By Jeffrey Owens, Director, and Piet Battiau, OECD Centre for Tax Policy and Administration, Alain Charlet, WTS France*
Like the OECD, VAT has also been around for about 50 years. Is it time to reform some of the older, more unwieldy versions and go for a trimmer, broad-base, standard-rate VAT system instead?
Value-added tax is but a baby in the history of tax. It was a German businessman, Wilhelm Von Siemens, who, in the 1920s, first described the idea of putting a tax on the additional value of output at each stage of its creation, rather than just at the retail sales end. But it was a Frenchman, Maurice Laurй, a tax official, who was the first to implement one in 1954. VAT in various guises subsequently appeared in former French West Africa in the 1960s and in Brazil in 1965. By the late eighties, some 48 countries had VAT, mainly in Western Europe and Latin America. VAT is now implemented in over 150 countries, where it accounts on average for as much as a fifth of the total tax revenue. Today, all OECD countries, with the exception of the US, have VAT systems in place. In fact, revenues from VAT represented on average 18.7% of total tax revenues of OECD countries in 2008, compared to 8.8% in 1975.
Why have VAT systems become so popular with governments? For a start, they are effective in terms of raising revenue and are cost-effective to administer compared with other taxes. VAT is actually collected by businesses at each stage of the production and distribution chain. Along the stages of the value chain, businesses are charged a tax for the inputs they buy to produce further goods or services, but can recover that tax in the price of the good or service they sell to the next supplier. This means the total tax levied at each output stage is a constant portion of the value added to the good or service. In the end the total tax collected throughout the chain should correspond to the VAT paid by the final consumer. This makes it a secure tax to collect, compared with conventional sales taxes which can be lost if evasion happens right at the final sales stage. It is also an attractive tax from the taxpayers’ perspective because of its transparent nature: people know what they pay, and if they don’t buy the goods, they don’t pay the tax.
There are economic attractions to VAT too. Research by OECD tax economists suggests that VAT is better for economic growth than income taxes, in that VAT has less negative impact on the economic decisions of households and businesses than income tax. For example, VAT does not discourage savings and investment. In international trade, VAT has proven to be the preferred alternative for customs duties in the context of trade liberalisation: it taxes importation and zero-rate exports, so it does not affect international competiveness. From a distributional perspective, there is more debate: some see it as a regressive tax, affecting lower income earners more than higher ones. This is countered by the argument that consumers have a choice to buy or not buy certain items, and, where daily essentials, such as food and shoes, are concerned, policymakers can intervene for equity purposes. Any debate on distributional issues needs to focus on the entire tax and benefit system, and not just on one tax in isolation.
Given the recent financial and economic crises, tax revenues from most other taxes, including personal and corporate income taxes, risk being relatively low for the next few years and could take time to recover. Companies will generate less income tax, and with unemployment higher, so will households. Also revenues from social contributions will be relatively low. The everincreasing importance of VAT is therefore likely to continue. Hence the importance of ensuring that VAT systems are designed and operated efficiently.
One area of priority is to make the system less complex. Simple tax systems are more efficient and offer more certainty to taxpayers. One way of reducing complexity is to broaden the VAT base, so that goods and services that are now subject to zero and reduced rates would be taxed at the standard rate. The use of multiple VAT rates creates complexity, and increases administrative and compliance costs.
OECD countries can be classified into two broad categories. The first group–mostly European Union countries–has a narrow base of goods and services taxed at a standard rate, with reduced or zero rates surrounding it. The European VAT Directive allows EU member states to apply a standard rate of minimum 15% and two reduced rates that can go no lower than 5%. However, there are “reserved rights” for older EU member states to a lower rate, if that rate was in place before 1991. The main motivation for Europe’s reduced rates was to alleviate tax on basic necessities, such as food and clothing, for poor households.
The second group of countries has a much broader base at the standard rate. This is how VAT is designed in Australia, Canada, Korea, New Zealand, Singapore and South Africa, drawing lessons from the difficulties of rate differentiation within the EU. New Zealand introduced a Goods and Services Tax (GST) in 1986 that is levied on a broad base at a low single standard rate, with few exceptions or exemptions. The result is that New Zealand scores the highest on the OECD “VAT revenue ratio”, which is an indicator that attempts to measure the gap between the revenues that would arise from a theoretically pure VAT system (a single rate with full compliance and full tax collection) and the revenues actually collected.
Developing countries seem to have taken to these single-rate systems too. Of the 21 African countries that adopted VAT between 1990 and 1999, 14 have a single rate system, as have eight of the nine African countries that have adopted VAT since 2000. South Africa, which implemented VAT in 1991, only exempts or zero-rates a limited number of items such as basic foodstuffs and paraffin to help the poorest households. It also keeps its system quite simple by applying a relatively high VAT registration threshold so as to exclude small firms from the system.
Another feature worth noting is that the standard rate in such countries is often lower than the European Union minimum of 15%, with 10% in Australia, 15% in New Zealand and 14% in South Africa for instance.
Switzerland, interestingly, began a public consultation in 2007 as part of an assessment of its VAT system 10 years after its introduction. The first part of the reform came into force in January 2010. This should help reduce compliance costs for businesses. The second part of the reform– at this stage still only a proposal–would replace the three existing VAT rates of 2.5%, 3.8% and 8% by a single rate of 6.1%. It would also remove 20 of the 25 exemptions. Some studies suggest that the reform could reduce business compliance costs by 20- 30% and add 0.1-0.7% to economic growth.
OECD economists have long supported the view that, all things being equal, a broad base, single-rate VAT is the preferred tax approach. It is a growth-oriented approach that allows governments to increase their revenues while cutting administration and compliance costs. The European Commission, in its December 2010 Green Paper on the Future of VAT suggests that a “broad-based VAT system, ideally with a single rate, would be quite close to the ideal of a pure consumption tax that minimises compliance costs”.
But what about equity and redistributive issues? After all, many countries justify reduced rates as a way of not penalising the poorest households for buying daily essentials, since these households spend more of their income on food and clothing. However, research on the merits of reduced rates suggests that they are rarely effective in achieving distributional objectives.
Reduced rates are not an effective way of alleviating the tax burden on lowincome individuals. Indeed, the wealthy also benefit from reduced rates and, as they consume more, they benefit more too. A more effective policy would be to apply a single VAT rate and to implement compensatory measures that are directly targeted at increasing the real income of poorer households. The Mirrlees review, for instance, produced by the Institute for Fiscal Studies, argues that, in the UK, “ending all current zero and reduced rates (except for housing and exports) while increasing all means-tested benefit and tax credit rates by 15% would leave the poorest 30% of the population better off …”. The Swiss government in its reforms appears to back this view and has argued that redistribution policy will be better achieved via a single VAT rate rather than via one with reduced rates or exceptions.
On the other hand, the 2007 Copenhagen Economics study on the use of reduced VAT rates in the European Union suggested that reduced rates in carefully targeted sectors may provide some benefits, for instance when the locally supplied services employ many low-skilled workers. This may discourage working in the informal sector, for instance.
However, overall experience with reduced VAT rates confirms that they rarely achieve their objective. The 2001 South Africa zero-rating of paraffin for instance proved ineffective as suppliers took the benefits but did not pass them on to consumers. Or take France’s decision in 2009 to cut VAT from 19.6% to 5.5% in a bid to bolster restaurant and catering services. According to the French national statistics institute, restaurant prices decreased by no more than 1.1% in July 2009 and actually edged up a little by October. Only 30% of the VAT cut would have been passed on to customers, studies showed, and the entire exercise could end up costing the government in lost revenue. More recently, this time in Germany, it has been reported that a VAT cut in the hotel industry, as part of the 2010 tax cut, would not have been passed on to consumers. Some suggested the money saved by businesses was invested in renovation and acquisitions, instead of being passed on to customers.
All this is not to say that VAT does not have a regressive effect on poor households, but rather, that other means can be more effective in achieving redistribution goals than reduced VAT rates. In any case, VAT reform should not be considered in isolation from the tax system as a whole. Accompanying measures to compensate those that would lose out should be studied carefully, notably in terms of personal income tax, as well as social benefits. New Zealand’s GST reform for instance, was balanced via changes in income tax and social welfare, with the revenues generated from the GST being used to finance redistribution.
In addition, a single standard rate decreases the cost of administration whereas multiple VAT rate structures with numerous exemptions make compliance more difficult for taxpayers and for tax administrations. Furthermore, they may lead to legal uncertainty, as similar products can be subject to the standard or reduced rate depending on their nature, as one famous UK case on whether a potato crisp should be taxed at the same low rate as a biscuit showed. Very recently a French high court decided that margarine should be taxed at the standard rate instead of the reduced rate that applies to butter, on the basis that public interest requires favouring dairy products. Such disputes can be timeconsuming, and open up opportunities for lobbying and unwarranted tax planning or avoidance.
Tax is ultimately the expression of political consensus and democratic debate. It reflects citizens’ preferences, which vary from culture to culture. For policymakers wishing to reform VAT into a single-rate system, careful assessment of the costs and benefits, followed by public consultation, is essential. This is not just a matter of principle, but political longevity. Consider New Zealand again, which introduced its GST system in 1986 after much public discussion. This GST was levied on all products, including food and clothes, and was accompanied by redistribution initiatives for the poor. But the public was won over. In contrast, the political party that proposed introducing exemptions in 1987 lost the subsequent election.
Just as with the OECD, after more than 50 years VAT may now be at a turning point in its life. As countries continue to deal with fiscal consolidation in the wake of the financial and economic crises and seek to restore growth, and as the increasing budgetary importance of VAT is likely to continue, it is time to consider reform to improve the efficiency of VAT systems. More analysis is needed, but the evidence so far suggests that reform to broaden the VAT base would be good for economic growth, revenues and social objectives. Political leadership is the value-added ingredient needed to make such reforms happen.
*Mr Charlet participates in the OECD Technical Advisory Group (TAG) on Consumption Taxes as a member of the Business and Industry Advisory Committee (BIAC).
**The authors would like to give special thanks to David Holmes and Stйphane Buydens for their suggestions.
Charlet, Alain and Jeffrey Owens (2010), “An International Perspective on VAT”, in Tax Notes International, Vol. 59, No 12, September 2010
Johansson, Вsa, Christopher Heady, Jens Arnold, Bert Brys and Laura Vartia, “Tax and Economic Growth”, OECD Economics Department Working Paper n° 620, 11 July 2008, p.42-45
Crawford, Ian, Michael Keen and Stephen Smith (2010), “Value Added Tax and Excises”, prepared for the report of a commission on Reforming the Tax System for the 21 st Century, chaired by Sir James Mirrlees, www.ifs.org.uk/mirrleesReview
Dickson, Ian and David White, ‘‘Tax Design Insights from the New Zealand Goods and Services Tax (GST) Model,’’ Working Paper 60, Apr. 2008, Centre for Accounting, Governance, and Taxation Research, School of Accounting and Commercial Law, Victoria University of Wellington
Go, Delfin S. Marna Kearney, Sherman Robinson and Karen Thierfelder, “An Analysis of South Africa’s Value Added Tax”, World Bank Policy Research Working Paper 3671, August 2005, page 19
OECD (2011), Consumption Tax Trends 2010: VAT/GST and Excise Rates, Trends and Administration Issues
©OECD Observer No 284, Q1 2011