregate receipt from a person in one year amounts to INR 1 million or more. For mutual fund houses, the limit is also INR 1 million and above. 6) Under the new norms, financial institutions must report the details of high-value transactions to the tax authorities online in a prescribed format. 7) The financial institutions must keep the records of these high-value transactions for six years to allow audits to check the genuineness of the transactions. 8) Financial institutions must verify the Permanent Account Number (PAN, the tax ID number) provided by the person. If a person does not provide a PAN number, the financial institution will take another declaration in a different format requiring other identification details. 9) These new reporting norms enable the tax department to verify tax returns filed by individuals by using details obtained from agencies quoting the PAN. The initiative has been very successful in practice as shown in Table 2.4. Table 2.4 Results Achieved by India’s New Income Tax Reporting Norms Fiscal year Number of potential Additional tax returns Additional tax collected non-filers detected filed 2013 1.29 million 2014 2.2 million Around 3.1 million US$750 million 2015 4.4 million Source: Ministry of Finance, India. Income received in cash from underground economy activities is not necessarily parked in domestic bank accounts. Experience in OECD countries has shown that even small business operators, such as butchers or painters, declaring only part of their income may have opened bank accounts in foreign countries to reduce the risk of detection in case of an audit. Therefore, access to information from foreign countries is an essential part of cash-economy compliance management.   25 Such access has traditionally been available through the exchange of information under tax treaty provisions, which largely follow Article 26 of the OECD Model Tax Convention. In the past, however, multiple obstacles prevented access to financial information about resident taxpayers from foreign countries. Many countries were particularly reluctant to exchange banking information, arguing that the national bank secrecy laws prohibited access to such information. This risk was addressed by the OECD in 2005 by adding a new paragraph 5 to Article 26 explicitly stating that requests for exchange of information cannot be denied on the basis that the requested information is held by a bank or other financial institution. Nevertheless, in practice, an international exchange of taxpayer financial information is still difficult in many cases, because the business operators’ home country tax administrations do not know the foreign bank account exists or the country in which it exists. The Commentary to Article 26 of the OECD Model makes clear that, while the home country tax administration is not required to determine precisely the financial institutions in which the taxpayer holds accounts or has deposited assets, general fishing operations asking, for example, for information about all bank accounts held by resident taxpayers in the other treaty country, are prohibited. Taking such difficulties into account, the possibilities for access to information from financial institutions in foreign countries have now been simplified and broadened substantially through the work of the Global Forum on Transparency and Exchange of Information for Tax Purposes, which has developed a legal framework and tools for the extensive automatic exchange of information. In the past, such broad-based automatic exchange of financial information was limited to small groups of countries signing special agreements; among EU member countries, for example, the European Directive on Taxation of Savings established a regular (at least annual) exchange of information on interest payments to residents in other EU countries, with the objective of counteracting cross-border tax evasion. A unilateral approach has been taken by the United States with the Foreign Account Tax Compliance Act (FATCA) as a basis for automatic exchange of bank account data. The Global Forum now has developed a global standard for automatic exchange of financial account information (the Common Reporting Standard, or CRS) and Global Forum members have committed to implementing the CRS. Currently, the Global Forum has 130 members, and 97 of these jurisdictions have already committed to implementing automatic exchange of financial information for tax purposes using CRS by either 2017 or 2018. The challenge, particularly for developing countries, will be to match the information received with internal information on individual taxpayers. This requires building capacity to process such information and attribute data received to a specific taxpayer. For compliance management purposes, the increased capacity to detect tax evasion proceeds parked in foreign countries should be communicated widely to the taxpayer community to raise awareness of the increased risks of tax evasion and thus to promote voluntary compliance. Many countries have introduced special withholding tax schemes targeting cash businesses operating in the underground economy. The benefits of applying a tax withholding approach are twofold: (i) ensuring a basic level of tax collection from informal businesses, and (ii) increasing pressure on informal businesses to register and apply for a TIN. In practice two types of withholding taxes are applied: withholding limited to payments to nonregistered businesses, or withholding applied more broadly to transactions in segments with high risk of noncompliance. Australia presents an example of the first type of withholding tax in section 12-190 of the First Schedule of its Taxation Administration Act 1953.   Box 2.6 Tax Withholding Obligations in Australia A business dealing with another business that does not quote its identification number must withhold taxes from any payment made at the rate of 48.5 percent. The high rate means the revenue is not at risk in relation to those transactions, since the withholding rate equals the maximum amount of income tax and social levy payable by an individual. The paying business must also complete a payment summary at the time of the withholding, giving full details of the payee and the transaction, and must send an   26 annual withholding report to the Tax Office detailing the transactions. This information enables the Tax Office to conduct income-matching checks on businesses that have not quoted an identification number.  The Australian approach has proved relatively successful. In its first year, taxes withheld amounted to US$16.1 million, in the second year US$32.2 million, and in the third year US$54.8 million. Ultimately, more than 40 percent of the businesses that had tax withheld because they did not present a registration number initiated business registration. An example of the second type of withholding tax is Ireland’s Professional Services Withholding Tax; it is a withholding and reporting regime covering prescribed professional services. The regime is supposed to have improved compliance, although its scope of application is rather narrow; data provided to the OECD show that 752 businesses were required to act as withholding agents, with tax withheld amounting to €527 million (US$576 million), representing income flows of over €3 billion.44 Again, as in the case of reporting requirements, such schemes are controversial and unpopular in the business community. The Dublin Chamber of Commerce, for example, continues to propose abolishing the withholding scheme. Box 2.7 The Relevant Contract Tax in Ireland All payments made by a principal contractor in the construction industry to a subcontractor are subject to tax withholding (the Relevant Contract Tax or RCT). Principals must notify the tax administration of all payments made on relevant contracts through an online information system. The subcontractor is also required to register for RCT, and subcontractors not already in the RCT database will be registered automatically after the first contract notification by a principal contractor. The principal must enter each payment to a subcontractor in the online information system before the payment is made and must deduct withholding tax in accordance with the deduction authorization issued by the tax administration. Of the three RCT rates (0 percent, 20 percent, and 35 percent), the applicable rate depends on the compliance records of the subcontractor, with the 0 rate applying to subcontractors that have been fully tax compliant for preceding three years, the 20 percent rate applying to subcontractors with a record of substantial tax compliance, and the 35 percent rate applying to all other subcontractors. In addition, if the tax administration forms the opinion that deductions from relevant payments at the standard 20 percent tax rate for the year of assessment will be insufficient to fully satisfy the income tax liability of the subcontractor for that year, the 35 percent rate may be applied. This provision can be used, for example, if the risk exists that the enterprise will go out of business before its tax debt has been cleared.   Several developing countries, particularly in the African region but also in Asia, have attempted to operate extensive withholding tax regimes, covering the entire cash economy segment. A typical example of such an approach is used in Pakistan. Box 2.8 Pakistan: Withholding Tax on Payments for Goods and Services and Payments for the Execution of Contracts Article 153 of the Income Tax Ordinance of Pakistan 2001 (ITO 2001) establishes extensive withholding requirements for payments for goods and services. The withholding rate schedule is complicated and consists of standard and reduced rates. Following are the core standard rates: • 3.50 percent for the sale of goods (other than rice, cottonseed, or edible oils) by a company; • 4.00 percent for the sale of goods (other than rice, cottonseed, or edible oils) by a person other than a company;                                                              44. OECD 2009.   27 • 2.00 percent for transport services; • 6.00 percent for other services rendered by a company; • 7.00 percent for other services rendered by a person other than a company; • 6.00 percent for the execution of a contract by a company; • 6.50 percent for the execution of a contract by a person other than a company. Only petty transactions (total price of goods sold by the trader to the withholding agent in a year is below PKR 25,000 (US$235) or total price of services provided in a year is below PKR 10,000 (US$95)) are exempted from the withholding obligation. Such regimes require many withholding agents, who are virtually impossible to identify and to monitor. In Pakistan, almost any person can become a withholding agent. Among those the law lists as being required to withhold income tax are any company and association of persons, nonprofit organization, persons registered under the sales tax act, and individuals with an annual turnover above PRs 50 million (US$470,000). Although such a regime creates additional incentives, particularly for retail businesses, to buy input from other cash economy businesses instead of from potential withholding agents,45 it imposes substantial compliance costs on buyers of goods and services, and the lack of proper monitoring capacity limits the impact of the withholding tax. This strongly suggests that a narrower withholding approach targeting specific high-risk cash sectors may be more effective than a broad and general withholding regime.   Section 3. Compliance Management through Incentives: Opportunities to Motivate Businesses and Consumers The risk of unrecorded cash transactions is pervasive, and the probability that sales will be made without issuing receipts is as high where POS systems are mandatory as where traditional paper receipts are used. Both cases require consent between the business operator and the customer, and the latter frequently is not interested in collecting invoices. Survey analysis in Kenya, for example, has shown that even after POS introduction, 85 percent of customers still did not ask for an EFD printed receipt,46 while a small mystery shopping case study in Rwanda showed that only 21 percent of the shops visited automatically handed out EFD receipts. Even with customers asking for receipts, the likelihood of EFD use increased by only 42 percent, while at the same time the price of the good increased by almost 10 percent on average.47 Analysis in the United States has shown that cash discounts made available to customers paying in cash and not requiring a receipt normally equals the per unit sales tax, so customers benefit from receiving the good without the indirect tax charge, while sellers benefit from reduced direct tax liability by not recording the sale in their income tax returns.48 Penalizing customers for not demanding an invoice for a purchase is an unpopular and questionable approach, and penalizing customers for using cash to get a price discount is not really feasible in practice. Therefore, the challenge for policy makers and tax administrators is to find ways to increase consumer interest in getting a tax invoice for transactions, and, one step further, to turn consumers into a control group checking suppliers’ tax compliance (that is, consumers serve as voluntary tax auditors). This is not possible                                                              45 . This issue is highlighted by Memon (2013, 40).   46. See Weru, Kamaara, and Weru 2013. 47. Eissa et al. 2014. 48. See Gordon 1990, 244.   28 on a mandatory basis;49 therefore, tax authorities in several countries have started exploring various incentive and awareness-building programs to influence customer behavior. The basis of any such approach is increasing awareness in the general society that underground economy activities have negative consequences both for the public and for the entire economy and that asking for receipts is an easy way to contribute to reducing the opportunities for underground economy activities. Such public awareness initiatives have operated in OECD countries for several years, with a pioneer role played by the Australian Cash Economy Task Force, established in 1996. More recent campaigns focus more directly on the role of consumer invoices for improving tax compliance. Box 3.1 Poland’s “Take a Receipt” Campaign The main objective of Poland’s “Take a Receipt” campaign, launched in 2011 by tax offices and chambers and fiscal control offices jointly, is to reduce the grey economy by means of control activities aimed at companies that avoid paying taxes and fail to obey the law. The campaign encourages customers to pay special attention to whether sales personnel conduct their businesses legally by providing receipts to customers after a transaction. One effect of this campaign has been to promote and support entrepreneurs who diligently fulfill their tax obligations. In Croatia, the Cash Transactions Fiscalization Act requires businesses to display on all electronic cash registers or in another visible place in the interior of the business premises a notice of the business’s obligation to issue receipts and of customers’ responsibility to take and keep the issued receipts. The business operator may be fined between HRK 5,000 and HRK 500,00050 (US$700 to US$7,000) for failing to display such a notice. A rather innovative approach is currently being tried in Romania. As of March 2015, economic operators obliged to use electronic cash registers must display a poster for their customers near the cash register. In accordance with Order No. 159/2015 of the Ministry of Public Finance, this poster must advise customers of their obligation to request a receipt if such receipt is not automatically issued by the vendor. Moreover, the poster indicates that, if the operator refuses to issue the receipt, customers are entitled to obtain the purchased good or service for free. The poster must also indicate that the receipt is the only document to be issued to the customer when purchasing a product or a service. Furthermore, restaurants, nightclubs, or other similar establishments are obliged to make this notice visible to customers by displaying it not only near cash registers but also on menus. Cash transactions between businesses and individual consumers are a typical risk area for underreporting. Individual consumers of business goods and services who pay cash provide businesses the opportunity to not report or to underreport those sales. Consumers generally do not have any incentive to receive invoices or receipts from businesses, however, since they ordinarily do not claim input tax credit for VAT or report purchases as an expense on their income tax filings. If business sellers offer lower prices than their competitors’ to customers who pay in cash, those customers are very likely to accept. Where this occurs, authorities find it very difficult to detect and curb the transactions, because no cross-checking takes place and the interests of sellers and buyers align. Proper incentive programs must therefore be designed to discourage consumers from making cash transactions. One possible approach to changing the incentive structure is to provide tax incentives for using cards instead of cash for payments in business-to-consumer transactions. The Republic of Korea has pioneered this                                                              49. See also the considerations expressed by Soled 1997. 50. HRK stands for the Croatian kuna.   29 approach. Private consumers can deduct from their salary income tax declaration a certain percentage of purchases paid by credit card or check card. When the system was introduced in 1999, 10 percent of credit or check card use up to a threshold of three million won (approximately US$2,900) could be deducted. Consequently, between 1998 and 2003, credit card use soared at an annual growth rate of 57.6 percent in terms of the number of transactions and 47.7 percent in terms of total value.51 In 2004, the incentives were reduced, reflecting the now widespread use of credit cards. As of 2013, 15 percent of credit card use above 25 percent of total wage income is eligible for income deduction. Other countries also have put in place measures to incentivize consumers to use electronic payment methods. For example, in Colombia, Article 850.1 of the Colombian Tax Code provides individuals a 2 percent VAT refund on purchases made by debit or credit card, as well as electronic banking for products and services paid at the general 16 percent rate or 5 percent rate. For purchases through mobile banking, mobile banking service providers must share specified information with the tax authorities: the base for calculating the two VAT points to be refunded, the total amount of VAT generated and paid in the transaction (at the general 16 percent rate or 5 percent rate), and information about the parties involved in the transaction. In 2001, Argentina introduced a point-of-sale VAT discount for debit card use (at the time of the corralito, when a separate official restriction on cash withdrawals from bank accounts was in place), giving cardholders a 5 percent reduction in the basic VAT rate of 21 percent on all purchases under Arg$1,000 (approximately US$220). In 2003, credit cards were added to the policy, giving cardholders a 3 percent reduction on their purchases. The government also introduced a further 2 percent reduction on purchases of gasoline via debit or credit card payment methods. The credit card rebates were abolished in 2009, but the debit card rebates remain in force. Several countries, particularly in Europe, have adopted tax incentives applicable to personal and household services (PHS), such as cleaning, cooking, laundry, and care services for children, the elderly, or the disabled, which are highly likely to fall within the underground economy. The scope of eligible services, the amount of tax deduction, and specific conditions for tax incentives differ among countries; however, the provisions all aim to turn undeclared household services into declared services by subsidizing them to reduce the expense gap between the formal and informal economies. For EU member countries, a European Commission report includes a detailed overview of tax incentives.                                                              51. See Bank of Korea (2013). Credit Card Usage in Korea Number of transactions  Annual growth rate  Total value (billion  Annual growth rate  Year  (thousands)  since 1998    KWN)  since 1998  1998        179,597.00             20,170.00      2003     1,747,510.00   57.6%          141,712.00   47.7%  2012     6,793,115.00   32.2%          359,670.00   24.8%  Source: Bank of Korea 2013.   30 Table 3.1 Scope of Personal and Household Services as Related to Public Policy Instruments Source: European Commission 2013. A relatively common approach to soliciting public participation in detecting tax evasion and underground economy activities is the operation of a whistleblower mechanism. One model of this approach gives customers an active role in verifying invoices. These systems allow customers to verify electronically if businesses have correctly reported invoices to the tax authority; if consumers find an incorrect or unreported invoice, they can report the business. Several countries have successfully deployed this approach; four key examples are Korea, Croatia, Portugal, and Brazil. The Cash Receipt System in Korea Korea’s cash receipt system was introduced in 2005; it is mandatory for retailers and other B2C businesses, except certain micro businesses. When a consumer purchases goods at a store and provides his or her mobile phone number, Cash Receipt card (an ID issued by the tax administration free of charge to consumers and businesses), or ID, the store issues a cash receipt via the issuing certified EFD. The issuance of the receipt is authorized by a licensed Cash Receipt System Operator (with currently 41 operators licensed). The Cash Receipt System Operator must send the transaction records to the tax administration no later than four in the morning of the following day. The tax administration stores the data in its database and utilizes them to keep track of businesses' revenues. Consumers and registered stores can look up their transaction records on the cash receipt website. In addition, a mobile cash receipts application developed in 2012 allows registered consumers and businesses to view cash receipt transaction records as well as report unissued or denied transactions. In addition, a cash receipt call center with 60 staff members allows consumers and businesses to inquire about transactions.   31 Figure 3.1 Korea’s Cash Receipt Reporting System Source: Korea National Tax Service 2013. Consumers have five years to report a violation of the obligation to issue a correct cash receipt for a transaction, and they can report a business’s incorrect cancellation of a transaction in the system by either contacting the call center or using the cash receipt website. Customers making such reports receive a reward of up to 20 percent of the amount of the transaction. In addition, to increase the incentive for requesting receipts, wage and salary earners can get an income deduction in the amount of a certain proportion of total cash purchases made and properly documented. The Tax Fiscalization Program in Croatia52 Croatia’s Tax Fiscalization Program started in 2012. Research was conducted to determine public views regarding cash business activities and the level of support for an anti-evasion program. An extensive national campaign called “No account does not count” was conducted in 2013 through TV spots, posters and advertisements, and seminars and public fora, combined with an award lottery. The objective of the campaign was to raise public awareness of the negative consequences of the underground economy, the importance of tax receipts for controlling tax evasion, and the contribution consumers can make to reducing evasion by asking for receipts. Research conducted by the Croatian Chamber of Commerce showed that more than 93 percent of citizens supported the initiative, and 32 percent of citizens indicated a change in their behavior by requesting an invoice more frequently. Customers are invited to report nonissuance or incorrect issuance of invoices to the tax administration, and around 10,000 notices of irregularities were reported in 2013 and 2014. Tax administration data comparing income tax returns from 2010 and 2013 for a sample group of taxpayers show a remarkable increase of declared turnover and income, particularly in the catering segment, where the average declared daily turnover had increased by 82 percent and the average declared daily income had increased by 110 percent in 2013 as compared to 2010. However, increases were rather modest in the trade segment, with a 12.67 percent increase in declared daily turnover and a 12.69                                                              52. For details see Kudeljan 2015.   32 percent increase in declared daily income. Results were also far more impressive in the small and medium taxpayer segment than in the large taxpayer segment.53 Customer Monitoring of Electronic Invoicing in Portugal Portugal introduced an obligation for companies with an annual turnover above €100,000 (US$ 109,000) to introduce electronic invoicing systems starting in 2013. More than 700,000 companies have implemented the system. Only in less than 10 percent of cases were invoice data transferred automatically to the tax administration, however; instead, companies generally submit electronically a Standard Audit File for Tax (SAF-T file) or directly insert invoice data into a special tax administration website. For small businesses, a tool for inputting invoices with authenticated access was made available on the tax administration website, and the self-employed can access an electronic invoice issuance system on the website. As of January 2014, more than 4.6 billion invoices had been processed, with a total VAT amount of almost €53 billion. A special password-protected tax administration website provides registered businesses and consumers with data on electronic invoices recorded. Consumers then have the possibility of adding to the site any invoices not listed. In four specific cash-intense sectors (catering, hotels, hairdressing, and car and motorcycle repair services) consumers who requested the insertion of their TIN into the invoice issued can request a tax benefit of 15 percent of the VAT charged. More than 2.7 million customers made use of this option in 2013. Figure 3.2 Electronic Receipts in Portugal: Customer Requests for Use of Their TINs on Invoices in Cash-Intense Sectors Source: Azevedo Pereira. 2014 Due to this system, VAT performance in the four sectors in which a tax benefit has been granted has substantially exceeded average VAT performance.                                                              53. Croatia, Ministry of Finance 2014; see also Katolik 2014, 545; and Marković and Pavić 2014, 575.   33 Figure 3.3 Electronic Receipts in Portugal: Variations from Average VAT Performance in Cash- Intensive Sectors  Source: Azevedo Pereira. 2014 The Nota Fiscal Paulista (NFP) Program in the State of São Paulo, Brazil The Nota Fiscal Paulista (NFP) Program introduced in the State of Sao Paulo, Brazil, in 2007, has three key elements: (i) a tax invoice lottery, (ii) an online account system to verify receipts, and (iii) tax refunds for consumers asking for receipts. When making a purchase, consumers can ask businesses to put their social security number (SSN) on the invoice. Businesses are obliged to send the tax administration all invoices electronically monthly, and the tax administration creates an account for each SSN and lists all invoices associated with it. Customers can create an online account on the tax administration website and cross- check the receipts listed with their SSN. Customers can also opt for getting an email alert every time a receipt is registered with their SSN. Customers can file complaints on the site if they did not get a receipt, the business refused to print the SSN on the invoice, an invoice has been issued but is not listed on the site, the invoice sent to the tax administration does not correspond to the invoice issued to the customer, or for other reasons. To promote the demand for receipts, customers get a VAT rebate for invoices sent to the tax administration that include the customer’s SSN. The amount of the rebate is calculated in a nontransparent way, however, using a complicated formula. In addition, customers can participate in a monthly tax lottery. From October 2007 to December 2011, a total of 13 million people enrolled online for access to the website, over 40 million customers asked for SSN receipts more than once, and 740,000 businesses submitted more than 3.5 billion receipts with SSNs to the tax authority. A total of 1,151,518 complaints were filed by 135,102 customers involving 134,054 different business entities. Naritomi’s analysis shows that the NFP led to an increase in reported turnover in the retail-business segment of, on average, 23 percent over the four years analyzed. This translates into an increase in tax revenue collection of around US$400 million net of consumer rewards over the four years. In addition to direct incentives to consumers for reporting violations of invoice issuance requirements, tax lotteries have become increasingly popular as a general incentive scheme for motivating customers to request a tax invoice and for increasing the risk to business operators of failing to report sales. Thus, a successful lottery scheme achieves two objectives: (i) an increased percentage of transactions supported by   34 correct issuance of invoices to customers, and (ii) provision to the tax administration of an additional tool for verifying declared business turnover by comparing the declared turnover with data from invoices submitted by lottery participants. Similar to the tax incentive schemes discussed above, tax lotteries incur costs; the tax administration must offer prizes sufficiently attractive to stimulate consumer participation. Lottery schemes are not a new phenomenon. The first tax lottery system was introduced in Taiwan, China, in the 1950s. In Europe, Malta in 1997 became the first country to introduce a lottery scheme, and many countries have followed, most recently Poland, with a shopping receipts lottery launched in 2015. Lottery prizes can be rather modest; Georgia, for example, offers in the range of US$5 to US$45. Some lotteries, not unusually, include cars as prizes, such as Portugal’s offer of an Audi A4. The top prize in the Uniform Invoice Lottery in Taiwan, China, is NT$10 million (US$300,000). A lottery’s success does not necessarily depend on the size of the prize. Most tax lotteries have draws every month (for example, Malta and the Slovak Republic) or every two months (Taiwan, China) or even every week (Portugal). The general experience is that tax lotteries are popular among taxpayers; in the small country of Malta, with a total population of only 400,000, a surprising 35.7 million receipts were submitted to the lottery in 2013. Experience with lottery schemes is rather mixed, however.54 While the costs of a lottery generally are well documented, the impact on tax revenue collection and compliance is rarely measured in practice.55 This is partly because, as in the introduction of POS systems, clear links cannot be established between the lottery and changes in taxpayer compliance attitudes. Attempts should nevertheless be made to assess the usefulness of a lottery scheme. Some conclusions can be drawn from analyzing the invoices submitted. Research in the Slovak Republic, for example, found that by far the largest share of receipts submitted for lottery participation had been issued by retail businesses. While this indeed is a major cash-business segment in the country, many invoices were issued by big retailers, such as supermarket chains, which generally have a good compliance history anyway. Only a small number of receipts had been issued by small independent retailers, and only 2 percent of all submitted receipts were issued by businesses in the service sector. Not surprisingly, therefore, the total fiscal impact of the lottery is estimated at a modest level of €8 million (US$8.5 million) additional revenues annually. A much higher impact has been estimated for the tax lottery in China, which started with a pilot phase in selected cities in 1998. The China Taxation Bureau in 2002, analyzing the period from January 1 to June 30, 2002, estimated the lottery’s cost-to- revenue ratio at 1:30 (cost of CNY 30 million versus increased tax revenues of CNY 900 million).56 Wan, analyzing 39 districts in Beijing and Tianjin over a period of five years, concludes that the effective growth rate of sales tax was 23.5 percent higher in areas practicing the lottery scheme than in areas not practicing it.57 Lottery schemes present several risks. A major one was mentioned above in discussing the Slovak lottery: a large majority of invoices submitted are issued by generally compliant larger businesses, particularly major supermarkets, meaning tax administration invoice-matching resources and lottery prizes are used without achieving any great benefit. On the other hand, the Slovak Republic example also shows that even major retailers are not necessarily fully compliant, making invoice checking potentially beneficial. The reverse side of the same issue, also reflected in the Slovak example, is that lottery schemes have limited impact on unrecorded cash transactions in the service sector. If a private customer can get a service, for example, have his house painted, for a cheaper price by paying cash without an invoice, the chances are rather limited that the customer will ask for an invoice on the remote chance of winning a lottery prize. In addition, experience in Taiwan, China, has shown that, in a scheme where each individual invoice has the same chance of winning regardless of the invoice value, customers have an incentive to pay separately for                                                              54. A detailed report on the experience of EU countries with invoice lotteries can be found in Fooken, Hemmelgarn, and Herrmann 2014. 55. For a theoretical discussion see Giebe and Schweinzer 2014. 56. Wan 2006. 57. Wan 2014.   35 every single item to multiply their chances by getting more receipts; customers understand this concept well.58 Finally, not all tax administrations were able to quickly and reliably cross-check invoice information with businesses’ tax return data, which reduces the impact of the lottery on revenue collection. Receiving thousands of invoices without having proper invoice-matching capacity is a pointless exercise, and selecting the small number of invoices requiring cross-checking is difficult. These issues have led some countries to abolish lottery schemes after a rather short period of operation. For example, Puerto Rico introduced a sales tax lottery (the IVU Loto) in 2011 and abolished it again in 2015; Georgia’s tax lottery operated for less than one year in 2012; and Korea decided to abolish its tax lottery after the introduction of the Cash Receipt System discussed above. An invoice lottery is definitely unsuitable as an isolated measure to counteract tax evasion in the cash economy. Offering occasional substantial monetary rewards for tax invoices does not effectively address cash-based tax evasion in some economic sectors, such as certain service areas, and it does not reduce the risk of collusion between buyers and sellers of a good or service. A lottery may be useful, however, to supplement a broader public awareness campaign to convey the importance of reducing underground economy activities and the contributions consumers can make to achieve this objective. Such campaigns should be built on an understanding of the negative effects of an underground economy and on reducing public tolerance for all tax evasion practices. Only with such a basic understanding, combined with incentives such as lottery schemes, can additional citizen control mechanisms become effective. Such comprehensive efforts require extensive outreach and public relations campaigns. Consideration should be given to adapting invoice lotteries to the environment and constraints of developing countries. For example, sending invoices to the lottery administration may not always be feasible in countries with unreliable or slow postal systems. Greece addresses this issue through its APODIXI system, which allows consumers to participate in the lottery by SMS: the buyer sends a text message to the lottery, including the necessary data from the receipt, especially the seller’s nine-digit fiscal code, the date and time of the transaction, and the invoice amount.59 Issuing a tax invoice is a business’s legal obligation, the violation of which is an offence subject to fines. This raises the question of whether additional incentives are appropriate and can be justified to increase the level of voluntary compliance. The taxpayer would be receiving financial benefits for behavior already required by law. The standard compliance management approach penalizes nonobservance of invoice issuance obligations rather than rewarding compliant behavior. In practice, however, cash economy environments make it impossible or extremely difficult to improve a low level of voluntary issuance of tax invoices either through enforcement or by increasing customer demand for tax invoices, raising the question of whether recourse to targeted incentive measures for businesses could be helpful. Several research studies have been conducted estimating the impact of reward systems on voluntary compliance. Not surprisingly, a survey of small business operators in South Africa found that rewards were more strongly favored than penalties as a strategy to encourage tax compliance, and the majority of respondents were not averse to receiving rewards for performing an action required by law.60 A compliance experiment with taxpayers in Costa Rica found the highest level of voluntary compliance in the group that was promised a monetary award for being fully compliant.61 Few countries have embarked on such a system in practice, however. Tanzania has tried a lottery for traders, giving prizes to those businesses that operate EFDs and declare a higher turnover as a result. The most prominent example of a business reward system is in Korea. Business operators issuing proper tax invoices can claim a VAT credit of 1.3 percent of the transaction value. In                                                              58. Giebe and Schweinzer 2014. 59. Fooken, Hemmelgarn, and Herrmann 2014. 60. Bornman and Stack 2015. 61. Torgler 2003.     36 addition, businesses can claim a KRW 20 (US$0.01) tax credit when they issue a tax invoice for a transaction with a value of less than KRW 5,000 (US$4.20). No analysis is available of the impact of such reward systems on businesses’ willingness to issue invoices. Compared to the potential reduction in the tax burden from not reporting transactions to the tax administration, the reward system seems hardly attractive, however, and doubts may therefore be justified over the usefulness of the costs and administrative burden of operating the system. Conclusion  Cash transactions create special challenges for tax compliance management. Cash payment for goods or services facilitates non-reporting of transactions and the business’s operation in the underground economy. A high level of tolerance in the society for informal sector operations, unfair competition for compliant businesses from informal sector entrepreneurs, and obstacles for businesses to enter the formal economy, particularly due to complex tax systems and high tax compliance costs, not only affect tax revenue collection but have a far broader impact on tax fairness and the quality of governance more generally. Promoting a change in payment methods and reducing the use of cash promises to be one of the most effective instruments for combatting the underground economy. This is a long process, however, and the willingness of economic actors to use noncash payment methods depends on various circumstances, including the availability of technology; the safety and reliability of the banking system; and costs. These circumstances generally are outside the sphere of influence of ministries of finance or tax administration. While new payment options, such as the development of mobile payment channels, contribute to a steady growth in the percentage of cashless transactions, in practice the level of noncash transactions per capita remains low, particularly in the retail sector, and even more so in developing markets, where they are only a fraction of those in developed countries. Although tax policy can only be one component of a broader strategy to increase the level of cashless transactions, the tax system can either provide a stimulus for moving to noncash payment methods or it can discourage the use of cash. The most direct method for enforcing payments through the banking system, used particularly in some OECD countries, is to simply set an upper limit for the permitted amount of payments in cash. But while this seems an easy and quick reform measure, the actual impact of such limits is heavily debated, and the forced move to a credit card or bank transfer payment method is not very popular. A more effective approach might be to introduce a less favorable tax treatment for cash transactions, possibly combined with targeted tax incentives for noncash payments. Approaches of this kind include disallowing deduction of business expenses from taxable income if the expenses were paid in cash and, on the VAT side, denying a deduction of input VAT for inputs paid in cash. Incentive schemes are operated in many European countries for the household-services sector, which tends to be a core of the underground economy; incentives include tax deductions or tax credits for noncash payments for such services. But incentives can also be designed more broadly to encourage the use of credit cards or bank transfers for paying for private goods and services. All such incentive schemes require careful monitoring of their costs and benefits. Given the impact of good governance and policy making, including tax policy design, on the level of the underground economy, it must be clearly understood that improving tax compliance in the cash economy is not just a tax administration problem. Tax policy should provide the necessary framework for improved voluntary compliance and reduced opportunities for tax evasion. This includes granting tax administrations the necessary access to third-party information; broader access to financial data, particularly bank account information and data on credit and debit card transaction volumes, can provide important information on the deposit and disbursement of undeclared income. Resistance from the banking sector is high, however, and access powers remains rather restricted in many countries. Scope also exists for further developing   37 income tax withholding or VAT reverse charge mechanisms for certain sectors with high underground economy participation; an example in many countries is the construction sector. Tax administrators risk overreliance on electronic tools to combat noncompliance in the cash economy, particularly given the increased sophistication of equipment such as electronic cash registers (ECR) in recent years. Technology does not change compliance attitudes, and it does not create a long-term commitment among cash economy actors to become more compliant. Short-term increases in the number of business transactions recorded are not necessarily a sign of sustained compliance improvement if ECR enforcement is an isolated approach and not part of a broader compliance strategy. The substantial ECR manipulation experienced in many countries documents the vulnerability of isolated technology-based approaches. An even more serious challenge is the widespread lack of interest among customers regarding tax receipt issuance, which undermines cross-checking of business transactions and facilitates nonuse of cash registers. The organization of tax invoice lotteries is an increasingly popular approach for the demand for tax invoices; experience with such lottery schemes is extremely mixed, however, and the resources required for operating a lottery and awarding prizes are not trivial. A lottery scheme may complement general measures to support the development of a tax invoice culture, but it should not be considered a substitute for them. Customers can be encouraged to report the non-issuance of invoices to the tax authority, and new technologies offer the opportunity for customers to check if invoices they received were recorded properly by the issuers. Even without offering rewards, such control schemes have attracted substantial interest from customers in some countries, for example, the Russian Federation’s app-based invoice validation option. While attempts to encourage customers to complain about not receiving invoices makes sense, underground economy activities are characterized in many cases by collusion between business and customer. In addition, emphasis on whistleblower practices creates an atmosphere of permanent control and denigration. Targeted and extensive outreach to taxpayers about the negative consequences of informal-sector activities and the risks of participation in the informal economy therefore should remain the cornerstone of a strategy to reduce cash economy noncompliance. The impact of such outreach programs can be increased by enlisting the cooperation of the private sector, particularly business associations, chambers of commerce, and the media, in the design and implementation of the initiatives. By combining extensive access to third- party data, increased data-matching capacity, incentives for improved voluntary compliance, and strengthened compliance enforcement, enhanced tax system effectiveness and increased formalization of business activities should be achievable, even in cash-intense sectors.   38   39 APPENDIX   Regression Outputs: Statistical Tables          Table A.1 Tax‐GDP on Shadow Economy, 143 Countries, Y2007      Least squares regression line      Tax – GDP Ratio (predicted)       =     27.957 + (‐0.278 * Shadow – GDP Ratio)    Table A.1.1 Tax/GDP, Shadow Economy Ratio, GDP per capita, 143 Countries, Y2007  Results suggest shadow economy ratio has a negative correlation with tax ratio significant at 5% level, when controlled for GDP per capita. As the shadow economy ratio grows in an economy, this leads to an estimated 12% decrease in the tax ratio’s numerical value.       40   Table A.2  Correlation Coefficients between Shadow‐to‐GDP and Regulatory Quality and Control of  Corruption                                 41   Table A.3  Regression of Shadow‐to‐GDP with Regulatory Quality, Control of Corruption                   Regression Line of Shadow Economy with Indicators                                                                               Regression Line of Shadow Economy with Indicators       Shadow – GDP Ratio (predicted)       =     32.0 + (‐3.0 * RQ) + (‐5.94 * CC)                                                                                 42 Table A.4  Regression of Tax‐to‐GDP with Regulatory Quality, Control of Corruption           43 Table A.5 Correlation coefficients and regression between Shadow‐to‐GDP and Technological  Readiness                                          44 Table A.6 Regression of Tax‐to‐GDP with Shadow‐to‐GDP and Technological readiness    Least squares regression line       Tax – GDP Ratio (predicted)      =     21.34 + (‐0.17 * Shadow – GDP Ratio) +                                                                                            (1.284 * Technological readiness)          List of Abbreviations    ATM    Automated teller machine  ATO     Australian Taxation Office  B2B    Business to business  B2C    Business to consumer  CRA    Canada Revenue Agency  CRS    Common reporting standard  EFD    Electronic fiscal device  ESS    Electronic sales suppression  EU    European Union  GDP    Gross domestic product  ICT    Information and communications technology  IMF    International Monetary Fund  IRS    Internal Revenue Service  NTS    National tax service  OECD    Organisation for Economic Co‐operation and Development  POS    Point of sale  TIN    Taxpayer identification number  US    United States  VAT    Value added tax    45       46 Bibliography    Ali, Merima, Abdulaziz Shifa, Abebe Shimeles, and Firew Woldeyes. 2015. “Information Technology and Fiscal Capacity in a Developing Country: Evidence from Ethiopia.” Working Paper 31, International Centre for Tax and Development. A. T. Kearney and F. Schneider. 2009. “The Shadow Economy in Europe: Using Payment Systems to Combat the Shadow Economy.” A. T. Kearney. https://www.atkearney.com/documents/10192/1f89864d-903f-4683-b6e4-df57e223a4d7. Azevedo Pereira, José A.. 2014. “E-invoicing and other on-going projects in Portugal”. Presentation for the CIAT Technical Conference. Berhan, Bahro A., and Glenn P. Jenkins. 2005. “The High Costs of Controlling GST and VAT Evasion.” Canadian Tax Journal 53 (3): 720. Bornman, Marina, and Lilla Stack. 2015. “Rewarding Tax Compliance: Taxpayers’ Attitudes and Beliefs.” Journal of Economic and Financial Sciences 8 (3): 790–807. Brat, Ilan. 2013. “Spain’s New Tax Crackdown Makes It Personal.” Wall Street Journal, February 8. Burgess, Elizabeth, B. Mazzotta, J.M. San Juan, G. Soltys, A.V. Zuccolotto Soto, D. Aldama Navarrete. 2013. “Cash Outlook: Mexico.” Working Paper 13-02, IBGC, The Fletcher School, Tufts University, Medford, MA. Cardoza, Martin. 2012. “Fiscal Printers: Dominican Experience.” CIAT/AEAT/IEF Tax Administration Review 33: 16–33. Casey, Peter, and Patricio Castro. 2015. “Electronic Fiscal Devices (EFDs): An Empirical Study of Their Impact on Taxpayer Compliance and Administrative Efficiency.” IMF Working Paper WP/15/73. IMF, Washington, DC. CIAT (Inter-American Center of Tax Administrations). 2014. “Taxpayer Assistance Channels and Platforms.” Portuguese presentation for the 48th CIAT General Assembly, Rio de Janeiro, May 5–8. Croatia, Ministry of Finance. 2014. “Fiscalisation: Successful Tool Fighting Against Tax Fraud: Experience of Its Implementation.” Croatia, Ministry of Finance, Tax Administration, Zagreb. Eissa, Nada, Andrew Zeitlin, Saahil Karpe, and Sally Murray. 2014. “Incidence and Impact of Electronic Billing Machines for VAT in Rwanda.” International Growth Center, London. European Commission. 2013. “Developing Personal and Household Services in the EU: A Focus on Housework Activities.” European Commission Report, Brussels. EY (Ernst & Young) 2014. “Global Survey: VAT/GST Electronic Filing and Data Extraction.” EY Global Headquarters, London. Fabbri, Marco, and Sigrid Hemels. 2013. “‘Do You Want a Receipt?’ Combating VAT and RST Evasion with Lottery Tickets.” Intertax 8/9: 430–43.   47 Fooken, Jonas, Thomas Hemmelgarn, and Benedikt Herrmann. 2014. “Improving VAT Compliance: Random Awards for Tax Compliance.” European Commission Taxation Paper 51. European Commission, Brussels. Giebe, Thomas, and Paul Schweinzer. 2014. “Consuming Your Way to Efficiency: Public Goods Provision through Non-Distortionary Tax Lotteries.” European Journal of Political Economy 36 (C): 1–12. Gordon, James P. F. 1990. “Evading Taxes by Selling for Cash.” Oxford Economic Papers 42 (1): 244– 55. IMF (International Monetary Fund). 1996. “Selected Issues in Administering the VAT: Cross-Checking Invoices and Controlling Refunds to Exporters” (mimeo). IMF, Washington, DC. ______. 2015. “Electronic Fiscal Devices (EFDs) – An Empirical Study of their Impact on Taxpayer Compliance and Administrative Efficiency”. Working Paper WP/15/73. Katolik, Andreja. 2014. “Impact of Fiscalization at the Height of Calculated Value Added Tax.” Interdisciplinary Management Research 10: 545–56. Koch-Billentis, Bruno. 2014. “E-Invoicing/E-Billing: Key Stakeholders as Game Changers.” Research report prepared for Basware, New York. Korea, National Tax Service. 2013. Annual Report. National Tax Service, Seoul. Kudeljan, Marijana Vuraić. 2015. “Fiscalization in the Republic of Croatia.” Paper presented at conference of the Ministry of Finance, Zagreb, June 16. Lee, Hyun Jung. 2013.“Marketing Strategies for the Social Security and Tax Number System in Japan.” Mimeo. School of International and Public Policy, Hitotsubashi University. Lee, Hyung Chul. 2016. “Can Electronic Tax Invoicing Improve Tax Compliance? A Case Study of the Republic of Korea's Electronic Tax Invoicing for Value-Added Tax.” World Bank Policy Research Working Paper 7592. Madzharova, Boryana. 2014. “The Impact of Cash and Card Transactions on VAT Collection Efficiency.” Paper presented at the Deutsch Bundesbank “International Cash Conference,” Dresden, September 15–18. Marković, Branimir, and Domagoj Pavić. 2014. “The Effects of Fiscalization in Suppressing the Underground Economy in the Catering Industry.” Interdisciplinary Management Research 10: 575– 84. Mativo, Peter Mutisya. 2015. “Factors Affecting Utilization of Electronic Tax Registers (ETRs) in Small and Medium Enterprises in Kenya: A Case of Nairobi CBD.” International Journal of Economics, Commerce and Management III (3). Memon, Najeeb. 2013. “Looking at Pakistani Presumptive Income Tax through Principles of a Good Tax?” eJournal of Tax Research 11 (1): 40. Mohammed, Hamdu Kedir, and Zinash Degife Gela. 2014. “Challenges of Electronics Tax Register Machine (ETRS) to Businesses and Its Impact in Improving Tax Revenue.” International Journal of Scientific Knowledge 5 (3): 17–22.   48 Naritomi, Joana. 2016. “Consumers as Tax Auditors.” Working Paper April 2016, The London School of Economics and Political Science, London. OECD (Organisation for Economic Co-operation and Development). 2000. “Improving Access to Bank Information for Tax Purposes.” OECD, Paris. ———. 2009. Forum on Tax Administration, Withholding and Information Reporting Regimes for Small/Medium-sized Businesses and Self-employed Taxpayers. OECD, Paris. ———. 2011. Global Forum on Transparency and Exchange of Information for Tax Purposes, Peer Reviews: France. OECD, Paris. ———. 2013a. “Electronic Sales Suppression: A Threat to Tax Revenues.” OECD, Paris. ———. 2013b. Global Forum on Transparency and Exchange of Information for Tax Purposes, Peer Reviews: Estonia. OECD, Paris. ———. 2015. “Improving Co-Operation Between Tax and Anti-Money Laundering Authorities: Access by Tax Administrations to Information Held by Financial Intelligence Units for Criminal and Civil Purposes.” OECD, Paris. Schneider, Friedrich, Andreas Buen, and Claudio Montenegro. 2010. “New Estimates for the Shadow Economies All Over the World.” International Economic Journal, 24: 4, 443 – 461.. Shome, Parthasarathi. 2011. “Financial Transaction Taxes.” Working Paper 254, Indian Council for Research on International Economic Relations, New Delhi. Skatteverket. 2013. “Requirements of Cash Registers: Impact Evaluation.” English Summary of the Swedish Tax Agency Report. Skatteverket, Stockholm. Slemrod, Joel, Brett Collins, Jeffrey Hoopes, Daniel Reck, and Michael Sebastiani. 2014. “Does Credit- Card Information Reporting Improve Small-Business Tax Compliance?” NBER Working Paper 21412, National Bureau of Economic Research, Cambridge, MA. Soled, Jay A. 1997. “Honest Taxpayers and the Tax Gap: It Takes Two to Tango.” Tax Notes 76: 983. Torgler, Benno. 2003. “Beyond Punishment: A Tax Compliance Experiment with Taxpayers in Costa Rica.” Revista de Analisis Economico 18 (1): 27–56. Véber, Zita, and Brosch, Judit. 2013. “Can cash payment be limited in a modern payment system?” MNB Bulletin, January 2013, 52. Wan, Junmin. 2006. “The Incentive to Declare Taxes and Tax Revenue: The Lottery Receipt Experiment in China.” Paper presented at the International Business and Economics conference “Economic Integration and Economic Development,” Beijing. June 24–25. ———. 2014. Consumer Casualties: Exploring the Economics of Habit, Information, and Uncertainty in Japan. London: Palgrave Macmillan.   49 Weru, Jane Muthoni, Mary Wanjira Kamaara, and Alice Njeri Weru. 2013. “Impact of Strategic Change: Introduction of Electronic Tax Register for Enhancement of Tax Collection at Kenya Revenue Authority.” International Journal of Economics and Finance 1 (5). World Bank Group. 2009. Doing Business. World Bank, Washington, DC     50