COMMISSION STAFF WORKING PAPER EXECUTIVE SUMMARY OF THE IMPACT ASSESSMENT
/* SEC/2011/1103 final */
|Bilingual display: EN|
Executive Summary of the Impact Assessment Report on "Instruments for the Taxation of the Financial Sector"
In its Communication "Taxation of the Financial Sector" of 7 October 2010, the Commission announced the launch of a comprehensive Impact Assessment (IA) to examine further the options for the taxation of the financial sector to be in a position to make appropriate policy proposals. This IA therefore analyzes various tax schemes and specifies the appropriate design of a FTT and its possible impact.
2. Procedural issues and consultation of interested parties
As required by the Impact Assessment Guidelines, this Impact Assessment has benefited from the consultation of all interested parties. It also benefited from the results of ad hoc external studies.
3. Problem definition and Subsidiarity 3.1. Problem definition
Costs of the crisis
A common claim in public debates is that the financial sector should bear its fair share of the costs of the financial crisis. Member States individually committed to support the financial sector for a total of about EUR 4.6 trillion (39% of EU-27 GDP in 2009). This has aggravated the situation of public finances and such situation is hardly sustainable from a fiscal point of view and imposes a heavy burden on the present and future generations.
VAT exemption of financial services
Article 135(1) of the VAT Directive provides an exemption from VAT for most financial and insurance services. The analysis suggests that the VAT exemption leads to a tax advantage for the financial sector in the range of 0.15% of GDP. It results in a preferential treatment of the financial sector compared with other sectors of the economy as well as in distortions of prices.
Market failure and systemic risks in the financial sector
The crisis resulted from the complex interaction of market failures, global financial and monetary imbalances, inappropriate regulation, weak supervision and poor macro-prudential oversight. It has been argued that taxes could be used as regulatory tools to address the problems such as implicit or explicit guarantees; automated trading (and more generally short-term rent-seeking); different tax treatment of debt and equity; excessive remuneration schemes that encourage risk-taking; large amounts of complex derivatives, and the existence of economic rents, leading principally to wage premiums in the sector.
Internal Market Aspects
In the aftermath of the financial crisis, Member States have reacted by imposing various tax and levies on the financial sector. Such unilateral action might lead to relocation and distortions of competition in the Single Market and calls for more coordinated actions. Coordination is also warranted because of the risk of double-taxation.
3.2. EU right to act and subsidiarity
The right for the EU to act in relation to taxes on the financial sector would be based on Articles 113 and 115 of the Treaty on the Functioning of the European Union (TFEU). The main rationale for the EU action is that the functioning of the Internal Market would be hampered if Member States decide to act unilaterally in this field. At least 12 Member States have introduced or, are analysing the possibility of introducing, bank levies on financial institutions. This uncoordinated introduction of taxes on the financial sector fragments the EU financial market, distorts competition and increases the risks of relocation of the financial activities both within and outside the EU. It would also increase the risk that the financial sector becomes subject to double taxation, which would in turn hinder the exercise of the fundamental freedoms protected by the TFEU.
It must therefore be concluded that EU action would respect the subsidiarity principle since the policy objectives cannot be sufficiently achieved by actions of the Member States, and can be better achieved at EU-level.
The following general objectives can be defined:
· Raising revenue and adequate contribution from the financial sector to tax revenues;
· Limiting undesirable market behaviour and thereby stabilizing markets;
· Ensuring the functioning of the Internal Market (avoid double taxation and distortion of competition).
General Problems || Specific problems || General objectives || Specific objectives
Costs of the crisis (Fiscal consolidation issues) Tax advantage for the financial sector || Substantial public financial support and other budgetary effects of the financial crisis led to need for budget consolidation VAT-exemption of financial services || Raising revenue Adequate (fair and substantial) contribution from the financial sector || (1) Identifying new revenue sources
(2) Recover costs of the recent financial crisis
(3) Cover the budgetary costs of potential future financial crises
(4) Compensate for VAT exemption of financial services
Market failure and systemic risks in the financial sector || (1) Undesirable market behaviour due to implicit guarantees (moral hazard) || Reducing undesirable market behaviour and thereby stabilizing markets || Reduce incentives for excessive risk-taking incentives
(2) Not properly managed and supervised automated trading || Address specific risks posed by automated trading
(3) Distortions of debt-equity choice: Leverage (4) Excessive executive compensation schemes encourage risk-taking || Reduce leverage – debt-to-equity ratio
(5) Complex products and counterparty risk || Ensuring the functioning of the Internal Market by a coordination of the measures to be introduced || Avoid distortions within the EU, ensure efficiency of the measures and safeguard relative competitive position within the EU
(6) Economic rents || Avoid double taxation
5. Policy options
The policy options analysed in the IA (Financial Transactions Tax (FTT) and a Financial Activities Tax (FAT) areassessed against the baseline scenario.
The baseline scenario is the scenario without introducing taxes. It is characterised by the VAT exemption of the sector that emerges as the main difference of tax treatment between the financial sector and other sectors of the economy, by high levels of executive compensations in the financial sector, by recent unilateral actions by Members States to implement bank levies as well as by a set of national taxes on either the sector or on financial transactions, by risks of double-taxation and by many recent initiatives in regulating the sector.
This baseline scenario also takes into account certain regulatory measures for the financial sector, which are currently being proposed and which also pursue some of the objectives defined above. More particularly, the IA assumes the adoption of those regulatory measures that are aimed at decreasing the individual as well as the systemic risk in the financial sector by increasing capitalization and capital buffers for both the financial institutions and the financial system as a whole. Some other policy options such as additional regulation, removing the VAT exemption or addressing tax differences for debt and equity by changes in the corporate tax systems are discarded on the ground that they do not address a sufficient number of the identified problems.
6. Comparing policy options 6.1. Raising revenue and adequate contribution from the financial sector to tax revenues
Identifying new revenue sources
Both tax instruments – FTT and FAT – could be new revenue sources as they are able to generate significant amounts of tax revenue for public budgets. However, there is a large uncertainty about the real revenue potential, given the number of unknown variables and assumptions in the estimations.
For the FAT, the revenues at an illustrative tax rate of 5% range between EUR 9.3 billion and EUR 30.3 billion depending on assumptions on relocation and design.
There is greater revenue potential for the FTT. Estimations may range between EUR 16.4 billion and EUR 400 billion depending on assumptions on decrease in volume, the scope of products covered and the tax rates (0.01% for the first estimate and 0.1% for the second). The relocation assumption would cover migration of transactions and reduction of volume of activity, notably of high speed trading transactions.
Recover costs of the recent and future financial crises
Imposing new taxes on the financial sector would contribute to recover the budgetary cost of the recent crisis if it is argued that the whole sector indirectly benefited from the financial aids granted to certain financial institutions. Furthermore, the so-called polluter-pays-principle would justify requiring that the financial sector further contributes to the government budget, with a view to covering the costs of future crises that go beyond the resolution mechanisms.
6.2. Compensate for VAT exemption of Financial Services
Although the addition-method FAT would partly address the VAT exemption for the financial sector, the IA concludes that this issue should ideally be solved in the context of the Green Paper on VAT.
6.3. Correcting undesirable market behaviour and thereby stabilizing financial markets
Reduce incentives for excessive risk-taking
Regulatory measures more closely linked to the sources of systemic risk might be more appropriate to deal with excessive risk taking.
However, the FTT might be an appropriate tool to reduce excessive risk-taking to the extent that short-term trading and highly leveraged derivative trading creates systemic risks. The FAT would only be an indirect measure to tackle risk-taking.
Address specific risks posed by automated trading
The FTT, de facto increasing the costs of transactions in financial markets, could be used as a measure to reduce Automated Trading, and notably its subset of High-Frequency-Trading. The effects of additional transaction costs might be particularly strong on some market segments characterised by very narrow spreads.
The FAT does not have a direct impact on the trading behaviour in financial markets.
Reduce leverage – debt-to-equity ratio
Neither the FTT nor the FAT address the distortions of financing decisions and the incentives to excessive leverage created by the different treatment of debt and equity under the current corporate tax systems. Regulatory measures are more appropriate for this purpose.
Taxing economic rents - Introduce measures to tax value added in the financial sector?
The FTT would have limited merits as a tax on economic rents. The FAT can be specifically designed to tax economic rents, although its effectiveness will depend on the applicability of the notion of normal returns to the tax base.
6.4. Ensuring the functioning of the Internal Market
In order to ensure the functioning of the Internal Market and to minimize economic distortions within the EU, both tax instruments should be ideally introduced with a high degree of harmonisation of the tax bases and the tax rates. More particularly, both the FTT and the FAT should be introduced in a harmonised way to avoid double taxation or non-taxation.
6.5. Cumulative effects
Although the cumulative effect of the tax with regulatory changes is more easily measurable for the FAT than for the FTT, it is likely that the impact of both FTT and FAT on most economic variables (cost of credit, macroeconomic variables and at the sector's level) simply adds to the effects of the regulatory measures currently being proposed.
6.6. Overview and concluding remarks
The analysis showed that both instruments – FTT and FAT – are technically feasible but the choice of one over the other is essentially a trade-off among the different objectives pursued and would also depend on the specific design features of the tax.
Both taxes seem to have the potential for raising significant tax revenues from the financial sector, although to a potentially greater extent for the FTT.
The FTT comes with a higher risk of relocation or disappearance of transactions, especially with respect to frequent short-term transactions. To this extend, Automated Trading in financial markets could be affected by a tax-induced increase in transaction costs, so that these costs would significantly erode the marginal profit, thereby affecting the business model of high-frequency trading.
Both taxes are also expected to have small but non-trivial effects on GDP and employment, with the negative effects of the FTT probably being somewhat higher. The reason for this negative effect is the increase in the cost of capital, as the taxed persons will try to pass the tax through to their clients, and which then negatively interacts with investment. The distributional implications of a tax on the financial sector are typically progressive, i.e. such taxes fall more on the richer parts of society than on its poorer parts, as the first make more use and benefit more from the services provided by the financial sector. This holds especially for the FTT in case it was limited to transactions with financial instruments such as bonds and shares and derivatives thereof.
On 29th June 2011, the European Commission put forward a proposal for the Multiannual Financial Framework 2014-2020. In this context, it proposed to introduce a FTT in the EU as a first step, taking into account its revenue potential and its impact on excessive specific risk-taking. The Commission Staff Working Paper accompanying the proposal identifies several elements which need be taken into consideration to mitigate the risks of relocation when a FTT would be implemented at EU level, which are discussed in the following section.
7. Analysis of impacts of different variants of a FTT 7.1. Products and transactions covered
The policy options distinguish between a tax on (i) currency spot and derivatives transactions, a tax on (ii) securities transaction on primary and on secondary markets, and a tax on (iii) derivatives markets, excluding derivatives on currency exchanges. All markets covered are both regulated exchange markets and trading "over the counter".
Currency transaction tax (CTT)
The collection would take place centrally at currency exchange systems, namely in Real Time Gross Settlement (RTGS). For transactions which are tracked in such systems, a tax could be levied centrally. Note that this option poses considerable legal issues. We therefore distinguish between spot currency markets and derivative markets for currencies.
Securities transaction tax (STT) without derivatives and currency transactions
It corresponds to the narrow-based transaction tax which would tax only the spot transactions of equities and bonds (on primary and secondary markets). This option is similar to the UK stamp duty when combined with the domestic issuance principle. Alternatively, one could exclude primary markets, so as to avoid making such capital-raising more expensive. Then such a securities transaction tax would only be levied on the trading of shares and bonds on secondary markets.
Financial tax on derivatives
It would cover all financial derivatives that are directly or indirectly derived from products traded on regulated exchanges or traded over the counter. This can include financial derivatives for currency transactions securities or commodity based derivatives, risk or interest rates based or all other financial derivatives that are not outright spot transactions and transactions for the raising of capital, such as bonds and shares.
Financial transaction tax
This option would cover all financial transactions as outlined above or a subset thereof. Such subset could e.g. exclude taxing outright spot transactions on currency markets and exclude taxing the raising of capital through the issuing of bonds or shares.
Pros and cons of the different options
The directive on markets in financial instruments already contains a well-established classification of financial instruments and products, which is taken as a starting point for defining the scope of the FTT proposal with respect to products to be covered.
All products that define financial markets and financial transactions and that could be close substitutes should form part of the basket of products covered. Going for narrower definitions would invite for large-scale tax-avoidance activities. Thus, an FTT covering all securities and all derivatives thereof seems to be the most suitable coverage of taxable products under an FTT.
As regards currency transactions, there might be a potential conflict between the objective of raising additional revenue and the objective not to impose restrictions to the free movement of capital. This holds especially for spot currency transactions. Especially taxing currency transactions involving different EU currencies is typically assumed being in violation of Article 63 TFEU, while taxing currency transactions involving non-EU currencies could potentially be made compatible in case the Council decided to do so in applying Article 64.3 TFEU that enables it.
7.2. Taxable event
The taxable event could either be based on the accrual or on the cash principle. For some transactions such a distinction could be rather irrelevant as both occur at the same time, especially when transactions are carried out electronically. This is typically the case in financial markets. However, in case one opted for a cash-based approach only, there might be an incentive for deferring payments so as to benefit from substantial cash-flow advantages. Also in cases where no exchange of instruments takes place, the taxable event would still occur. For derivatives, the taxable event would then typically be the moment when the contract is agreed upon. The IA suggests that the best option would be a hybrid of the cash and accrual based principles, depending on products considered.
7.3. Place of Taxation
In order to define the financial transactions that each jurisdiction would be entitled to tax under the FTT, different principles could be used (residence of the parties, place of the transaction, and place of issuance of the financial instrument). Those approaches have different challenges with regard to tax collection mechanisms, enforcement and revenue distribution. The IA concludes that the residence principle is suggested as the best option, taking into account the need to mitigate relocation risks and the requirements for a simple tax administration scheme.
Of the various options discussed, the FTT could be levied on the financial institutions being party or intervening on behalf of a party. This would have the advantage of targeting the legal incidence on the financial sector and of avoiding the problematic of potentially high administrative costs of a FTT on non-financial actors. In order to catch also intra-group financing and shadow-banking activities, it is suggested that all the enterprises conducting more than a certain threshold of financial activities should become subject to the FTT too.
The IA suggests that the option of making the parties to a transaction established in the EU liable to pay the tax due in case they trade with non-EU financial institutions is an appropriate way to fight tax avoidance.
7.5. Taxable base
The definition of the taxable base for spot transactions should not pose serious problems. Although some derivatives do have their own value at the moment of contracting, for many the only readily available reference value, which could serve as a taxable base is the value of the underlying instrument or asset. The IA suggests using (i) the gross transaction value for spot transactions, and (ii) the value of the asset underlying a derivative contract (i.e. notional value).
7.6. Tax rate
A low flat statutory tax rate is assumed to be essential to avoid strong negative impacts on markets and to ensure some revenue collection, since the incentives for avoidance increase with the tax rate. The IA also suggests differentiated rate by category of products as a good way to mitigate the relocation risks while ensuring appropriate revenues.
7.7. Conclusions on the design
For an efficient application of any version of an FTT in the EU, a common definition of the scope of the tax, tax rates as well as on the precise tax bases and other essential features of the tax is highly advisable. Firstly, there are strong economic reasons for a high degree of harmonisation and co-ordination in order to avoid substitution and loopholes. Secondly, there are technical arguments. Some national exchanges have merged in recent years and use the same technical infrastructure. These systems are highly integrated and co-ordination of tax administration and enforcement between the countries is advisable. Again, applying the residence principle as outlined above could mitigate these issues. Therefore the preferred option for a harmonised FTT is a tax on the trading of financial instruments and derivatives thereof by financial institutions and covering shares, bonds and related derivatives – at notional value – and based on the residence principle. The inclusion of spot currency transactions is subject to its legal feasibility. Primary markets for bonds and shares would be exempted as to mitigate the direct impacts on the financing of companies in the real economy.
7.8. Impact of FTT
The revenue estimates for an FTT at a tax rate of e.g. 0.01% are between EUR 16.4 (with an elasticity of -2 and high volume decrease) and EUR 43.4 billion (with an elasticity of 0 and low volume decrease) (0.13% to 0.35% of GDP) when all sources are accounted for and the value of the asset underlying the transaction was taken as the taxable amount. If the rate is increased to 0.1%, total estimated revenues are between EUR 73.3 (with an elasticity of -2 and high volume decrease) and EUR 433.9 billion (with an elasticity of 0 and low volume decrease) (0.60% to 3.54% GDP).
The revenue estimates for the variants of FTT heavily depend on the assumptions on volume decrease and on the elasticity of remaining trade volumes to the tax. The IA has retained two working assumptions on volume decrease (including relocation and disappearance), one (70% decrease) in line with existing studies on reaction in derivative markets, the second one (90% decrease) slightly higher and based on the Swedish experience. These assumptions are also based on the development of high-frequency trading which now account for about 40% of equity markets and would be severely affected by a FTT. The IA suggests that some elements of the design of the tax such as the place of taxation based on residence principle and the level of the rates are the most promising elements to mitigate the relocation risks.
The introduction of additional taxes bears the intrinsic risk of agents relocating their activities to reduce the fiscal burden. Relocation might take place by moving the relevant activities to jurisdictions where they are taxed less, or by shifting to products/suppliers outside the scope of taxation within the same jurisdiction. Obviously, the risk of physical relocation of markets/market players and migration to non-taxed products decreases the more widespread is the adoption of the taxes and the broader is their scope.
Macroeconomic impacts and employment
The model used to analyse the macroeconomic impacts suggests that at 0.1%, a transaction tax on securities could, without the application of mitigating effects, reduce future GDP growth in the long run by 1.76% of GDP and of 0.17% at a rate of 0.01%,. However, these results should be interpreted with caution given certain limitations of the underlying models. A limitation of the model is that it only takes into account the effects of an FTT on one source of financing, namely the issuance of equity. The effects on the costs of debt financing are not taken into account. Due to this limitation the model could overestimate the negative GDP effects of the tax. Also, there is no available model to assess the proposed mitigating effects from the design of the proposal (e.g. exclusion of primary markets, excluding most transactions that do not involve at least one financial institution, etc.) and the channels through which they impact macroeconomic variables. The only available approach is therefore to proxy the effects with the large caveats and uncertainties that such an exercise may carry. Under the assumption that all mitigating effects play in full, the output losses of the scenario without mitigating effects could be reduced from 1.76% to 0.53% of future GDP growth. In the modelled scenarios, securities transaction taxes at rates of 0.01% and 0.1% would have limited negative impacts on employment of respectively -0.03% and -0.20%.
Risk-taking and behavioural effects
The aspects of dealing with risk and behavioural effects of the FTT relate to the possibility of the FTT to curb speculation, noise trading and technical trade, and to decrease markets' volatility. While the economic literature concludes that the effects of the FTT on volatility is largely inconclusive and depends on market structure, it would be a very effective tool to curb automated high-frequency trading and highly leveraged derivatives.
Impact on capital costs and growth
The effects of transaction taxes on the cost of capital (and thus firms' investment behaviour) have been investigated. A general theoretical result is that higher transactions costs, including those imposed by transaction taxes, are associated with lower asset prices. This may increase the cost of capital faced by firms. The effects on the security prices and on the cost of capital increase with the tax rate and are dampened by longer holding periods. Empirical studies of the impact of the tax on financial markets generally confirm the theoretical result. The IA suggests that the exclusion of primary markets for bonds and shares from the scope of the tax could be a promising way to mitigate the impact on capital costs.
Incidence and distributional impacts
A large part of the burden would fall on direct and indirect owners of traded financial instruments. Specific design features of a FTT, such as those suggested in terms of definition of the taxpayer, could mitigate the impact of increased capital costs on the non-financial sectors.
Impacts on market structures and competitiveness effects
The effects of a transaction tax depend on the market structure. This structure differs between market segments and also between countries. If this structure is heterogeneous, the tax might affect the markets in question very differently. The empirical literature comes thus to different results when evaluating the effects of transaction taxes. While most studies find that trade volume is reduced, the effects on volatility and prices is less clear even though results based on panel data and estimation approaches that better identify transaction cost effects seem to find more often a positive relationship between transaction costs and volatility.
The impacts of the taxes on the competitiveness of the real economy (industry and services) depend to a large degree on the design of the tax. The IA suggest that limiting the taxable products to financial instruments as defined in the directive on markets in financial instruments (MiFID), thus excluding everyday financial transactions (such as the payment of bills or of the salary) and transactions such as loans from banks to enterprises or private households (such as mortgage loans or the provision of consumer credits) goes in the direction of ring-fencing the real economy from the direct effects of levying this tax.
8. Monitoring and Evaluation
The evaluation of the macroeconomic and microeconomic consequences of the application of the legislative measure could take place three years after the entry into force of the legislative measures implementing the Directive. The Commission could then submit to the European Parliament and the Council a report on the technical functioning of the Directive.
 European Commission. SEC(2011) 876 final, p. 29-30.