6.2 Better management of the Commonwealth's balance sheet

The government balance sheet is an important measure of its financial position at a point in time. The balance sheet lists the government assets and liabilities that can be reliably measured in accordance with relevant accounting standards.

There would be advantages in raising awareness within the community of the importance of the Commonwealth’s balance sheet.

Background

Having a strong balance sheet is important because:

  • the state of the balance sheet is a key determinant of the borrowing costs a government faces and the willingness of the private sector to lend to government. Credit ratings agencies carefully examine government balance sheets;
  • liabilities on the balance sheet have to be serviced and repaid which reduces the government’s capacity to fund other services;
  • repairing poor balance sheets requires a higher burden on future taxpayers and reductions in government services;
  • a strong balance sheet allows government to respond to economic shocks (such as financial crises) without calling into question their fiscal sustainability or the government having to raise taxes or cut spending sharply; and
  • sustainable government debt levels contribute to macroeconomic stability.

Governments can take risks on to their balance sheet when they provide guarantees and loans to entities. As has been seen recently in Europe and the United States, these risks can materialise quite quickly and have significant impacts on government finances.

Current structure of the Commonwealth Government balance sheet

The Commonwealth Government balance sheet is structured somewhat differently to other jurisdictions. Most assets are financial assets, such as the Future Fund, rather than physical assets.

In terms of non-financial assets, the Commonwealth plays a significant role by providing funds to the States, who invest and own assets such as schools, roads and hospitals. This contrasts with the situation for unitary systems of government such as New Zealand and the United Kingdom, where the national government owns and administers road systems, hospitals and schools on its own accord.

The Commonwealth Government general government balance sheet as at the end of 2012‑13 (Australian Government, 2013a) is made up of:

  • Financial assets totalling $250.8 billion, including the Future Fund ($88.9 billion), $25.1 billion of equity investments in public sector entities (including equity in NBN Co, Medibank Private and Australia Post), the Higher Education Loans Program receivable ($21.6 billion) and taxes receivable ($20.5 billion);
  • Non-financial assets totalling $110.1 billion, of which over 60 per cent is held by the Department of Defence and related agencies; and
  • Liabilities totalling $563.5 billion, including $285.7 billion of Commonwealth Government Securities and $193.3 billion in public sector superannuation liabilities.

While the balance sheet provides a relatively detailed position of the Commonwealth Government, it is far from comprehensive. For example, it does not reflect the Commonwealth’s biggest potential asset which is the power to tax and it does not include future age pension obligations as a liability.

State balance sheets

The State and Territory and local government balance sheets have a greater proportion of non-financial assets to financial assets than the Commonwealth. The aggregate State and Territory and local general government sector non-financial assets in 2009-10 were $1,184 billion compared with financial assets of $216 billion (Australian Bureau of Statistics, 2012).

For the States, non-financial assets include physical assets such as public schools, hospitals, police stations and roads. For example, in NSW non-financial assets were in the order of $146 billion at 30 June 2013 (NSW Government, 2013).

Balance sheet metrics

There is a range of metrics that can be used to measure balance sheet strength, each of which provides information which is useful for different purposes.

  • Net debt (interest bearing liabilities less a pool of liquid financial assets) is the most common metric used for international comparisons.
  • Net worth (total assets minus total liabilities) is the broadest measure of balance sheet strength.
  • Net financial worth (financial assets minus total liabilities) is considered a better measure of fiscal sustainability than net worth as it captures government’s ability to withstand adverse economic shocks by drawing on its stock of liquid assets to finance its budget.
  • Gross debt, commonly referred to as Commonwealth Government Securities (CGS) on issue.

Trends

Data from the 2013-14 Mid-Year Economic and Fiscal Outlook shows that each of these measures of balance sheet strength has deteriorated considerably since 2007-08.

Net debt has grown by $198 billion since 2007-08 and at the end of 2012-13 was 10 per cent of GDP, as seen in Chart 6.2.1.

 

Chart 6.2.1: Net debt

This chart shows the change in net debt between 2002-03 and 2012-13.

Source: Australian Government, 2013b.

Since 2007-08, net worth and net financial worth have fallen by $274 billion and $298 billion respectively and are now -13.3 and -20.5 per cent of GDP, as seen in Charts 6.2.2 and 6.2.3 respectively.

Chart 6.2.2: Net worth

This chart shows the change in net worth between 2002-03 and 2012-13.

Source: Australian Government, 2013b.

Chart 6.2.3: Net financial worth

This chart shows the change in net financial worth between 2002-03 and 2012-13.

Source: Australian Government, 2013b.

The main driver of the deterioration of these metrics has been the increase in gross debt via the issuance of Commonwealth Government Securities (CGS) (up by $227 billion since 2007-08) to fund government deficits, as seen in Chart 6.2.4.

The face value of CGS on issue has increased by $202 billion since 2007-08.

 

Chart 6.2.4: Commonwealth Government Securities on issue – face value

This chart shows the change in Commonwealth Government Securities between 2002-03 and 2012-13.

Source: Australian Government, 2013b.

The annual interest expense on CGS has increased from $3.8 billion in 2007-08 to $11.8 billion in 2012-13. Growth in the public sector superannuation liability (of $93 billion) has also contributed to the fall in net worth and net financial worth.

These increases in liabilities have been partially offset by increases in a number of financial assets including the Future Fund (by $24.7 billion), Higher Education Loans Program receivable (by $11.0 billion) and investments in Residential Mortgage-Backed Securities (by $9.1 billion).

Issues

Increasing awareness of the importance of the balance sheet

In assessing the government’s fiscal position most attention is on the annual ‘bottom line’ as measured by the underlying cash balance which broadly reflects annual receipts and payments. However, such a focus largely ignores the state of the Commonwealth balance sheet.

It is the position of the Commonwealth balance sheet which heavily influences credit ratings and borrowing costs and is an important indicator of short term fiscal sustainability and the government’s ability to respond to economic shocks. The balance sheet also reflects the debt that must be repaid by future generations.

A greater focus on the balance sheet position would be beneficial as it would broaden the debate on fiscal policy which can often be fixated on which year a government returns to surplus. For instance, following the global financial crisis, the primary public focus was on whether the Budget would return to surplus while substantial increases in deficits in 2010‑11 and 2011-12 which resulted in significant balance sheet deterioration were largely overlooked.

A balance sheet focus would also provide impetus for better asset management. The responsibility for managing assets and liabilities on a day-to-day basis has been devolved to individual agencies. Sound asset management suggests that governments should adequately fund upfront expenditure on assets, but also ongoing capital maintenance and operating costs.

Unfunded superannuation liabilities

A significant pressure on the balance sheet is unfunded superannuation liabilities for public servants and military personnel, as seen in Chart 6.2.5 below.

 

Chart 6.2.5: Unfunded Commonwealth Government superannuation liability projections

This chart shows the increase in the Commonwealth unfunded superannuation liability from 2013-14 and 2049-50, with most of the new growth occurring in the military schemes.

Source: Department of Finance.

The Commonwealth’s defined benefit superannuation schemes – which broadly pay a set benefit as a proportion of salary on retirement – are not funded over time through regular contributions from government. They represent a major pressure on future budgets with projected liabilities totalling $353 billion by 2050.

Accumulation schemes, on the other hand, provision over time an appropriate level of funding ready for when retirement benefits need to be paid by the Commonwealth. The Commonwealth’s defined benefit schemes include the Commonwealth Superannuation Scheme, the Parliamentary Contributory Superannuation Scheme and the Public Sector Superannuation Scheme. These schemes closed to new members in 1990, 2004 and 2005 respectively (Department of Finance and Deregulation, 2008a, b, and c).

The Military Superannuation and Benefits Scheme (MSBS) remains open to new members, and is projected to be the main driver of the Commonwealth’s superannuation liability growth from 2030.

Budgeting for loans and guarantees

The Commonwealth Government has taken on risks including:

  • guaranteeing approximately $13.1 billion of aged care accommodation bonds in case of provider default;
  • guaranteeing $3.3 billion of liabilities taken on by the Export Finance and Insurance Corporation;
  • guaranteeing nearly $700 billion of deposits as part of the Financial Claims Scheme;
  • guaranteeing large deposits and wholesale funding for banks during the global financial crisis, with $40.7 billion of funds still under guarantee;
  • committing $10 billion to the Clean Energy Finance Corporation (in the process of being abolished) for concessional loans and guarantees to clean energy companies; and
  • providing $420 million of concessional loans to financially stressed farmers (Australian Government, 2013c).

The information provided to ministers and released publicly on these risks is relatively poor. The Statement of Risks in Budget Paper 1 identifies a list of guarantees provided by the Commonwealth, but has no information on the likelihood of risks materialising and the expected cost of the risks.

The Commonwealth Government does not recognise the expected cost of these risks within the Budget estimates and these issues are not budgeted for on an equivalent basis to the costs of more direct forms of government expenditure, like grants. This may create perverse incentives as loans and guarantees appear to be ‘costless’ forms of assistance. For example:

  • If the government provides a $200 million grant to a firm, then this will reduce the underlying cash balance by $200 million in the year the grant is made.
  • Alternatively, the government could provide a loan guarantee to a risky firm, of $200 million value to the firm (through the lower borrowing costs it receives). Despite the risk that the government will have to pay out (and the value provided to the firm) this will have no impact on the ‘bottom line’ (underlying cash balance).
  • In contrast, the government underlying cash balance could actually improve by providing concessional loans to a risky firm. Assuming a government cost of borrowing of 4 per cent, a $1 billion loan over 10 years at a concessional lending rate of 5 per cent would improve the bottom line by around $30 million over 4 years.

Because the risk taken on by government is not properly priced, guarantees and loans are not subject to an effective budget constraint - they cannot be prioritised or ranked against other direct forms of expenditure. The government has an incentive to take on more risk on behalf of taxpayers than is optimal.

The risk to taxpayers can eventually materialise, as was the case with the support from the Commonwealth and Queensland Governments for the Australian Magnesium Corporation, see Box 6.2.1.

Box 6.2.1: Guarantee to Australian Magnesium Corporation

In 2001, the Australian Magnesium Corporation planned to establish the world's largest magnesium smelter at Stanwell, near Rockhampton in central Queensland. The project attempted to commercialise the light metal for use in the motor industry, but had difficulty raising capital from the private sector given uncertainty about the viability of the technology.

To assist the $1.3 billion project, the Queensland Government provided a $100 million loan facility to support dividend payments to AMC shareholders, as well as $50 million in infrastructure. In addition, the Commonwealth Government agreed to support AMC by providing a loan guarantee of up to $100 million, and a $50 million contribution through the CSIRO to assist with commercialising the technology.

The loan guarantee appeared to be 'costless' as it had no impact on the Budget estimates, but was noted in the Budget's Statement of Risks.

Construction costs on the project soon escalated. Work on the project eventually ceased in June 2003 due to significant project cost overruns. As a consequence, as reported in the 2004-05 Budget, the Commonwealth finalised its obligation to pay $90 million to the ANZ Bank who provided the original loan to AMC.

Experience from overseas suggests that guarantees and loans are used more frequently during times of fiscal adjustment, where governments want to provide assistance and subsidies without immediate impacts on their reported budget position (Polackova‑Brixi, 2012).

Budgeting for equity investments

The government also takes on the risks of ownership and receives returns (such as dividends) on behalf of taxpayers when it partially or fully holds equity. Equity holdings in major government-owned entities include the NBN Co, Medibank Private, and Australia Post.

Under the relevant accounting standards, the Commonwealth’s equity investments are classified as having no impact on the Budget ‘bottom line’ (the underlying cash balance) if the Commonwealth has a reasonable expectation of a real return on its investment, at least covering the effects of inflation.

For example, the Budget papers record the financing of NBN Co as an equity investment on the basis that the Commonwealth expects to recover its outlay in real terms over time through dividends and sale of the investment in the future.

This ‘hurdle rate of return’ (of around 3 per cent) for the Commonwealth is well below that required by a private investor, who would require a rate of return commensurate with the cost of borrowing and the level of risk undertaken. Under the Government Financial Statistics accounting standards there is no requirement for NBN Co or other equity investments to achieve a risk-adjusted rate of return.

Without commercial discipline, government-owned entities may accept a lower rate of return and take on more risk. While the Commonwealth has a low cost of borrowing by virtue of its ability to source debt and raise tax, this does not mitigate the risks taken on by taxpayers from the financing of risky projects.

The annual change in the value of equity in NBN Co is presented as an ‘other economic flow’ on the balance sheet. Should the Commonwealth no longer expect to receive a real return on its investment, any future equity injections would be classified, partially or fully, as grants. This would worsen the underlying cash balance.

Potential areas for reform

Military superannuation

The 2007 Report of the Review into Military Superannuation Arrangements recommended there be no change to the indexation arrangements for the current Military Superannuation and Benefits Scheme (MSBS), that the scheme be closed to new entrants and a new scheme based on an accumulation plan be opened for new Australian Defence Force members (Department of Defence, 2007).

The MSBS is now the only major Commonwealth scheme with unfunded defined benefits that remains open to new members, and hence is generating uncapped and increasing unfunded liabilities. These liabilities will have to be paid for by future generations.

Closure of the Military Superannuation and Benefits Scheme would result in a significant reduction in the Commonwealth’s unfunded superannuation liability potentially in the order of $200 million by 2020. This would make a significant contribution to repairing the long-term financial position of the Commonwealth.

If the Military Superannuation and Benefits Scheme is closed and replaced by a funded accumulation scheme for new military personnel there will be an immediate increase in cash costs to the government. This arises because employer contributions would be made directly to the new scheme in respect of members’ services, whereas the reduction in outlays from closing the Military Superannuation and Benefits Scheme arises from reduced pension payments in the longer term.

While the increased cash outlays would depend on the design of the new scheme, including the contribution rate and death and disability arrangements, these outlays over the first four years could be around $400 million.

The Commission considers that any new scheme should continue to provide a defined benefit where a member dies or is medically discharged from the Australian Defence Force. This death and disability element will complement the accumulation component of the scheme.

As outlined above, the Future Fund was established to enhance the ability of the Commonwealth to discharge annual unfunded superannuation benefit payments. Under current arrangements, draw downs from the Future Fund will occur from 2020.

The effect of drawing down pension payments, from that date, will be a decrease in cash costs to government, as payments will be met from Future Fund monies rather than consolidated revenue.

In relation to civilian superannuation, while the major defined benefit schemes have been closed already, the liability will nevertheless continue to grow until around 2030 in relation to existing contributors.

The Government should move over time to a ‘funded model’ for those defined benefit schemes, where the employer contribution would be paid to the superannuation trustee, the Commonwealth Superannuation Corporation, rather than retained in consolidated revenue. This is the arrangement the Commonwealth currently follows for its accumulation schemes.

This would reduce the Commonwealth’s medium to long-term liability, but lead to a significant and immediate increase in cash costs to government, estimated to be around $1.4 billion per year. If this option is also implemented in relation to the existing military scheme, this cost would rise to about $2.7 billion per year.

Given the significant cash impacts of funding future employer contributions, the Commission recommends that the Government consider allowing earlier drawdowns from the Future Fund (that is, before 2020). This would effectively offset, in part or full, the costs of a new accumulation scheme and, or the funding of defined benefit schemes in future years. This action would require amendments to the Future Fund Act.

Military superannuation is also discussed in Section 9.8 of the Appendix.

Guarantees and loans

The Government could ensure that the expected costs of guarantees and loans are budgeted for within the forward estimates on both a cash and accrual basis. This would ensure that the best estimate of the costs of these transactions is reflected in estimates and improves incentives for decision-making.

A similar approach was adopted by United States through the Federal Credit Reform Act of 1990 following the failure of numerous government guaranteed Savings & Loan Associations.

The Government could also improve transparency by reporting on the likelihood of risks materialising and the expected cost of guarantees and loans. For example, the Chilean Government discloses information on the costs of revenue and exchange rate guarantees over time. It also publishes estimates of the amounts it expects to pay on such guarantees over the next 20 years and the value of each of these guarantees.

Guarantees

Rather than only reporting information in the Statement of Risks, the Commonwealth could explicitly include the expected cost of existing and new guarantees and loans in the Budget ‘bottom line’, where these can be reliably estimated.

The expected cost of the guarantee should accurately reflect the expected cost to the Commonwealth of taking on the associated risk. A provision in the Contingency Reserve should be made in the Budget year and forward estimates to account for the expected cost of the guarantee. The Contingency Reserve is explained in Box 6.2.2 below.

Where these costs materialise they would be accounted for within the appropriate agency and the amount in the contingency reserve would be removed. Where these costs do not eventuate the amount would be removed from the Contingency Reserve.

Moving to expected-cost pricing of guarantees may take some time to phase in, as there are methodological difficulties in developing an accurate estimate of expected costs.

For the purposes of the actuals published in the Final Budget Outcome and Consolidated Financial Statements there would be no change from current practice. A worked example of the proposed treatment for guarantees is at Attachment 6.2.1.

Box 6.2.2: The Contingency Reserve
The Contingency Reserve is an allowance used to ensure that the Budget estimates are based on the best information available at the time of the Budget and to ensure that aggregate estimates are as close as possible to expected outcomes.

Amongst other things it includes a provision for events and pressures that are reasonably expected to affect the Budget estimates.

Under this proposed treatment, the risk taken on by taxpayers would be more appropriately incorporated in the Budget papers. This would create better incentives for ministerial decision making as:

  • there is an immediate impact on the ‘bottom line’ (UCB); and
  • the proposing minister could be required to offset the cost; or preferably, charge the beneficiary for the risk taken on by the Commonwealth.

Budgeting for the expected costs of such transactions would not only improve the incentives that the government faces as a whole, but also the incentives faced by individual ministers. This is because ministers could now be required to offset the expected costs with savings from their portfolio. This would require ministers to consider whether such a guarantee or loan is a better use of government resources than other portfolio priorities.

Loans

For loans, it is proposed that the Commonwealth recognise the expected amount of loan defaults as a grant at the time of the loan origination.

There is already scope under existing accounting standards to recognise the expected costs of loan defaults as a grant in the Budget estimates.

It appears that in practice these expected defaults are only sometimes included in the Budget estimates as a grant. For instance, no expected default costs are currently recognised for the Clean Energy Finance Corporation and Farm Finance loan packages. It is recommended that the Government include provision for such costs.

In a similar way to the inclusion of expected costs of guarantees in the Budget process, the proposing minister could be required to offset the cost or charge the beneficiary for the risk taken on by the Commonwealth.

A worked example of the proposed treatment of loans is at Attachment 6.2.1.

Existing guarantees

Applying the recommended treatment to all existing guarantees would ensure that the Budget estimates most accurately reflect the expected costs of significant commitments such as the Financial Claims Scheme and Aged Care Accommodation Bonds Guarantee. It would also allow governments to consider the benefits of such guarantees in light of their expected costs.

Although there is a large aggregate value of quantifiable guarantees and other contingent liabilities, in the order of $900 billion in 2013-14 in the Budget Statement of Risks, only the expected cost to the Commonwealth associated with those guarantees would be booked to the ‘bottom line’. The expected cost takes into account the probability that the guarantee will be called upon.

An example of a Commonwealth Government guarantee with a large value ($13.1 billion in 2013-14) with a comparatively low expected cost ($25 million since 2006-07) is the Aged Care Accommodation Bonds Guarantee (see Box 6.2.3).

Box 6.2.3: Aged care accommodation bonds guarantee
Under the accommodation bonds guarantee scheme, the Commonwealth Government guarantees the repayment of the bond balance if the aged care provider becomes bankrupt or insolvent and is unable to repay the outstanding bond balance. In return, the residents' rights to pursue the provider to recover the accommodation bond money transfers to the government.

In the event that the government cannot recover the full amount from the defaulting provider, it may levy all providers holding accommodation bonds to recoup the shortfall. Since the scheme was introduced in 2006-07, it has been activated five times requiring payment of $25 million. The government has not sought to levy providers to recover this cost.

Including the expected costs of guarantees and loans in the Budget estimates is likely to result in a deterioration in the Budget bottom line in the short term. However, this will provide a more accurate reflection of the costs which the government expects to face.

Companies are likewise required to provision for bad and doubtful debt. Often this is set on a historical basis, relying on previous outcomes.

The proposed changes would affect the estimated Budget bottom line in future years, but would not affect the final Budget position in any year; as actual costs of guarantees and loans would continue to be reflected in each Final Budget Outcome.

The valuation of expected costs of guarantees and loans requires particular expertise. It may be prudent for all estimates to be conducted by a single team within the Department of Finance to ensure the necessary capacity building and consistency amongst estimates of different guarantees and loans. This team could draw on the expertise of the Australian Office of Financial Management, the Australian Government Actuary and private sector firms.

For administrative simplicity, a minimum threshold could be applied, so that expected costs of new risks are only included in the Budget where they can be reliably estimated and have an expected cost above $5 million in any year.

The expected costs of guarantees may fluctuate over time as market conditions affecting the recipient entities change. The expected costs would therefore need to be re-estimated over time, with revisions affecting the Budget estimates.

This process of revising estimates already occurs for many other items in the Commonwealth Budget, such as forecasts of revenue and the costs of demand-driven programmes.

The recommended treatment for guarantees should also be applied to other contingent liabilities including indemnities and risk of litigation, where they can be reliably estimated.

Equity investments

As outlined in Section 10.1 of the Phase One Report, the Commission considers that Commonwealth bodies that operate and compete in contestable markets should be considered for their privatisation potential.

Similarly, as a matter of policy, the Commonwealth should not take equity positions where the activity can be undertaken by the private sector, with private investors risking their own money rather than taxpayers.

Where the Commonwealth does take an equity position it should disclose the rate of return it expects to receive and how this compares to the risk-adjusted rate of return that a private investor would need to make the same investment.

To ensure transparency around the government’s equity holdings, the Commonwealth should continue to regularly and independently re-assess the value of its equity investments.

Box 6.2.4 provides notes for implementation of the Commission’s recommendations on guarantees and loans.

Box 6.2.4: Implementation notes – guarantees and loans

The expected costs of new and existing guarantees and loans should be budgeted for within the estimates on both a cash and accrual basis:

  1. The expected costs of all existing and new Commonwealth guarantees should be included in the Contingency Reserve for the Budget year and forward estimates (and removed from the Contingency Reserve at the end of each year if the costs have not eventuated).
  2. The expected non-repayment amount of loans should be recognised as a grant at the time of the loan origination.
  3. These expected costs should be included in the Budget where they can be reliably estimated and have an expected cost above $5 million in any year.

As the expected cost of guarantees and loans would now impact on the underlying cash balance, the cost could be:

  1. borne at the whole-of-government level; or
  2. be offset by the proposing portfolio minister; or
  3. be charged in full or part to the beneficiary.

The Government should apply this treatment to other contingent liability items such as indemnities, where they can be reliably estimated.

References

Australian Bureau of Statistics 2012, Year Book Australia, cat. no. 1301, Canberra.

Australian Government 2013a, Final Budget Outcome 2012-13, Canberra.

Australian Government 2013b, Mid-Year Economic and Fiscal Outlook 2013-14, Canberra.

Australian Government 2013c, Budget 2013-14, Canberra.

Department of Defence 2007, Report of the Review of Military Superannuation Arrangements, (Podger Review), Canberra.

Department of Finance and Deregulation 2008a, Commonwealth Superannuation Scheme (CSS), Canberra.

Department of Finance and Deregulation 2008b, Parliamentary Superannuation Act 2004, Canberra.

Department of Finance and Deregulation 2008c, Public Sector Superannuation Scheme (PSS), Canberra.

New South Wales Government 2013, Monthly Statement 2012-2013, NSW Government, Sydney.

Polackova-Brixi, H 2012, Avoiding Fiscal Crisis: Accounting for Contingent Liabilities to Manage Fiscal Risk, World Economics, Vol. 13, No.1.