An accounting system in which expenses are recognised when goods or services are used in earning revenue; and revenue is recognised in the period in which output is sold.

Budget Sector

That part of government which is dependent upon the budget (i.e. upon taxes) to finance its operations. Contrasts with the public enterprise sectors, which are self-financing from commercial revenues. (Properly speaking, the correct government accounting terminology is 'general government' sector. The term 'budget sector' is substituted in this paper to aid comprehension.)

Capital Expenditure

Expenditure which generates a flow of direct future benefits.

Cash Accounting

A method of accounting under which revenues and expenses are recorded only when cash is received or when cash is paid out.

Commercial Assets

An asset the costs of which are entirely met through the imposition of user charges (i.e. through payments by consumers), and which does not therefore require tax finance.

Cost-Based 'Valuation'

The practice of recording assets such as plant and equipment in a balance sheet at a 'valuation' which is based upon the cost of the asset minus depreciation to date. 'Cost' in this context has a number of possible meanings, one of which is historical cost (see below). Measures of asset 'valuation' which adjust historical cost for inflation (such as in so-called Constant Purchasing Power Accounting) are also versions of cost-based 'valuation'.

Current Expenditure

Expenditure which gives rise solely to benefits in the financial reporting period in which it takes place, or which at a minimum generates no flow of direct future benefits.

Depreciation

An allocation made, as an expense, of the cost or other value of an asset, in each financial reporting period over the life of the asset. The standard meaning of the term, consistent with cost-based asset 'valuation', is the allocation of the cost of the asset over time. (Advocates of 'economic' valuation (see below) substitute for this the alternative concept of depreciation as the reduction in each time period of the economic value of the asset.)

Economic Valuation

Working out what an asset or business is worth either (1) based upon the income or service potential it is expected to generate in the future for its owner, or (2) the price the asset or business would achieve if sold, which reflects the service potential of the asset to someone else.

Fiscal Sustainability

The (political) feasibility of government continuing in the future to meet its financial commitments.

Historical Cost

The use in cost-based valuation of the original cost of an asset, without any adjustment to take inflation into account. Contrasts with alternative approaches which attempt to adjust for inflation through such means as marking up the historical cost according to a index of general inflation (Constant Purchasing Power Accounting).

Intergenerational Equity

The principle that taxpayer contributions to the costs of 'social' assets should be spread over time in accordance with the manner in which benefits from such assets accrue.

Net Worth

The value of all assets held by an organisation minus the value of all its liabilities.

Social Assets

An asset the benefits of which are made available to citizens without charge, and which is therefore ultimately tax-financed.

Solvency

The capacity of an entity to meet the interest costs and repayment schedules associated with its long-term financial obligations.

Major Issues

If everyone were an expert accountant with all the relevant information at their fingertips, the accounting practices used by corporations and by government to report their financial results might not matter much. Because, however, both expertise and information tend to be in short supply, the impressions created by the choice of specific accounting policies matter enormously. Given that individual voters have little incentive to go to the trouble of digging below the surface of government accounting, the impressions created by public sector accounting can have considerable political impact.

Traditionally, the public sector has employed what is known as 'cash' accounting. There is little dispute that cash accounting can promote serious false impressions about the overall financial position of the public sector. Governments are aware of these false impressions and at times actively exploit them.

Cash accounting uses the level of public debt to indicate the state of government finances at any point in time. The standard cash accounting deficit basically indicates the change in debt between two points in time. The trouble with debt as a measure of the financial position of government is that it ignores a whole range of other assets and liabilities - including fixed assets, land, and liabilities such as public service superannuation - which are an important part of the picture of government finances. Focusing on debt in isolation without taking these other assets and liabilities into account creates a variety of false impressions. One of the best known of these is the assumption, often made by the uninformed, that when an asset is sold the impact of that sale upon the financial position of government can be measured by the reduction in the deficit and debt produced by the proceeds of the asset sale. (This is, of course, a view which overlooks the loss of the benefits which the asset would have generated had it remained in public hands). A related fiscal illusion is that all 'blowouts' in the budget deficit are equally bad, irrespective of whether they result from, say, major unfunded tax cuts or, alternatively, a major enhancement of the country's transport and communications infrastructure. (This overlooks the fact that in the one case physical assets are being created, whereas in the other case there are no additions to the stock of assets).

There is a strong school of thought which holds that the solution to this problem is to be found in using 'accrual' accounting to report the financial position of the public sector. This involves replacing debt as the fundamental indicator of the financial position of government with what is known as 'net worth'. Net worth is the value of all assets minus the value of all liabilities. Within this accounting framework, a new type of deficit measure would come into play, which would indicate changes in net worth rather than changes in debt alone.

The implementation of versions of accrual accounting is well on the way in Australia and in some overseas countries. For example, both New South Wales and New Zealand report the overall financial position of government on an accrual basis. The Commonwealth has been moving in this direction and the National Commission of Audit has recommended the adoption of full accrual accounting by December 1997. The Australian Bureau of Statistics has also announced a timetable for implementing accrual presentation of its Government Finance Statistics series.

Accrual accounting enthusiasts see in these developments the solution to the fiscal illusions which can be generated by cash accounting. They note that the private sector employs accrual accounting, and that if private corporations were to use government-style accounting the results would be truly bizarre. Accrual accounting, in their view, offers vastly superior indicators of the way in which past and present fiscal policies will impact upon future governments and taxpayers. They suggest, for example, that if an asset were sold, accrual accounts would show up the impact of this upon net worth, which would automatically take into account the loss of the asset as well as debt reduction from the asset sale proceeds. Similarly, the difference between capital and current expenditure would be brought out clearly, because the assets created by capital expenditure would be highlighted in the net worth measure and in the corresponding deficit measure.

Are these expectations of accrual accounting advocates realistic? Concretely, can net worth deliver the results expected of it as a 'headline' measure of the financial position of government which allegedly transcends the limitations of traditional measures? Can it provide an objective measure free of - or at least much less prone to - accounting manipulations which mislead the electorate and perhaps even the parliament?

These are crucial questions which have not received anything like the attention they deserve. It has all too often been considered sufficient to contrast the familiar failings of real-world cash accounting with the near perfection of an idealised version of accrual accounting. This approach is, however, unrealistic and indefensible. It is an approach which fails adequately to acknowledge the great practical significance of 'technical' controversies, including issues about which a range of possible versions of the net worth concept may be appropriate. It also fails to accord due recognition to the severity of the practical problems which confront attempts to measure net worth, or to the policy implications of such measurement problems.

This paper aims to restore balance to this issue by subjecting the concept - or rather concepts - of net worth to a searching analysis. Regrettably, there is no way of doing this without pursuing issues which may at first sight seem unnecessarily technical. Considerable effort has, however, been made to present these issues as accessibly as possible.

Even the limited experience to date has made it very clear that the definition, meaning and interpretation of public sector net worth is no mere technical matter. Rather, it can be a matter of great political importance in the evolution of fiscal policy. In New Zealand a couple of years ago, the presentation to the public of a balance sheet purporting to show substantial negative public sector net worth played a very important role in persuading the New Zealand electorate to accept a fundamental shift towards tighter fiscal policy and smaller government. This negative net worth was widely seen in New Zealand as evidence that the government was in some sense insolvent. On the other side of the coin, the former Victorian Government attempted in the early 1990s to make use of newly-developed public sector balance sheets to argue, in defiance of reality, that the State did not face a debt problem.

Both the New Zealand and Victorian examples highlight the important fact that measures of net worth may vary greatly depending upon technical choices as to accounting methodology. Critics of the New Zealand balance sheet pointed out, with justice, that a failure to take inflation into account in valuing many assets had the effect of artificially depressing measured net worth. In Victoria, the great variability of net worth measures was graphically demonstrated when, in 1993, the Commission of Audit appointed by the incoming Kennett Government produced a revised State balance sheet which increased the estimate of liabilities, and reduced the estimate of the value of assets, and which was then interpreted politically as providing evidence that the financial position of the State was in a far more parlous condition than had hitherto been revealed.

In reality, the net worth concept cannot even in theory meet all of the claims made for it. To get some understanding of why this is so, it is useful to consider the nature and role of a balance sheet in the annual accounts of a private business. Typically, the fixed assets (plant, equipment etc) of such a business will be 'valued' in the balance sheet at cost minus 'depreciation' to date. Corresponding to this is the recording each year of a depreciation charge in the business' 'profit and loss' statement, with depreciation simply representing an appropriate portion of the cost of those fixed assets. (Those with an accounting background should note that 'cost' in this context may mean either 'historical' cost - the original cost of the asset - or one of the forms of modified historical cost which adjust historical cost for inflation).

The key thing about such a balance sheet is that the 'valuations' of fixed assets upon which it is based bear no necessary relationship to what those assets are really worth, either to the business or to some other potential owner. To know what an asset - or, rather, a business - is really worth involves a separate process in which the focus is not upon what assets might have cost, but rather upon the future profit expectations of the asset or business.

What is implicit in this is that there are two fundamentally different valuation concepts. The first is cost-based valuation. The second, based upon profit expectations, is 'economic' valuation. Corresponding to each of these is, in a sense, a different net worth measure (or class of measures), derived by bringing these asset valuations together with valuations of liabilities and other assets. Whereas the function of a cost-based balance sheet is to permit the estimation of annual profit, the function of a net worth measure based upon 'economic' asset valuation is, by contrast, to tell one what the company is worth. Economic valuation also measures corporate 'solvency' - that is, whether a business is likely to be able to meet its financial commitments - whereas the cost-based balance sheet in the business' annual accounts tells one absolutely nothing about solvency. This is not to criticise the typical balance sheet, but simply to point out that different valuation concepts serve different purposes. An additional strength of the 'cost-based' asset valuation in a typical private sector balance sheet is that the information it provides is inherently much less subjective than any estimate of economic value can be. Because economic valuation depends upon expectations of future profits, the estimation of economic value is a highly subjective exercise, reliant upon 'guesstimates' and very prone to manipulation. This is demonstrated by the significantly different valuations of companies typically made by the opposing camps and other analysts at the time of a corporate takeover battle.

Which type of approach to valuation, and thus to the measurement of net worth, might be relevant to the public sector?

The cost-based approach certainly has its merits. It is, for a start, the natural approach if one is interested in what is known as 'intergenerational equity'. Intergenerational equity is essentially concerned with the equitable sharing over time of the costs of long-lived assets. It therefore links naturally to the idea of charging part of the costs of such assets to the budget each year as depreciation, and thence to the idea of a balance sheet based upon the 'valuation' of assets at depreciated cost. Another related argument for the cost-based approach is that it facilitates a good accounting measure of what is referred to as 'net' savings by government - that is, of government savings minus depreciation.

What, on the other hand, about a net worth measure based upon economic valuation? Here the story is rather more complex. For instance, it is not possible even in principle to design any version of the net worth concept which measures the 'solvency' of government, for the simple reason that government depends upon its power to tax in order to meet its liabilities, and no net worth measure can factor in future taxation. Beyond this, there are a variety of other reasons to doubt even the theoretical value of an 'economic' measure of public sector net worth. And this is before one even considers the practical measurement problems which confront any effort to measure 'economic' net worth - measurement problems which are of an order of magnitude greater even than those which confront the measurement of economic net worth in the private sector.

For reasons which are not fully clear, the main accounting policies which have been developed to guide the development of public sector net worth measures in principle favour an economic approach to valuation, while in practice having extensive recourse to cost-based valuation as a fallback when (as is often the case) economic valuation is impracticable. Not only is this highly questionable at a theoretical level, but it has the effect of creating considerable scope for eclectic mixes of valuation methodologies. The inevitable result is net worth measures which are not only of highly questionable meaning, but which are highly susceptible to accounting manipulation - perhaps even more than traditional cash accounts are.

Regrettably, all this means that recent 'progress' in the development of accrual accounts by State and Commonwealth governments may be essentially a futile exercise, and an expensive one at that, given the high administrative costs involved with accrual accounting.

This is not to say that the development of an appropriate form of accrual accounting for the public sector is necessarily an impossible task. Its feasibility remains to be seen. Clearly, however, if it is going to be possible to develop a conceptually appropriate and practicable measure of public sector 'net worth' (if, in fact, that is what such a measure should be called), it will only be by going back to the drawing board.

It also needs to be borne in mind that, for all the fundamental limitations of cash accounting techniques, the inadequacies of 'traditional' public sector accounting are nowhere near as severe as one might be led to believe by some critics. There are, for example, already in existence more sophisticated measures which distinguish capital and current accounts, and which at least partially adjust for potentially distorting factors such as asset sales and State repayments of debt to the Commonwealth. Some of the less problematic elements of accrual accounting are, moreover, increasingly being incorporated into so-called 'cash' accounting. A prime example of this is the presentation of broader 'debt' measures which include employee liabilities such as superannuation.

The implementation of accrual accounting has gone so far that it is probably here to stay. Moreover, it may be argued that there are managerial benefits to be derived at the departmental level from the operation even of a flawed accrual system. So the best approach may be to let the steps which have been implemented to date run on for a while, and use them to gather further valuable experience. It is, however, clear that, rather than 'advance' accrual accounting developments further along the current track, the Commonwealth should place a moratorium on further policy implementation in this area until such time as a defensible framework of accrual accounts is developed.

Introduction

In the public sector, as in the private sector, accounting policies and practice matter enormously. It is these policies and practices which determine the 'results' which a firm or government reports to the world. This paper discusses an accounting choice with enormous implications for the manner in which government reports its overall performance. The 'headline' financial performance measures generated by traditional public sector accounting are measures of debt and of changes in debt (deficits). Over at least the last decade, the adequacy of these measures has been questioned. A substantial school of opinion has argued that the deficiencies of these traditional measures are such that the only way forward is through the major conceptual leap from 'cash' accounting to full 'accrual' accounting. This substitutes net worth in the place of net debt as the fundamental 'stock' measure of the financial position of government. The 'flow' counterpart of this is the replacement of cash deficit measures with a bottom line measure in an accrual operating statement which measures changes in net worth. The net worth measure requires the existence of a public sector balance sheet.

Accrual accounting appears to be sweeping all before it. Most States and the Commonwealth have moved towards agency-level adoption. In Australia, New South Wales has taken the lead in the regular reporting of accrual concepts on a public sector wide basis. Overseas, the New Zealand Government is one which has moved strongly in the same direction. The reporting in 1993 of a negative public sector net worth of $7.7 billion in New Zealand was of great political importance in winning political support for a tough program of fiscal reform.

Notwithstanding the seemingly irresistible progress of accrual accounting, there is in fact not consensus support for these moves either within the accounting profession or within the public sector. There is a considerable body of opinion either opposed to, or having great reservations about, these changes. The matter is not assisted by the fact that highly reputable individuals and organisations are to be found on each side of the debate.

In addition to this disagreement about the merits of accrual accounting per se, there are also some quite major disagreements amongst proponents of accrual accounting concerning appropriate accounting methodologies. These are not mere 'technical' disagreements, but have important practical ramifications.

This issue is particularly important at the Commonwealth level at the present time because, having moved towards the implementation of accrual accounting at an agency level, the Commonwealth must determine whether to move its financial reporting for the budget sector and/or the public sector as a whole onto an accruals basis.

The objective of this paper is to clarify, in as non-technical a manner as possible, some of the key issues which arise in considering whether it is appropriate and useful to develop public sector net worth measures. 'Public sector' in this context means either the budget sector of government or the public sector more generally (i.e. incorporating public enterprises). The fundamental issue at stake is whether the development of a balance sheet and net worth measure for the public sector is likely to produce substantial improvements in the transparency and quality of fiscal policy or management, and if so whether these improvements come at an acceptable cost. The paper does not directly consider the merits of agency-specific balance sheets, although the relevant issues overlap. As is well known, public enterprises already generally use the accrual accounting framework, but government departments till recently have not.

Weaknesses in Traditional Public Sector Accounting

The case against traditional public sector accounting arises primarily from the desire to facilitate better financial decision-making through an accounting framework which

brings out accurately and clearly the implications for the future of fiscal decisions,
does so more adequately than does the accounting framework which has traditionally been used within the core 'budget sector' of government, and
in this way facilitates the application to fiscal decision-making of the relevant policy criteria.

There are two fundamental criteria against which to assess the implications of fiscal decisions for the future. These are intergenerational equity and fiscal sustainability.

Intergenerational equity principles are those which govern whether, and to what extent, it is viewed as equitable for government to incur financial liabilities. There are those who take the view that it is, as a matter of principle, inappropriate to create financial liabilities which future taxpayers will have to meet. The conventional formulation of intergenerational equity(1), however, rejects this notion on the grounds that if expenditures are undertaken which create benefits for future taxpayers it will also be reasonable that those future taxpayers make a commensurate financial contribution. The operative principle, from this standpoint, is one of matching costs and benefits over time.

Fiscal sustainability refers to the ability and willingness of citizens in the future to meet financial liabilities entered into by past governments. For financial liabilities to be sustainable requires not simply that it is arithmetically possible to impose taxes and charges sufficient to service those liabilities, but also that the electorate is prepared to acquiesce to such a level of taxes and charges without reacting in such a manner as to induce their government to choose to default on its liabilities.

Traditional public sector accounting confines its coverage of the future implications of fiscal decisions to debt. This includes debt owed by and to the public sector, the latter constituting monetary (or financial) assets. Since debt represents an obligation to pay interest in the future, and monetary assets an entitlement to receive interest in the future, one can subtract the latter from the former to obtain a net debt measure. This is, of course, a 'stock' measure, in that it measures the stock (quantity) at a point in time of a certain class of assets and liabilities. The 'flow' counterpart of this stock measure is the change in net debt over a period of time. This is, of course, referred to as a deficit if there is an increase in net debt, and a surplus if there is a decrease(2). It is as a reflection of this that the standard public sector budget statement is essentially a record of cash payments and receipts.

The use of debt and deficits as 'headline' measures of fiscal performance tends to promote the view that increases in debt leave the future worse off, and reductions in debt leave the future better off, irrespective of what is behind the change in the level of debt. The problem with this is that, on the one hand, conventional debt is not the only type of future financial encumbrance into which government enters and, on the other hand, interest earned by monetary assets held by government is not the only type of future benefit which accrues as a result of past fiscal actions.

On the liabilities side, the conventional debt measure leaves out important items such as accrued employee entitlements (superannuation, long service leave), pension and social security commitments, and future commitments under service contracts for the provision of 'contracted out' goods and services. This blind spot in conventional fiscal measures has made it easier for governments, for example, to store up troubles for the future through the run-up of big public sector superannuation commitments which imply a heavy peak in superannuation payments.

This is serious enough, but on the asset side there are equally serious problems as a consequence of a failure to take into account the existence of assets other than monetary assets. An asset is an economic resource which generates a stream of future benefits. The most obvious category of asset which is overlooked in the traditional net debt measure is fixed assets. The future benefits generated by such assets may be financial or non-financial, or a combination of both. The profits earned by government business enterprises are a good example of a financial benefit. An example of a non-financial benefit is the service ('consumer surplus' in economic jargon) provided to the community by ordinary (i.e. non-toll) roads and bridges.

The traditional net debt measure fails to register (other than in the case of monetary assets) the addition or diminution in future flows of financial and/or social benefits which occurs when assets are acquired or sold. Instead, all it registers is any increase in debt when an asset is acquired, and any reduction in debt when an asset is sold. This tends to encourage serious misapprehensions about the impact of capital expenditure and capital (i.e. asset sale) revenue upon fiscal sustainability and intergenerational equity.

Foremost amongst the misapprehensions this may promote is the assumption that foregoing capital expenditure, or entering arrangements under which the private sector undertakes the provision of 'public' infrastructure, makes for a better fiscal future because it reduces debt or keeps it lower than it would otherwise be. It is for this reason that traditional public sector accounting has often been considered to create a bias against capital spending.

A closely related illusion is that the use of privatisation to reduce the deficit (and, consequently, debt) improves fiscal sustainability. It is unquestionably a mistake to make this assumption without considering the future financial returns foregone as a consequence of the sale of the public enterprise concerned. Yet it is precisely the type of crude thinking on the part of the uninformed which is encouraged by a narrow focus upon debt. And the unfortunate reality is that governments of all persuasions are not above exploiting popular misapprehensions in order to 'sell' their policy agendas.

The Broad Case for a Public Sector Net Worth Measure

It is in the light of this critique of traditional public sector accounting that the case for the development of a public sector balance sheet arises. A balance sheet is a statement which seeks to identify and value all liabilities and assets. Net worth is the 'bottom line' of a balance sheet, obtained by subtracting the aggregate value of liabilities from the aggregate value of assets.

Like the net debt measure, net worth is a 'stock' measure. The corresponding 'flow' measure, representing the change in net worth within the accounting period, is referred to as an 'operating deficit/surplus'. Just as the conventional deficit is the bottom line of a traditional 'cash' budget statement recording cash outlays and receipts, so the operating deficit/surplus is the bottom line of an 'accrual' operating statement which instead records 'expenses' and revenues.

The contrast between a cash outlay and an expense is exemplified by the case of a fixed asset. In cash accounting, the cost of the asset appears as an outlay at the time the asset is paid for. This is not the case for accrual accounting. Instead, depreciation of the asset concerned is entered as an expense during each period in the life of the asset. Depreciation reduces the recorded 'value' of the asset - hence the link between the operating statement and the balance sheet. Another example illustrating the difference between expenses and cash outlays is superannuation. Cash accounting records superannuation payments when they are actually made to retired public servants. Accrual accounting treats as an expense in each time period the accumulation of superannuation entitlements by public servants during that time period. The accrual principle, in short, is that expenses are recognised in the period to which they relate, which is not necessarily the period in which the relevant cash payments takes place.

Proponents of public sector balance sheets believe that net worth provides the best summary measure of the overall 'financial position' of the public sector (e.g. AARF, 1995). The belief is that by explicitly factoring in the assets and liabilities which traditional public sector accounting does not cover, the full implications of fiscal decisions for the future will become more readily evident and fiscal illusions and misrepresentations considerably reduced.

Concretely, it has been claimed (e.g. Rowles, 1992; AARF, 1995; JCPA, 1995a; Mellor, 1996) in support of the development of a public sector balance sheet that

net worth provides the best measure of fiscal sustainability,
net worth provides a means of assessing the compatibility of fiscal policy with intergenerational equity principles,
the change in net worth (or, to be more precise, an adjusted variant of this) is a superior measure of the government's contribution to national savings because it measures net savings (i.e. savings after depreciation) rather than gross savings,
a focus upon net worth eliminates fiscal illusions associated with asset sales,
the same applies to a whole host of analogous transactions used by governments these days to raise money or fund service provision through the alienation of future revenue flows and/or through commitments to make future capital usage payments in a form other than interest (for an outline of the range of transactions concerned see, e.g., Robinson, 1996a),
conversely, the net worth focus encourages the sale of public assets (as opposed to passively holding those assets out of habit) where it is genuinely beneficial to do so,
it removes incentives for governments to run up unfunded superannuation and similar liabilities in order to give themselves more short run fiscal freedom,
the valuation of the physical asset stock in a balance sheet removes the anti-capital expenditure bias of traditional cash accounting,
the valuation of physical assets also facilitates assessments of the adequacy of the capital stock, and draws attention to future replacement costs, and
it removes the illusion, arising from the failure of traditional accounting to recognise the future implications of reduced service potential, that deferring maintenance of fixed assets across government somehow improves the budgetary position.

These are impressive claims. But before being carried away by the breadth of these claims and the force of the critique of traditional accounting, it is essential to closely examine the conceptual and practical issues which determine whether net worth reporting can deliver on these expectations. There are many more such issues than can feasibly be covered in this paper, and the discussion which follows is therefore highly selective and focuses principally on alternative valuation concepts for non-monetary assets. Technical as this issue might appear, it has profound implications for the meaning and usefulness of net worth measures.

Before considering asset valuation concepts and their ramifications, it is important to note that, notwithstanding the clear deficiencies of the net debt measure, 'cash' accounting is not as utterly ineffective as is often portrayed. Although it has nothing better to offer in terms of stock measures, there certainly exist alternative cash accounting flow measures which address at least in part many of the problems identified above. In particular, there are cash accounting measures which distinguish between capital and current payments and receipts, and decompose the cash budget balance into the current account deficit/surplus and the capital account deficit/surplus. The great advantage of this is that the budgetary current account is not distorted by irregular capital expenditures and asset sales.

Moreover, there are accounting options which lie between the extremes of fully cash and fully accrual accounting. In fact, so-called cash accounting is in practice substantially modified to include less problematic elements of accrual methodology. One of the more recent examples of this is the development of an augmented version of net debt which includes public service superannuation liabilities and certain other non-public debt claims and entitlements.

What this means is that any decision against the development of public sector balance sheets and the rest of the framework of accrual accounting is not necessarily a decision to live with all the weaknesses of the most basic and unsophisticated version of cash accounting. There are other options.

Asset Valuation Concepts

Economic Value

In the private sector context, the principal meaning of the 'economic' value of an asset is the sum of money which represents the present-day ('discounted') equivalent(3) of the future net cash receipts which the asset can be expected to generate for its owner. Two variants of economic value may be distinguished. If the owner keeps the asset, 'future net cash receipts' refers to net receipts from the sale of the goods or services produced by the asset(4). The discounted sum of these receipts from the sale of goods or services is commonly referred to as 'value-in-use'. If, however, the owner decides to sell the asset, 'future net cash receipts' means the 'market' price of an asset. This is termed 'value-in-exchange', the presumption being that value-in-exchange to the asset's owner corresponds to value-in-use for potential purchasers of that asset.

If the firm is making rational decisions about holding or selling assets, it will regard the worth of an asset as being the higher of its value-in-exchange and its value-in-use (Henderson and Pierson, 1994: 56). In the language of the Australian accounting standards, this private sector concept of 'economic' value is referred to as the 'recoverable amount' (AASB 1010, paragraph 9), with recoverable amount corresponding to value-in-use if the asset is held for use, and to value-in-exchange if the intention is to sell the asset (Henderson and Pierson, 1994: 658).

The net worth of a firm calculated using this economic valuation of its assets provides a solvency measure, in the very real sense that for the financial value of a firm's total portfolio of assets to be less than the value of its liabilities means by definition that it is not expected to be able to meet those liabilities. At the same time, at least in a simple textbook world, economic value is the foundation of capital budgeting decisions. The textbook model of decisions about the creation or purchase of a capital asset uses economic asset valuation to calculate 'net present value', which is the value-in-use variant of the economic value of the asset minus the cost of constructing/acquiring it. If net present value is positive, the asset is worthwhile constructing/acquiring, but if net present value is zero or negative, the firm would be well advised to keep its money in its pocket. The choice between construction of the asset concerned and purchase of a second hand asset (if available) will be determined by the relationship between construction cost and its value-in-exchange. By extension, it is economic value which is critical to decisions about the takeover or sale of enterprises.

How relevant is this type of asset valuation methodology to the public sector? In these days when public enterprises are intended to operate in something close to a purely commercial manner, it is arguably not unreasonable to apply this version of economic asset valuation to public enterprises. One certainly cannot, however, do likewise in the departmental core of government (the 'budget sector'), the purpose of which is service delivery, not profit-making. Within the budget sector, most assets earn no revenue and are not intended to do so. Their function is to provide unrequited services to the community.

It helps to distinguish between these various types of assets by describing as 'social assets' those assets the benefits of which are provided to citizens without charge. At the opposite end of the spectrum are 'commercial assets'(5), the benefits of which are charged for at a rate sufficient to at least fully cover costs. There is also, of course, an intermediate class of assets where charges partially cover costs. Generally speaking, the costs of social assets, together with that portion of the costs of intermediate assets not covered by user charges, are met from the annual government budget, thus bringing them within the budget sector and making them wholly or partially 'tax financed'.

Because social and intermediate assets will only be acquired in the first place because of the non-monetary benefits they yield, it is clearly not sufficient to adopt the private sector approach to capital expenditure evaluation of recognising future expected monetary returns from assets, and disregarding non-monetary benefits to consumers. Both are relevant. In the standard language of Australian accounting, one must consider the future 'service potential' of an asset in general, rather than simply the cash component, if any, of that service potential. This leads to an extension of the economic value concept in respect to such assets. If the asset concerned is a purely social asset yielding no monetary returns, then it can only be valued in terms of the sum of discounted non-monetary benefits. This requires, of course, assigning a dollar value to those non-monetary benefits. If the asset is of an intermediate variety, then both monetary and non-monetary returns are discounted to value the asset. Those familiar with cost/ benefit analysis will immediately recognise that this is essentially what that methodology does, deducting initial cost from this measure of asset value in order to derive what is sometimes referred to as the 'net benefit' to be derived from the creation of the asset - equivalent to net present value given the broader class of benefits taken into consideration.

To highlight the broadening of the range of future benefits taken into account in the valuation of 'social' asset, it is useful to refer to this broader version of economic valuation, including both monetary and non-monetary benefits, as the social value of an asset. By contrast, we can refer to the private-sector version of economic valuation as measuring the financial value of an asset. It should be emphasised that both social value and financial value are economic valuation concepts.

To aid exposition, the assumption is made in what follows that all public sector assets fall either into the category of pure social assets or the category of pure commercial assets. The assumption is, in other words, that there are no intermediate assets. It is further assumed that all social assets lie within the budget sector of government, and all commercial assets within the public enterprise sectors.

Cost-Based 'Value'

The other key valuation concept differs fundamentally from the concepts discussed above. This is the concept of cost-based 'valuation'. Under this approach, an asset is initially recorded in the balance sheet at cost. The depreciation by which this recorded 'valuation' of the asset is reduced each year during its service life represents an attribution of the cost of the asset. In principle, the depreciation schedule should allocate the cost of the asset between time periods in accordance with the manner in which the benefits generated by the asset are distributed over time.

It should be stressed that cost-based 'valuation' does not refer solely to what accountants refer to as 'historical cost' asset valuation - which is when cost is taken to refer simply to the cost of the assets in the dollar values prevailing at the time the asset was paid for. The notion of cost-based valuation also refers to modifications of historical cost which adjust cost for inflation(6).

In the private sector, the balance sheets produced routinely as part of company financial statements are overwhelmingly drawn up using cost-based valuations of fixed assets. Even though the accounting standards (AASB 1010) seek to provide some encouragement for the use of economic valuation, the reality is that 'in practice, the measurement basis employed in Australia is the modified historical cost method' (Pierson and Ramsay,1996: 438). This is partly for reasons of practicability (see below), but also because the primary function of annual financial statements is the determination of income, which requires the subtraction of costs from revenues.

The absolutely crucial point to be made here is that, because the returns/benefits yielded by an asset bear no necessary relationship to its cost, cost-based measures of 'value' bear no necessary relationship to economic measures of value(7).

This means that balance sheets produced upon this basis actually indicate nothing at all about what a firm or asset is 'worth'. The balance sheet is merely a step between successive operating statements. When the issue at stake is the potential sale or acquisition of assets, rather than the reporting of profit, analysts within and outside the firm will conduct separate 'economic' valuations. They will do so on an ad hoc basis, rather than as a periodic reporting exercise. Given this, to describe the bottom line of the balance sheet of a private-sector company (which is known technically as 'owner's equity') as a measure of 'net worth' is inherently misleading.

The Policy Relevance of Contrasting Net Worth Concepts

Overlooking the inappropriateness of the term 'net worth' in regard to a cost-based balance sheet, one can say that for each of the asset valuation concepts outlined above, there is a corresponding net worth concept(8), obtained by valuing the total asset portfolio and then subtracting public debt and other liabilities (such as superannuation). Which if any of these concepts is the appropriate one, given the purposes which proponents of a public sector balance sheet expect the net worth concept to serve?

As it turns out, there is no single 'correct' valuation concept which could, even in principle, serve all or most of the purposes envisaged by balance sheet advocates. This should not come as any surprise. In the private sector accounting context, after all, it is well established that 'a single (valuation) concept cannot serve all purposes equally well' (Hendriksen,1970: 255).

The following analysis considers a number of the key proposed uses of public sector balance sheets, and identifies the net worth concept which might be considered appropriate to each of these purposes. It then critically evaluates the in principle case for using the specific net worth concept for that purpose. 'In principle' means that the analysis overlooks at this stage the formidable practical obstacles which stand in the way of any attempt to apply certain of these asset valuation concepts to actual assets. These practical measurement problems and their ramifications are considered subsequently.

Net Worth and Fiscal Sustainability in the Public Sector

As mentioned above, in the private sector a measure of net worth based upon the economic value of assets provides a measure of solvency. That is, if net worth is positive, future earnings are expected to be more than sufficient to meet existing liabilities. It would be manifestly absurd to judge a private sector firm's solvency solely by reference to its debt or even to its liabilities more generally, without regard to its expected future earnings. There is, therefore, in principle everything to be said for using such a net worth measure to assess solvency, and nothing to be said for using a net debt measure.

By contrast, in the public sector it is standard to use net debt or a closely related indicator as a measure of financial health. Yet clearly the public sector has substantial business assets which yield financial returns, and the existence of those expected financial returns is as relevant as the existence of debt in judging the financial health of the public sector. Would it not, then, be better to use a relevant public sector net worth measure instead of net debt for this purpose? The fact that advocates of the development of a public sector balance sheet see such a balance sheet as reporting the 'financial position' of the public sector seems to imply that this is precisely what they have in mind.

The relevant valuation concept for this purpose would obviously be financial value as defined above - that is, the same concept of value employed in the private sector. The corresponding balance sheet would be one in which assets would be valued only if and to the extent that they had financial value. Non-financial benefits to the community would be ignored in this context.

As sensible as this idea may appear, it is one with many problems even at the in principle level.

The first point is that in the public sector there is no determinate net worth concept which provides a measure of solvency in the private sector sense. To see this, consider the way in which net worth measures the solvency of a private sector. The value of the firm's liabilities measures future payments which must be made to service debt and other commitments. On the other hand, the value of the firm's assets (taken as a whole) measures expected future gross earnings(9) from the firm's existing assets. Subtracting the former from the latter gives a more or less determinate measure of profitability - and therefore of solvency - because the maximum future earnings of a firm from its existing assets is determined by the nature of those assets and the (uncontrollable) external environment. The only 'discretionary' factor is the degree of efficiency demonstrated by the firm.

Contrast this with the position of the budget sector of government. The role of the budget sector is to provide non-marketed goods and services which are financed through taxes rather than charges. The budget sector equivalent of gross earnings is the so-called 'primary' budget surplus/deficit, which is the difference between tax revenue and outlays excluding interest payments. So a public sector net worth concept designed to measure solvency would have to include not only the conventional asset and liabilities entries, but also the discounted value of future primary budget surpluses/deficits (i.e. the discounted value of future tax revenues minus the discounted value of future outlays on 'free' goods and services).

The trouble with this is that future taxation revenues and non-interest outlays are discretionary matters in a manner which future commercial revenue and costs are not, so that the concept of the 'discounted value of future primary budget surpluses/deficits' is an indeterminate one. The ultimate public sector solvency constraint is the political limit upon the capacity of Governments to increase the primary budget surplus by raising taxes and/or cutting outlays. This debt-servicing constraint cannot be quantified. The only manner in which this 'comprehensive' concept of public sector net worth can even be made determinate is to assume that a specific set of tax and outlays policies will apply into the future. On this basis, one could in principle calculate a net worth measure which would measure the sustainability of given fiscal policy settings into the future, given the assets and liabilities which the public sector has at the outset. Something like such a 'comprehensive' net worth measure has been conceptualised by the economist Willem Buiter (1990: 62-64), and effectively the same idea is embodied in what economists refer to as the 'intertemporal budget constraint'. These measures are, however, theoretical constructs rather than practical propositions. In the real world, the task of assessing fiscal sustainability employs the cash flow counterpart of this net worth concept. That is, finance departments estimate the trend fiscal deficit assuming the continuation of existing tax and outlays policies, and from this they work out whether there is any implicit trend towards an unsustainable 'explosive' growth in debt/national output.

Problems of a 'Financial' Net Worth Measure

The 'financial' net worth measure postulated above as a potential improvement upon net debt for the purpose of assessing fiscal sustainability - together with all the other variants of the net worth concept discussed below - is a partial rather than comprehensive net worth measure, given that it ignores future tax and non-interest outlays. Without exception, the balance sheets and net worth measures actually constructed by governments are partial measures which ignore both taxes and non-interest outlays. Superficially, they look like private sector net worth measures and/or balance sheets, but in reality they are fundamentally different.

The contrast between public and private sectors is illustrated nowhere better than in the fact that negative 'financial' net worth would neither imply that the public sector was in some sense insolvent, nor even indicate that financial management had been unsound (unless one for some reason adopts the crude 'balanced budget' view discussed below). Negative financial net worth of this sort could, for example, result from nothing more sinister than the partial funding of social infrastructure assets from debt in accordance with the conventional concept of intergenerational equity outlined earlier. The most that could be claimed for the net worth measure drawn from a balance sheet based upon a financial valuation of the capital stock is that it might give a better indication than net debt of the degree of fiscal pressure implied by future financial commitments - that is, of the extent to which taxes will in future have to be levied (over and above those required to pay for current outlays) to service inherited liabilities.

Even in attempting to play this modest role, a financial net worth measure faces further substantial difficulties. There are two difficulties in particular worth mentioning here.

Firstly, the financial valuation of many public sector business assets is as fundamentally indeterminate as the valuation of future tax receipts, and for the same reason. Public enterprises with substantial monopoly power enjoy considerable discretion in setting prices, and as a result there is commonly a substantial and variable taxation component built in to the prices charged by public enterprises such as electricity utilities and water authorities. Consequently, the financial valuation of such assets is a moving target. In Victoria, to take one relevant example, the Kennett Government in 1993 imposed substantial surcharges on utility charges to assist in addressing that State's serious financial difficulties. How would one handle significant unanticipated changes in public enterprise pricing policies of this type in a 'financial' balance sheet? Is one to substantially revalue assets every time a discretionary pricing policy change occurs?

Secondly, because by definition the commercial assets of government are largely located outside the 'budget sector' of government (where, in contrast, most assets will be 'social' in nature and therefore have no 'value-in-use' financial value), the use of a financial net worth measure implicitly involves the accounting consolidation of the budget sector of government with the public enterprise sector(s). Despite the fact that 'whole of government' consolidation has for some time been fashionable in public sector accounting, the use of consolidated measures for the specification of 'headline' fiscal measures and fiscal targets raises very serious difficulties which have not for the most part been recognised, let alone satisfactorily addressed, by exponents of the idea. Although this issue cannot be properly canvassed here, the broad point is that the fundamental differences in the objectives of the budget sector and public enterprise sectors, coupled with the substantial (and much increased over recent years) autonomy of public enterprises, render the meaning and significance of whole of government consolidated measures unclear.

Asset Sales Transparency

The potential role of net worth measures in removing fiscal illusions associated with asset sales and analogous transactions would also require the use of a 'financial' net worth measure. More precisely, it would require a balance sheet in which (non-monetary) assets were valued in terms of the 'value-in-use' variant of financial value-that is, in terms of the discounted sum of future expected earnings from the asset if left in government hands. With a balance sheet drawn up on this basis, the change in net worth arising from the sale of an asset would be equal to the difference between the sale price and the value of the asset in the continued possession of government. In other words, the change in financial net worth would provide a measure of the 'true' impact of the asset sale upon the financial position of government(10).

Net Worth and Intergenerational Equity

Having considered the potential application of net worth concepts to the assessment and monitoring of fiscal sustainability, we turn to the potential role of net worth in regard to the other key criterion for the appraisal of fiscal policy - intergenerational equity. Expanding on the description given earlier, the conventional version of the intergenerational equity principle can be summed up in the propositions that

it is reasonable to defer some of the costs of an item of government expenditure if and only if benefits will accrue in the future from that expenditure (if, in other words, it is capital rather than current expenditure) and those benefits are not less than any deferred costs, and
the contribution made by taxpayers in each time towards covering the costs of such an item of capital expenditure should be proportionate to their share of the benefits generated by that expenditure.

A crucially important point is that this principle is applicable only to the financing of social assets, and not to commercial assets. (The presumption behind this is that commercial and efficiency criteria should take precedence in determining pricing policy for the services yielded by commercial assets). In sectoral terms, this means roughly that the intergenerational equity principle applies to the budget sector, and not to the public enterprise sectors. (Remember that we are assuming for expository purposes that all budget sector assets are social assets, and all public enterprise sector assets are commercial assets).

How do concepts of net worth tie in with the intergenerational equity principle? It turns out that a policy commitment to intergenerational equity has implications for two net worth concepts - those based respectively upon the social valuation of assets and upon the cost-based valuation of assets. Moreover, because intergenerational equity principles apply to the budget sector and not the public enterprise sectors, the policy-relevant net worth measures would - notwithstanding the current popularity of 'whole of government consolidation' - be measures of budget sector net worth, rather than total public sector net worth.

The social value of an asset is, as outlined above, the sum of the discounted benefits generated by the asset (expressed in monetary equivalents). Therefore, given that intergenerational equity demands that costs be deferred to be future via debt and similar liabilities only if future benefits equal or exceed those costs, the consistent application of the intergenerational equity principle must result in the social value of assets in the budget sector being not less than the total liabilities of that sector. In other words, social net worth will certainly not be negative(11).

To go beyond this, we need to turn to cost-based valuation. As outlined earlier, the essence of cost-based valuation is the writing off of the cost of the asset over time in proportion to the benefits generated by the asset(12). This clearly parallels the conventional version of the intergenerational equity principle, in that depreciation charges in a cost-based accounting system would in principle allocate costs over time in accord with this principle. The only difference is that whereas in the case of a commercial asset the depreciation schedule should in theory accord with the manner in which the monetary benefits from the asset are apportioned over time, in the case of a social asset the depreciation schedule should be based upon the spread of non-monetary benefits.

If the depreciation schedule in an accrual operating statement does in practice allocate costs over time in proportion to the allocation over time of the benefits derived from the asset, then the consistent application of the intergenerational equity principle implies a consistently balanced operating statement (i.e. one registering no deficit or surplus)(13). And, as explained above, the bottom line in an accrual operating statement is equivalent to the change in net worth. This means, in a nutshell, that the consistency of fiscal policy with the conventional formulation of intergenerational equity is assessable by reference to whether the cost-based net worth measure remains constant at zero (or remains constant at some other level if intergenerational equity policies were not applied consistently at all times in the past(14)).

Government Savings

Cost-based valuation is also the appropriate accounting basis upon which to measure net government saving. Gross saving is cash income used to acquire assets or reduce liabilities, from which arises the fact that capital expenditure represents an application of, rather than a reduction in, government saving in the time period when it occurs. The idea behind the concept of 'net' government savings, on the other hand, is that this capital expenditure should gradually be re-classified as consumption expenditure rather than saving as the asset's service potential is exhausted. The accrual operating balance is therefore an appropriate measure of government saving because it expenses the costs of fixed assets not in the period when the capital expenditure occurs, but progressively over the service life of the asset. Expressed in either gross or net terms, 'savings' remains a measure of the cash income input directed towards increasing the capital stock or reducing liabilities. It bears no necessary relation to the value of assets created through the application of that cash income. Consequently, no balance sheet based upon economic concepts of asset value will produce a net worth figure which indicates anything about government savings(15).

If one were to accept the view that increasing government savings is an effective way to increase aggregate national savings, then one is likely to favour, not the maintenance of constant net worth as suggested by the traditional intergenerational equity approach, but rather the augmentation of net worth. (Of course, there is great controversy not only about the link between government 'saving' and national saving, but also about the significance of the choice of means by which government savings might be augmented).

The Balanced Budget Rule and 'No Borrowing for Social Infrastructure'

Brief comment should be made in passing about the net worth implications of the view, alluded to above, that it is inappropriate for government to create liabilities which will have to be borne by future taxpayers. This is the 'no borrowing for social infrastructure' view, so described because it rules out debt financing for infrastructure which cannot be self-financed through user charges, while accepting borrowing to finance commercial assets. It is a view which is more or less equivalent to the 'balanced budget' rule, since these two fiscal rules mean the same thing if (as assumed earlier) all social assets lie within the budget sector and all commercial assets lie within the public enterprise sectors.

Evaluating this fiscal rule is beyond the scope of this paper (but see Robinson 1996b). Its implications in net worth terms are, however, straightforward. It implies simply that the financial net worth of the budget sector should never be negative. This contrasts with the net worth rules derivable from the traditional intergenerational equity approach (that the social net worth of the budget sector should never be negative, although financial net worth might well be).

Capital Stock Adequacy and Future Replacement Needs

By contrast to the other potential uses of balance sheets and the net worth measure which have been discussed above, there is no balance sheet concept which, even in theory, will deliver on expectations that it will provide a means of assessing both capital stock adequacy and future capital replacement needs. This is partly because, paradoxically, an erosion in the value of the capital stock over time could be due to either of these two factors without the other, or to both factors operating together. It might, for example, be that there is a considerable portfolio of fixed assets, but that many are reaching the end of their service lives. Alternatively, there may simply be a paucity of fixed assets. This is just not something that one can judge by looking at a set of figures in a balance sheet. Figures, moreover, tell nothing about key determinants of public infrastructure needs such as demographic outcomes and technological change.

The Macroeconomic Impact of Fiscal Policy

The lack of any mention of macroeconomic policy in the above discussion of the functions of balance sheets and net worth measures may appear to the reader as a glaring omission. In fact, it reflects the substantial consensus amongst economists that cash measures are most appropriate for the task of assessing the impact of fiscal policy on economic activity and the price level. Without going into unnecessary complexities, implicit in this is a belief that the macroeconomic impact of fiscal policy is principally the result of the infusion or withdrawal of cash from the economy through government expenditure and taxation(16).

This is a point which is not essentially in dispute, so that even strong advocates of accrual accounting accept that in an accrual environment it is necessary to produce cash data as well. And as they point out, this is not a problem given that cash data is a subset of the accrual data system.

The Gulf between Concept and Practice in the Valuation of Fixed Assets

The above analysis indicates that no single version of a public sector balance sheet and accompanying net worth concept would be capable, even in principle, of delivering all or even most of the results anticipated by advocates of accrual accounting. One would, rather, need a number of different balance sheets prepared. Even then not only would there remain real ambiguities, but some of the putative functions of a balance sheet would remain out of reach.

So far, we have considered the role and interpretation of net worth at a purely conceptual level, ignoring practical issues which arise when one attempts to put these concepts into practice. Yet there are great practical problems, foremost amongst which is the measurement problem.

Measurement difficulties are most profound by far in respect of economic concepts of asset value and net worth. These difficulties are, moreover, strongly apparent even in a commercial private sector context. The future flows of cash or benefits which must be discounted to derive an economic measure of value are inherently uncertain, so that any projection of these flows is necessarily a matter of judgment in respect of which there can be considerable and quite legitimate differences of opinion. (The only, limited, exception to this is the market value ('value in exchange') of specific widely-traded standard assets).

This creates an inevitable unreliability about economic asset valuations - an unreliability which is greatly compounded by the fact that insiders with the understanding and information necessary to formulate the best estimates of economic asset value often have a direct personal or corporate interest in the impression created by such a valuation. Unreliability therefore becomes manipulability. This is an important additional factor in the reliance placed by private sector accounting upon cost-based valuation for non-monetary assets.

These measurement problems might be considerable in a private sector commercial context, but they are of a greater order of magnitude altogether in respect to 'social' assets in the budget sector core of government. For in estimating the economic value of such assets one has to contend not only with uncertainty about the future, but also with the enormous practical and conceptual problems of expressing non-monetary benefits in monetary terms.

Although a public sector balance sheet drawn up from cost-based valuations is not as impracticable as is the notion of one based upon economic valuations(17), it is important to recognise that cost-based valuation is not free of significant measurement problems. One of the most significant problems is the prevalence of depreciation schedules (that is, rules for the apportionment of cost over time) which are essentially arbitrary. This is a manifestation of precisely the same information problem which renders economic valuation impracticable. In principle, a depreciation schedule should be asset-specific, and should reflect the manner in which the benefits derived from the particular asset are spread through time. In practice, the allocation of these benefits over time is just as fraught with uncertainty and measurement problems as is their magnitude. The arbitrariness of depreciation is therefore a matter of practical necessity. The public sector is in this respect little different from the private sector, where 'off the shelf' depreciation schedules - whether straight line, declining balance or whatever - are adopted because such is the only practical course of action(18).

'Horses for Courses': the Potential Role of Different Types of Balance Sheet

It is useful to summarise the implications of the above analysis for the principal variants of the balance sheet.

Balance sheets based upon economic valuation of fixed assets are essentially impracticable. Moreover, analysis suggests that even in principle they may not have much to offer as a means of monitoring fiscal sustainability.

Cost-based balance sheets are somewhat more practicable, although the same essential problems manifest themselves in a different manner (especially through the arbitrariness of depreciation schedules). It nevertheless may be that the development of a balance sheet of this type could ultimately be worthwhile. There are, however, a host of technical - but nevertheless highly policy-relevant issues - which need to be resolved in relation to such an approach.

In principle, the prime policy rationale for a cost-based balance sheet is, as discussed above, the monitoring of the ramifications of fiscal policy for intergenerational equity. One key problem is that much more will tend to be read into such a balance sheet. There will almost inevitably be misinterpretation, with the reading into cost-based 'net worth' measures of interpretations which they cannot, even in principle, sustain. In particular, it is probably inevitable that any 'net worth' measure will be treated as a measure of solvency or of fiscal sustainability. Yet clearly the cost of a fixed asset bears no necessary relationship to the future financial returns (either in use or through sale) from that asset, so that a cost-based balance sheet indicates absolutely nothing about fiscal sustainability.

This is not surprising. After all, as noted above, the standard annual balance sheet of a private sector company tells one nothing about solvency or even about the 'worth' of the company in the standard sense of the word. The very term 'net worth' is a misnomer in such a balance sheet. The Victorian Commission of Audit recognised this point in its 1993 Report. In the context of presenting the balance sheet for the State of Victoria which it was required by its terms of reference to prepare, it disclaimed any pretence of estimating net worth, commenting that

... a balance sheet does not normally purport to measure net worth ... Even a business enterprise which utilizes its assets to generate profits would not generally claim that its balance sheet measures net worth and the Commission believes that it is inappropriate as an objective of financial reporting for the public sector. (1993: 25).

For the same reason, a cost-based balance sheet cannot fully deliver on the worthy objective of removing fiscal illusions arising from asset sales. The most that the change in a cost-based 'net worth' measure due to an asset sale could signify, even in principle, is whether or not the asset was sold for more, or less, than depreciated cost. While this is an advance upon cash accounting, it is still flawed. Because the economic value of an asset retained in the public sector need bear no particular relation to its cost (either historical or inflation-adjusted), differences between cost and economic value will be a potential source of fiscal illusion.

Analogous comments apply to the use of a cost-based 'net worth' measure for one of the other prime functions for which balance sheets tend to be employed. This is the notion that the existence of a stock of fixed assets 'valued' at, say, $50 billion in some way justifies the existence of debt of the same magnitude. In practice, the lack of a link between the cost and (economic) value of assets (and the arbitrariness of depreciation schedules) makes this an unwarranted interpretation. It might, for example be the case that the assets concerned were inappropriately costly because of construction inefficiencies or so-called 'gold plating' (i.e. construction to an unnecessarily high standard). Alternatively, the assets concerned may simply represent a poor investment. The example of expensive Olympic games capital works being constructed in Sydney springs to mind. In such cases, the asset might never have been worth its original cost, and the application of an arbitrary depreciation schedule to allocate cost over the service life of the asset might do no more than to spread the adverse impact upon social or financial net worth.

The classic use of a balance sheet for the purpose of justifying debt is to be found in the Victorian balance sheets of the early 1990s, referred to above. These were clearly prepared to rebut suggestions that Victoria had a growing debt problem by shifting attention from net debt to net worth, as was made explicit by the State Treasurer of the time (Victorian Treasury, 1991a: iii-iv).

The best argument that one can advance for the relevance of such a balance sheet for such purposes is a very second-best one - namely that, even if the net worth number, or the change in the net worth number, cannot be relied upon as a performance indicator in respect to fiscal sustainability or to the fiscal impact of asset sales, at least the balance sheet perspective refocuses attention from debt in isolation to debt in the context of assets.

It should be emphasised that none of the above is intended to be critical of cost-based balance sheets per se. Indeed, a conceptually sound cost-based balance sheet may be the least unreliable and most useful of all balance sheet options. But it has to be recognised that such a balance sheet can only ever even in principle serve some of the multiple demands which accrual accounting enthusiasts tend to place upon balance sheets. And in practice, it will in practice serve even those objectives only imperfectly.

Flawed Accounting Standards and the Development of Balance Sheets

The asset valuation methodologies recommended in accounting guidelines and put into practice in actual public sector balance sheets developed to date in Australia appear to be characterised by lack of conceptual clarity, eclecticism and a failure to tie accounting choices firmly to intended policy purposes. It is not possible in this paper to provide a detailed analysis of these developments. Some overview comments will have to suffice.

The principal accounting standards are AAS 29: Financial Reporting by Government Departments (December 1993), ED 62: Financial Reporting by Governments (March 1995). The 'ED' in the second of these stands for 'exposure draft', meaning that the latter standard has not yet been formally promulgated. The general accounting guidelines on the revaluation and depreciation of non-current assets (AASB 1010 and AASB 1021) also have significant application to government. The local government equivalent of the above standards is AAS 27, which adopts much the same approach to questions of fixed asset valuation and depreciation.

The valuation methodology for fixed assets recommended (for it is not mandatory) in AAS 29, ED 62 and AAS 27 is that of 'written-down current cost'. One might think that this represents the adoption at the conceptual level of a cost-based valuation methodology, given that 'current cost' accounting is often interpreted as referring to asset valuation based on historical cost adjusted for changes in the construction cost of the particular type of asset. However, it is apparently not that simple. It would appear that, in principle, these standards actually favour economic asset valuation - in its 'social' variant in the case of social assets(19). There appears to be a considerable degree of confusion between value and cost(20). Written-down cost might be assessed by, say, market price (i.e. value-in-exchange) rather than replacement cost, and their appears to be a belief that the measurement process approximates economic value. This impression is reinforced by accounting standards issued by governments, which in many cases nominate economic value as the first-best valuation approach, and treat other methodologies as compromises when direct measures of economic value are unavailable (see, for example, Victorian Department of Finance, 1995: 11-20).

Essentially, the accounting practice tends to 'solve' the practical problems which obstruct attempts to measure what it sees as the ideal value concept - economic value - by laying down what is effectively a hierarchy of asset valuation compromises, the last of which is generally a variant of cost-based valuation. And in the final analysis, the balance sheets actually developed tend to rely very heavily upon cost-based valuation of non-monetary assets, albeit for reasons of practicability rather than conceptual choice. Indeed, there is a tendency to scorn balance sheets based explicitly upon cost-based valuations, such as those developed in Victoria in the early 1990s which were drawn from the cost-based 'PIM' methodology long used by the Australian Bureau of Statistics to measure capital stock and depreciation (Victorian Treasury, 1991a, 1992; Macmillan, 1991: 12).

As this paper has shown, there is considerable reason to doubt even in principle the usefulness of a net worth measure based upon economic valuation of capital assets. Regrettably, an analysis of the work of those involved with the development of accrual accounting standards suggests that these fundamental issues concerning the meaningfulness of the net worth and operating deficit/surplus measures have not been properly addressed. In any event, the idea that the menu of 'compromise' valuation methodologies employed within the 'written-down cost' framework approximates economic value seems without foundation. Ma and Mathews are without doubt correct in observing that 'at the measurement level, it is not possible to value in any meaningful sense, for disclosure in the balance sheet, assets which cannot conceivably be traded in the market place and hence have no market value' (1993: 79).

The Problem of Eclecticism

This situation is fraught with problems. There is, firstly, the problem of eclecticism - that is, the de facto mixing and matching of valuation concepts in the same balance sheet. Although the accounting standards recommend consistency in valuation methodologies, the hierarchy of valuation options embraced within the preferred methodology makes it clear that consistency need not actually mean consistency.

Eclecticism gives the numbers in balance sheets a highly rubbery quality. The inherent subjectivity of economic valuation which was discussed above is compounded by the considerable discretion available over the choice of valuation methodology within this eclectic context. The result is that it is not difficult to come up with quite different net worth figures, even while complying with the accounting standards. Given that there is a tendency for the concrete quality of accounting numbers to create an illusion of precision, this is a very important problem. There is a misplaced tendency of some advocates of public sector balance sheets to downplay this issue, and to suggest that it is more important to get some numbers on paper than to worry too much about the justification for those numbers. This was the impression which the Commonwealth Joint Committee of Public Accounts formed of the New Zealand approach (JCPA, 1995b: S 219). Yet to take this view is implicitly to gravely underestimate the significance which will inevitably be read into the numbers.

Rubberiness means manipulability. Proponents of accrual accounting have frequently represented its adoption as a prime means by which to minimise the accounting manipulation which characterises traditional cash accounting in the public sector (e.g. Churchill, 1992: 18; Mellor, 1995: 80). But such claims are not only without theoretical foundation, but are also demonstrated to be inaccurate by experience with accrual accounting in the Australian public sector. As Professor Bob Walker - himself an advocate of the development of an appropriate form of accrual accounting in core government - has noted, '[accrual] accounting methods being adopted within the public sector are so flexible (and amenable to manipulation) that often, little credence may be placed on many of the numbers presented in published accrual-accounting based reports' (Walker, 1995: S 156; Walker, 1988: 82).

The most fundamental problem of eclecticism is, however, that it greatly compounds the problem of meaning. What policy relevance could one possibly assign to a balance sheet drawn from a diversity of different valuation methodologies? Whereas one might hope that an appropriately constructed cost-based balance sheet and operating statement might provide guidance on intergenerational equity, government savings and some other policy matters, what possible meaning can be assigned to accounting measures which end up falling between the stools of economic valuation and cost-based 'valuation'?

Such a balance sheet ends up with all of the limitations of a cost-based balance sheet but with none of its virtues. It is also intrinsically prone to misinterpretation.

The Problem of Misinterpretation

As explained earlier, cost-based net worth measures indicate precisely nothing about fiscal sustainability, and even economic net worth measures are of highly questionable value in this respect. Clearly, therefore, eclectic balance sheets designed in accordance with present accounting standards also indicate precisely nothing about fiscal sustainability. Yet it is perhaps inevitable that any 'net worth' measure drawn from these balance sheets will be widely interpreted in the general community as indicating something about solvency or fiscal sustainability. As the Commonwealth Auditor-General has noted (Barrett, 1995: S 38), the mere act of valuing an asset tends to create 'a misleading impression .... that these are resources available to the government to meet its obligations'. The fact that the opening balance sheet in New Zealand in 1993 purported to show substantial negative net worth was of great political significance in helping to 'sell' the fiscal policies preferred by the Government of the time, despite the fact that the balance sheet concerned was afflicted not only by the problems discussed above, but also that there was extensive and quite unjustifiable use of the 'historical cost' version of cost-based asset valuation.

It is regrettable that this type of misapprehension is encouraged by accrual accounting advocates who characterise the net worth measure as a summary of the 'financial position' of government. The fundamental trouble appears to be that, because there is no clear distinction between economic value and cost in the accounting standards, there is a corresponding failure to clearly recognise the respective limitations of each type of measure.

A natural extension of this misinterpretation is the unfounded, but eminently understandable, notion that the greater is net worth, the better. The result of such a misinterpretation would be that, having moved away from the net debt measure because in part of the tendency of that measure to encourage debt elimination as a policy objective, one would be faced with an equally inappropriate policy bias based upon simplistic misconceptions about the new headline measure.

The growing use of balance sheets in the public sector can be expected to produce new types of misconceptions in coming years. A case in point is the treatment of asset sales. As indicated above, one of the core claims made by accrual accounting advocates is that a balance sheet approach will eliminate fiscal illusions associated with asset sales. As the Australian Bureau of Statistics puts it in its recent Exposure Draft, 'asset sales have no effect when the asset is sold for its real value' (ABS, 1995: B-3). Yet, as explained above, this is only true if the asset concerned is valued - and valued accurately - in the balance sheet in terms of the 'value-in-use' variant of economic value. Insofar as balance sheet asset valuations will commonly not meet this criterion, it is clear that the expectations held by the enthusiasts about asset sale transparency through accrual accounting are not fully warranted(21).

The most that can reasonably be argued for the development of public sector balance sheets is not that they will remove fiscal illusion arising from asset sales, but rather that the degree of fiscal illusion will be much reduced. The use in balance sheets of valuation methodologies other than 'value-in-use', and inaccuracies in valuations even where they are intended to be on a 'value-in-use' basis, will ensure that there remains substantial scope for fiscal illusions concerning asset sales even within an accrual accounting framework. It would, moreover, be unrealistic not to assume that fiscal illusions arising from the failure of the public to understand the technicalities of public sector accounting will be exploited by politicians just as much in an accrual accounting environment as they have been in the traditional cash accounting environment.

Are Balance Sheets and Net Worth Measures Worth the Trouble?

The aim of the above discussion has been to restore some balance to the debate about cash versus accrual accounting. Years of experience has taught us the defects of cash accounting - to the extent, in fact, that commentators very quickly point out what is going on when governments employ many of the customary accounting 'rorts'. By contrast, despite the substantial movement towards its adoption, there has so far been quite limited experience in core government with accrual accounting, balance sheets and net worth measures. Not only that, but many proponents of accrual accounting have shown little interest in realistic appraisal of the experience to date in Australia and other countries. Instead, they tend to argue by contrasting 'warts and all' cash accounting with an utterly unrealistic idealisation of accrual accounting.

It is important to stress that this paper has in a sense only scratched the surface in discussing the conceptual and practical problems which arise in relation to public sector balance sheets and net worth measures. Amongst the issues which it has not been possible to discuss are, for example:

Other valuation issues: there are, for example, significant issues relating to the practice of valuing monetary liabilities and assets at 'market value' as distinct from 'face value'. And on the assets side, issues such as land valuation have not been touched upon. At a broader level, one of the crucial issues which it has not been possible to discuss here is the distorting effect of valuation shocks upon balance sheets and accrual operating statements. Also, there are the much-debated issues concerning the valuation of heritage and cultural assets.
The question of the scope of a public sector balance sheet: that is, what 'assets' and what 'liabilities' should be included. On the liabilities side, there is an important debate on the adequacy of the standard 'control' criterion for inclusion. The treatment of 'private' infrastructure projects which involve direct or indirect public guarantees is one issue here. Another is the treatment of the 'liabilities' which can be regarded as accrued if social security provisions remain unchanged. On the assets side, there are questions as to how far, if at all, a public sector balance sheet should go in including items such as the value of publicly-owned mineral and energy deposits, or even the 'human capital' created by, for example, the education system.

Notwithstanding that it has not been possible in this paper to comprehensively discuss all the key issues relating to balance sheets and net worth, enough has been said to demonstrate that it is at this stage far from clear that accrual accounting is desirable for the budget sector of government. Nor, more specifically, is it yet clear that the development of public sector balance sheets and net worth measures is a desirable reform.

On the one hand, attempts to implement balance sheets founded upon economic valuation of fixed assets are impracticable and of doubtful value even in principle. It would therefore be premature and inappropriate for the Commonwealth to commit itself on the basis of presently available accounting policies to the regular provision of a balance sheet and net worth measure. Moreover, although the ABS has laid its plans for regular accrual reporting of government finance statistics on the table, it would be desirable for implementation to be put on hold while further consideration takes place.

It may, by contrast, be feasible to design a system of accrual accounting for the public sector which would be a valuable policy tool. This would probably be founded upon cost-based asset 'valuation', with principal relevance to the assessment of intergenerational equity and public sector savings. Such an approach would, incidentally, be much closer to private sector accounting practice than is the approach currently embodied in public sector accounting standards. There are, however, a range of substantial conceptual and practical issues which need to be assessed in determining whether or not this is an appropriate direction for public sector accounting. In evaluating these issues, it is necessary to go 'back to the drawing board'.

This may appear a somewhat dour stance to take at a time when the accrual bandwagon is moving forward so steadfastly along the road constructed for it by those who develop accounting standards. Why spoil the party when everyone appears to be having so much fun? Three points are worth emphasising strongly in answering that question.

The first is that, although those concerned should speak for themselves, it is very clear that some of the public organisations and officials closest to the action on these matters have, at very least, major reservations about what is happening. The Commonwealth Department of Finance has made clear its reservations about the policy relevance of net worth measures, expressed concerned about misinterpretation, and counselled that if public sector accrual accounts are to be progressed, this should be restricted to trialing rather than full-blooded implementation (Department of Finance, 1995: S77). Treasury has expressed similar concerns, and has observed that (notwithstanding the work of accounting standards bodies) 'at this point in time the appropriate methodology for measuring net worth has yet to be fully resolved' (Treasury, 1995: S118). The head of the Northern Territory Treasury has opposed accrual accounting outright, on strongly reasoned grounds which resonate at a number of points with the analysis of this paper (Conn, 1996). It may be tempting for individuals outside government to suggest that these concerns simply reflect bureaucratic stodginess, but in reality they reflect the old adage that 'the devil is in the detail'. It may be easy for enthusiasts in private accounting firms and elsewhere to restrict their vision to the glorious 'big picture' of accrual accounting and to avoid troublesome details, but those within government who deal continuously with the nitty gritty of public finances can enjoy no such luxury.

The second point is that the costs associated with the implementation of the present approach to accrual accounting are very substantial, and certainly well in access of the costs of administering traditional cash accounting. In particular, the costs associated with asset valuation are quite high. Part of the continuing consideration of accrual accounting options ought to be a proper weighing of the considerable administrative costs involved in the operation of what is fundamentally a much more complicated system of accounting that the traditional cash system.

The third point is that, as noted earlier, the 'cash' system is not as utterly hopeless as many exponents of accrual accounting would have one believe. The system has not in any sense been a static one, but has been through significant reforms and improvements, including increasing national standardisation around a reformed 'Government Finance Statistics' system (Albon, 1994). The standard critique of cash accounting applies with greatest force to the traditional debt measure and to the associated budget deficit measure. Yet even without the development of accrual accounts, alternative cash accounting measures have existed for some time which address, at least to a significant extent, many of the difficulties with the present deficit measure. In particular, the ABS publishes accounts which distinguish current and capital expenditures and receipts, and which provide a measure of the budgetary 'current account' balance, excluding these items. Other even more sophisticated adjustments of cash measures are available. In fact, some of the more relevant of these alternative measures are reported in the Commonwealth budget papers themselves. Not only that, but so-called 'cash' accounting is increasingly incorporating less problematic elements of the accrual accounting schema, an example being the supplementation of net debt measures with measures of the value of superannuation and similar liabilities.

It is, above all, crucial that the Australian public sector does not travel prematurely down an inappropriate and costly accrual accounting path simply because of an unthinking belief that to do so is desirable because it in some sense places the public sector on a private sector footing. The aims, objectives and modus operandi of the core budget sector of government differ fundamentally from those of the private sector, and the difference is one which no amount of fashionable 'reinventing' of government can remove. It is therefore absolutely crucial to tailor any form of accrual accounting adopted within the departmental core of government to the concrete policy objectives which such accounting reform is intended to serve.

Endnotes

There is considerable theoretical complexity and disputation surrounding not only the concrete operationalisation of this criterion, but also (thanks particularly to the vogue enjoyed by the now-discredited doctrine of full 'Ricardian equivalence') about the meaningfulness of the criterion in any form. More recently, advocates of what is known as 'intergenerational accounting' have claimed (with exaggeration, in the author's opinion) to have overturned traditional preoccupations with the budget bottom line. It is impossible to canvas these issues here. Instead, we simply assume the traditional formulation of this concept.
This is a simplification, which abstracts from a number of factors which will give rise to differences between the deficit and the reported change in net debt. One such factor is the (questionable) practice of reporting net debt at market value rather than at face value.
Because a dollar now is worth more than a dollar later, future expected earnings have to be 'discounted' to bring them back to their present day equivalents.
Which makes it clear that this valuation methodology applies to assets or groups of assets which constitute a complete production process.
Sometimes referred to, in somewhat misleading terminology which will be avoided here, as 'economic assets'.
One such modification is 'current purchasing power accounting', which adjusts historical cost in line with an index of general inflation. What is known as 'current cost accounting' is also a cost-based valuation methodology insofar as it is defined by the adjustment of historical cost in line with price trends for the specific type of asset. As discussed subsequently, however, there appears to be some confusion on this, with current cost accounting being confused by some with economic valuation.
The only link between the two is that the depreciation schedule should, in principle, allocate the costs of an asset over time in proportion to the allocation over time of the benefits which the asset yields and which constitute its economic value. And even here, as is discussed further below, the practical reality is that informational and other problems mean that essentially arbitrary depreciation schedules are commonly employed.
We set aside here certain issues pertaining to the valuation of non-monetary liabilities, which further complicate matters.
Gross earnings i.e. earnings before interest.
A caveat is, however, necessary here. As important as information about the financial impact upon government of a major asset sale might be, the case for or against an asset sale cannot be made simply upon such information. For example, it may be that a 'favourable' sale price for a public enterprise reflects the intention of the new owners to raise prices. It may even be that the government has effectively sold discretionary monopolistic pricing powers along with the physical assets of the enterprise. In these cases, the improvement in the fiscal position of the public sector would simply reflect a worsening in the position of consumers.
Or if there was a failure at some time in the past to abide by the principles of intergenerational equity, the application of those principles from a specific starting point would imply that there be no decline in social net worth from that time on. It should be noted that these conclusions only strictly apply under condition of certainty. Unexpected adverse events can, of course, reduce the social or financial value of assets.
A technical issue often raised in the public sector accounting debate concerns the appropriateness of depreciation for assets which are maintained and renewed indefinitely. At a conceptual level, this issue presents no difficulty - replacement charges simply replace maintenance in the operating statement (the so-called 'renewals' method). The problem occurs at a practical level, in large measure because of the informational deficiencies discussed below.
This abstracts, of course, from a number of real-world complications, and most particularly from the impact of the business cycle upon the budget.
In New Zealand, the Fiscal Responsibility Act 1994 targets a level of net worth which provides a buffer against future adverse events. The thinking behind this is unclear. Recent New Zealand fiscal policy documentation suggests (New Zealand, 1996: 20) that one consideration is the need to build up net worth in anticipation of a bunching in social security expenditure for demographic reasons. This makes sense if one is working with a balance sheet which does not include social security as a liability, and is equivalent to running a constant net worth policy in the context of a balance sheet which does include social security. However, given that the New Zealand balance sheet does indeed encompass social security, the rationale for building up a substantial positive net worth in that country awaits further clarification.
Notwithstanding the discussion on this topic in the ABS Exposure Draft on Introduction of an Accrual Basis in Government Finance Statistics (ABS 1995: B-3).
Although there may be other effects which fit in with elements of the accrual framework (such as alleged 'wealth effects' from changes in the market value of government debt held by the private sector), the idea that variations in the value of public fixed assets may have a direct impact on the macroeconomy is one which would find little if any support amongst economists.
This is true in respect of historical cost and Current Purchasing Power Accounting, which adjusts historical cost via an index of general inflation. Current Cost Accounting (CCA) raises significantly more difficult measurement problems (Henderson and Pierson, 1994: 618; Pierson and Ramsay, 1996: 578). Even the measurement difficulties of CCA are, however, not of the same order of magnitude as those involved in economic valuation.
Moreover, for assets in respect of which continual repair and maintenance implies a more or less indefinite life span (see note 12), there is the likelihood that unthinking application of depreciation will lead to a 'double whammy' impact upon the operating statement (Aiken and Capitanio, 1996: 566), making the operating deficit appear worse than it should.
In the language of AAS 1010 (Paragraph vii), the 'carrying amount' of the non-current assets of 'not-for-profit entities ..... should reflect their remaining service potential as at the reporting date'.
In a monograph on asset valuation issues, one of the key personnel in the standards-development process refers to written-down cost as valuation 'on a current value basis', equates written-down cost with what he calls 'economic cost', and asserts that 'assets held for their value-in-use provide the controlling entity with an economic benefit equal to their economic cost' (Rowles, 1992: 67, 63).
Insofar as some saleable assets are to be valued at market value, rather than cost, under the various valuation policies which currently exist, one has the equally bothersome problem that the sale of such an asset will necessarily have zero impact upon the balance sheet (assuming the estimate of market value is accurate). The balance sheet will therefore indicate precisely nothing about the real impact of the asset sale concerned.

References

Aiken, M. and C. Capitanio (1995), 'Accrual Accounting Valuations and Accountability in Government: a Potentially Pernicious Union', Australian Journal of Public Administration, 54 (4), December, 564-576.

Albon, R. (1994), 'Evaluating Changes in a Government's Net Economic Position', 1994 Research Lecture in Government Accounting, Australian Society of Certified Practicing Accountants, Melbourne.

Australian Accounting Research Foundation (AARF) (1995), Financial Reporting by Government, Exposure Draft 62, Melbourne, March.

Australian Accounting Standards Board (1993), AASB 1010: Accounting for the Revaluation of Non-Current Assets.

Australian Bureau of Statistics (ABS) (1995), Introduction of an Accruals Basis in Government Finance Statistics, Canberra, September.

Barrett, P. (1995), 'Submission', in JCPA, Reference: Inquiry into Fiscal Responsibility Legislation and Whole of Government Reporting for the Commonwealth, Submissions, Volume 1, Canberra.

Buiter, W. (1990), Principles of Budgetary and Financial Policy, Cambridge, Mass., MIT Press,

Churchill, M. (1992), 'The Winds of Change', Australian Accountant, March: 17-20.

Conn, N. (1995), 'Reservations about Governments Producing Balance Sheets', Australian Journal of Public Administration, 55 (1), March,: 82-85.

Department of Finance (1995), 'Submission', in JCPA, Reference: Inquiry into Fiscal Responsibility Legislation and Whole of Government Reporting for the Commonwealth, Submissions, Volume 1, Canberra.

Henderson, S. and G. Pierson (1994), Issues in Financial Accounting, Melbourne.

Hendriksen, E. (1970), Accounting Theory, Homewood, Illinois.

Joint Committee of Public Accounts (JCPA) (1995a), Accrual Accounting-a Cultural Change, Report 338, Canberra, August.

Joint Committee of Public Accounts (JCPA) (1995b), [Transcript of Evidence, 11 September] in JCPA, Reference: Inquiry into Fiscal Responsibility Legislation and Whole of Government Reporting for the Commonwealth, Submissions, Volume 2, Canberra.

Ma, R. and R. Mathews (1993), 'Financial Reporting by Government Departments: ED 55 - a Dissenting View', Australian Journal of Corporate Law, 3 (1).

Macmillan, G (1991), 'Victoria's Financial Statements: the Accounting Agenda for Reform', Paper Presented at Public Sector Management Institute Seminar, Melbourne, October.

Mellor, T. (1996), 'Why Governments Should Produce Balance Sheets', Australian Journal of Public Administration, 55 (1), March, 78-81.

Neilson, J. (1993), 'AAS27: Expensive and Unnecessary', Charter, March: 53-55.

New Zealand (1996), Budget Policy Statement, 1996, Wellington.

NSW Audit Office (1995), 'Accrual Accounting in the New South Wales Public Sector', in JCPA, Reference: Review of Auditor-General's Reports 1993-1994, Accrual Reporting, Submissions, Volume 2, Canberra.

Pierson, G. and A. Ramsay (1996), Financial Accounting: an Introduction, Melbourne.

Robinson, M. (1996a), 'Fiddling the Figures', Australian Accountant, May, 28-29.

Robinson, M. (1996b), 'The Case against Balanced Budgets', Australian Journal of Public Administration, 55 (1), March, 48-62.

Rowles, T. (1992), Financial Reporting of Infrastructure and Heritage Assets by Public Sector Entities, AARF Discussion Paper No. 17, Melbourne.

Victorian Department of Finance (1995), Recognition and Valuation of Non-Current Physical Assets, Melbourne.

Treasury (1995), 'Submission', in JCPA, Reference: Inquiry into Fiscal Responsibility Legislation and Whole of Government Reporting for the Commonwealth, Submissions, Volume 1, Canberra.

Victorian Treasury (1991a), Victoria's Public Balance Sheet, Information Paper No. 1 Melbourne.

Victorian Commission of Audit (1993), Report of the Victorian Commission of Audit, Volume 1, Melbourne.

Walker, R. (1988), 'NSW Sorts out Its Debt', Australian Business, 4 May: 80-83.

Walker, R. (1995), 'Submission' in JCPA, Reference: Inquiry into Fiscal Responsibility Legislation and Whole of Government Reporting for the Commonwealth, Submissions, Volume 1, Canberra.

Comments to: web.library@aph.gov.au
Last reviewed 19 July, 2004 by the Parliamentary Library Web Manager
© Commonwealth of Australia

Public Sector Net Worth: Is it Worth Measuring?

Marc Robinson Consultant to Economics, Commerce and Industrial Relations Group

Glossary of Terms

Accrual Accounting