Australia's Future Tax System

Final Report: Detailed Analysis

Chapter D: Taxing consumption

D4. Taxing financial services

D4–2 The nature and consequences of current arrangements

Current treatment compared to consumption benchmark

Services to facilitate the deferral of consumption should be included in the consumption tax base, like other goods and services. That is, the value of domestic private consumption of financial services should be taxed, while financial services provided to non-residents or business should not be taxed.

However, estimates based on the existing GST system suggest that the current tax treatment of financial services under the GST over-taxes business by around $760 million in 2010–11, while the failure to fully tax household consumption of financial services results in a $3.9 billion shortfall from the consumption tax benchmark (see Table D4–1).

Table D4–1: Cost of input tax treatment of financial supplies

  2008–09
$m
2009–10
$m
2010–11
$m
2011–12
$m
Households 3,580 3,710 3,890 4,090
Businesses –690 –720 –760 –790
Total 2,880 2,990 3,140 3,290

Source: Treasury estimate, against benchmark of taxing household final consumption at 10 per cent rate.

If final financial services were taxed on a consumption basis, this would result in increased bank fees or interest-rate spreads on financial services for private consumers, while reducing fees or interest-rate spreads for business consumers, and therefore the prices of other goods and services for which financial services are inputs.

Principle

The value added by the financial services sector to household consumption should be taxed in an equivalent way to consumption in other parts of the economy, while inputs to production should not be taxed.

Box D4–1: A Tobin tax?

Many submissions to the Review have supported the introduction of a tax on foreign currency transactions, commonly known as a 'Tobin tax', named after Nobel prize-winning economist James Tobin (1974), who first proposed the tax in a 1972 lecture. In the wake of the global financial crisis the idea has received new attention in the international policy debate. In August 2009, Lord Turner, chair of the UK Financial Services Authority, canvassed the possibility of a similar tax on all financial transactions to promote an efficient financial sector, particularly more stable financial markets. Keynes (1936) made a similar suggestion during the Great Depression.

The goal of a Tobin tax is to dampen de-stabilising speculative financial activity. By putting 'sand in the wheels' of the financial system, proponents believe that financial prices (such as foreign exchange rates) would be less likely to overshoot or undershoot economic fundamentals. If de-stabilising speculative transactions are more typically short-term and high-volume, they would be disproportionately affected by the tax, even though it would be levied at a low rate, based on value, to limit its impact on real activity. More recently, proponents have argued that the revenues could be used to finance international public goods, such as the United Nations or world poverty alleviation.

Transaction taxes like the Tobin tax are generally inefficient because the tax rate rises according to how often an asset changes hands, rather than any underlying economic value. There is no 'economic base' for transaction taxes. In general, transactions tend to create value because they shift resources to higher-value purposes. If these prices are publicly available, the transactions also provide the public information that assists wider resource allocation in the community.

Financial markets are not perfectly efficient. Notably, the global financial crisis resulted from a widespread mispricing of risk by financial markets. However, a financial transactions tax would not directly address the sources of financial market failure, such as moral hazard arising from implicit or explicit government guarantees, incentive structures skewed toward short-term gains, and human psychology. There is no necessary correlation between trading volume and the creation of systemic risk. The tax would apply indiscriminately to transactions that are socially useful — including those that contribute to financial system stability — and those that are costly.

In fact, transaction taxes could potentially reduce financial stability. They would reduce market liquidity, which could lead to prices becoming more volatile and more prone to misalignment. They would also impede hedging activity, which can involve a large volume of transactions to disperse risk. Although the great majority of financial transactions occur between financial firms, much of this is generated by the process of reallocating risk between financial firms rather than speculation. Further, speculation is not inherently destabilising as it can sometimes help correct misalignments.

It would be difficult to prevent activity shifting to unregulated sectors or jurisdictions. Businesses would also have an incentive to structure themselves to avoid the tax. For example, large, vertically integrated businesses use fewer transactions to make the same product and would pay less tax. Even if levied at a low rate, a tax would cause some impediment to real activity (for instance, currency transactions are essential for international trade and investment) and may impede some necessary adjustments.

Current treatment of financial services under the GST is inefficient

Australia and most other countries with a value added tax (VAT) use the 'invoice-credit' approach (see Section D2 The goods and services tax). Each business incurs a tax liability for its sales and claims a credit for purchases of taxed goods and services. Taxing financial services under this system is complex and inefficient, mainly because it is very difficult to measure the value of the services provided in individual financial transactions.

In many cases, the consideration received for financial services such as a loan is not explicit, but implicit in a margin or investment return. In the case of interest, part of the interest payment is to compensate the lender for financial services associated with the loan, such as assessment, monitoring and account keeping. However, part of the interest paid to a lender is to compensate the lender for the use of the loan funds and for the risk of default. In more complex transactions, a bank may obtain consideration for the services it provides in arranging a loan by means of an implicit margin in the various financial flows making up the transaction. Determining the value of the implicit consideration for supply of intermediation services, for each party to the transaction, poses considerable challenges.

Taxing financial services can also be problematic due to the nature of the financial institution as an intermediary. A bank, for example, typically does not make a loan using just its own capital. Instead, it obtains money from another source which it then lends. Often, the bank may receive consideration by way of a margin applying to both the borrowing and lending. Properly taxing the value of the services provided to the borrower and the lender requires an allocation of value between these various parties that can be problematic to achieve. This becomes particularly complex where the supply to one party needs to be treated differently to the other (for example, if one supply is an export).

The same problem would arise for the cash flow tax (see Section D1 A cash flow tax), which would also exclude financial cash flows from the tax.

Most countries with a VAT have opted for a 'second-best' solution to taxing financial services, using 'input taxation'. This means that inputs of the service provider are taxed, but not the value they add. The purchaser of financial services does not receive an input tax credit for the GST incurred on their business inputs. The treaty under which members of the European Union impose VAT requires this approach.

While businesses receive a credit for GST on their inputs, they are not able to recover all embedded tax when they or their suppliers have input taxed financial inputs. This approach results in various biases for both businesses and consumers. These can result in efficiency costs, including:

  • cascading of taxes through the supply chain, flowing through to higher prices of goods and services to consumers, businesses and exports;
  • businesses organising themselves to 'self-supply' goods and services to reduce the tax payable on their inputs. This gives large, vertically integrated businesses an advantage over smaller competitors;
  • complexities in apportioning the cost of inputs between taxable, input taxed and GST-free uses, including tracking the use of individual assets; and
  • adverse impacts on the financial sector's international competitiveness.

These impacts influence the way financial supply providers operate, and change the prices faced by consumers. For example, treating financial services as input taxed means that consumers do not bear the full GST and may encourage them to use more financial services over other consumption. Conversely, the relative price of financial services for business is higher, as the GST paid on other inputs is typically refunded in full.

This embedded tax is likely to be passed forward to consumers — resulting in effective tax rates above 10 per cent for taxable goods and services for which financial services are an input. Where an Australian exporter makes use of Australian financial supplies, the price of their exports — which should face no tax under a destination-based GST — can also include embedded tax from input-taxed financial services (see Table D4–2).

Table D4–2: GST with input taxation of financial services

  Non-financial goods and services
Explicit price = wages + economic rent
Financial services
Implicit price in interest margin
Households 10% tax on price 10% tax on some inputs into financial services
Business GST refunded, except for embedded tax 10% tax on some inputs into financial services
Exports GST free - some embedded tax GST free - some embedded tax

Australia's GST law includes additional, complex provisions to reduce some of the efficiency consequences of input-taxing financial services. While these provisions mitigate some of the inefficiencies of input taxation, they do not amount to efficient taxation of domestic consumption. These provisions include:

  • introducing a reduced input tax credit equal to 75 per cent of the full input tax credit for a defined range of acquisitions that would otherwise be fully input taxed. This is a unique feature of Australia's GST regime;
  • adopting a narrow definition of what constitutes a financial supply for GST purposes;
  • introducing a financial acquisitions threshold that excludes many financial supplies made by non-bank financial institutions from input taxation;
  • including a special exemption for certain borrowing costs where the funds are used in making GST-free or taxable supplies, to reduce tax cascading; and
  • avoiding an incentive for input taxed entities to acquire supplies from other countries by requiring those acquiring such supplies to make equivalent GST payments.

Finding

Financial services paid for through an interest margin, rather than explicit fees, cannot be taxed directly using an invoice-credit GST or a simple cash flow tax. The use of input taxation under GST potentially biases production and consumption decisions, resulting in large efficiency costs and additional complexity from special provisions designed to reduce the inefficiency.

Impact on Australia as a regional financial services centre

Other countries in the region — notably Singapore and New Zealand — have also attempted to address these problems by modifying their GST. However, neither has succeeded in treating financial services equivalently to other forms of taxed consumption. Nevertheless, these innovations may have given these countries an edge that may harm Australia's efforts to position itself as a regional financial services centre.

Singapore

Singapore input-taxes financial services but allows the service provider an input tax credit under either a 'special method' or the 'fixed input tax recovery method'.

The former requires separate reporting of the value of certain services as a proportion of total services. This adds to compliance costs. The latter calculates the input tax credit entitlement by applying a ratio to total acquisitions. The ratio is determined by the tax authorities annually, and varies for different types of banking licences. For example, banks with full banking licences, wholesale banks, offshore banks and finance companies each apply a different ratio. This is similar to Australia's 75 per cent reduced input tax credit, although in Australia this applies only to a limited range of inputs.

New Zealand

Since 1 January 2005, New Zealand's GST has allowed business-to-business supplies of financial services to be GST-free where, over a 12-month period, the recipient's taxable supplies exceed 75 per cent of their total supplies. This GST-free treatment was introduced to remove embedded tax on business inputs caused by input taxation.

This approach requires financial institutions to obtain information about the eligibility of their customers to claim input tax credits. This information is usually unnecessary under a GST. However, to reduce compliance costs, suppliers can refer to Australian and New Zealand Standard Industrial Classification codes to determine the recipients' status.

Treating business-to-business transactions as GST-free requires an appropriate way to determine eligibility for claiming input tax credits, particularly for overhead costs. There is also a revenue risk that GST-free supplies may be made for final consumption. New Zealand's GST includes additional anti-avoidance measures to address this threat.

Finding

The input taxation of financial services under the GST, and associated provisions to give relief from it, are complex and affect Australia's position as a regional financial services centre.