II. How Should the Fiscal Position Be Assessed?
- James Daniel
- Published Date:
- August 2006
Without data, all you are is just another person with an opinion.
Just as fiscal adjustment has many objectives, it can be measured in many ways. Even the best designed adjustment will fail if the fiscal indicators on which it is based are flawed. Different problems, objectives, and economic structures imply that no single measure will fit all circumstances. Individual country practices reflect this diversity, and international statistical standards themselves are changing as countries gradually adopt the IMF’s Government Finance Statistics Manual 200110 (Box 2). The main issues relate to what should be included in the public sector, when fiscal transactions should be recorded, and what indicators should be used.
Coverage of the Public Sector
The components of the fiscal sector are the central government, subnational government, social security funds, and public corporations.
Central government refers to the activities of a country’s central authority. Transactions at this level should reflect the legal budget of the central government as well as fiscal actions of any extrabudgetary funds or autonomous agencies relevant to central government policies or under the central authorities’ effective control.
Subnational governments consist of the budgetary and extrabudgetary activities of decentralized governments operating only in parts of the country, such as regional, state, and local governments. The central and subnational governments together constitute the general government.
Box 2.Key Differences between the 2001 and 1986 GFSMs
The 2001 edition of the Government Finance Statistics Manual (GFSM) differs from the 1986 edition in three main ways:
Accrual basis. Unlike GFSM 1986, the 2001 manual emphasizes recording fiscal statistics on an accrual basis—at the time the economic event occurs, not necessarily when cash is paid or received.
Integration of stocks and flows.The GFSM 2001 is underpinned by a set of well-defined relationships between flows and stocks. Specifically, the government’s opening and closing balance sheets are reconciled with the flows derived from government operations and other economic flows (e.g., valuation changes or the extinction of assets from natural disasters).
The analytic framework. This consists of three main tables and four main balances.
The Statement of Government Operations distinguishes among transactions (1) affecting net worth, i.e., revenue and expense; (2) in nonfinancial assets; and (3) in financial assets and liabilities. The difference between revenue and expense is the net operating balance (similar to the current balance in GFSM 1986). Subtracting net acquisition of nonfinancial assets gives the net lending/borrowing balance (similar to the overall balance in GFSM 1986, except for net lending now excluded). This, in turn, is equal to the net acquisition of financial assets minus the net incurrence of liabilities.
The Balance Sheet shows the government’s net worth at the end of the period. It is equal to the stock of nonfinancial assets plus net financial worth (the stock of financial assets minus liabilities). The change in net worth during a year is the sum of changes attributable to revenue and expense transactions and to other economic flows.
The Statement on Sources and Uses of Cash shows purely cash flows associated with revenue and expense transactions and transactions in nonfinancial assets, which yields the cash surplus/deficit. Adding cash flows in financial assets (other than cash) and liabilities to the cash surplus/deficit yields the net change in the stock of cash.
While full implementation of the GFSM 2001 will be challenging for many countries, many of its benefits can be reaped in the interim just by presenting existing fiscal data in this new framework.
Social security funds either form their own subsector or are part of the level of government at which they operate and are consolidated with either the central or general government.11
Public corporations consist of financial public corporations (FPCs, including the central bank) and nonfinancial public corporations (NFPCs). Consolidating NFPCs with the general government yields the nonfinancial public sector, and adding FPCs constitutes the consolidated public sector.
Fiscal policy can be carried out by different levels of government and through a range of institutions. Normally, it is implemented by entities wholly devoted to the economic functions of government, such as central and local governments. But public corporations (both financial and nonfinancial) can also carry out fiscal policy, typically without being explicitly recorded in the budget. For example, central banks may extend subsidized loans and nonfinancial corporations may not operate at market prices, or may provide social services. Such activities—known as “quasi-fiscal operations” (Mackenzie and Stella, 1996)—can have a fiscal impact comparable with that of more traditional government activities, although they are often difficult to measure.
Determining the specific coverage of the public sector for a given country entails striking a compromise between what is administratively feasible and what is important for fiscal policy. As a rule, fiscal policy should be assessed for policy purposes based on general government plus public corporations—whether financial or nonfinancial—that pose a significant risk to public finances. If some levels of government (e.g., local governments) are constrained by the need to run balanced budgets, it may be possible to abstract from them for some analytic purposes.
Whatever the exact coverage of fiscal indicators for policy purposes, fiscal statistics should be compiled both for general government and the public sector. Given that any public corporation, financial and nonfinancial, is a potential source of fiscal risk, the operations of all public corporations should be reported and monitored. This will improve transparency and accountability and also help identify any emerging problems.
When to Record Government Transactions
Governments generally record transactions on four bases: commitment, accrual, due-for-payment, and cash. Under the commitments basis, transactions are recorded when commitments to them have been entered into (usually when purchase orders are issued); under the accrual basis, transactions are recorded at the time of the economic event (e.g., when ownership of goods changes or services are provided); under a due-for-payments basis, at the latest time they can be paid for without incurring additional charges or penalties (or, if sooner, when the cash payment is made); and under the cash basis, when cash is received or disbursed.
Many countries, especially those with weaker public expenditure management systems, use a cash (or modified cash) basis. While operationally easier, it can distort the analysis of fiscal policy (e.g., its impact on aggregate demand and fiscal sustainability). This is because it records transactions after they have occurred economically and ignores noncash transactions, such as arrears and grants-in-kind.
Information on cash flows will still be important, even with accrual accounting—for example, to manage government liquidity and to assess the impact on monetary variables. This does not require a cash basis of recording, but a separate statement on cash flows. Cash-based systems can also be modified to allow some items to be recorded on a noncash basis (e.g., recording interest on a due rather than paid basis).
Main Fiscal Indicators
The fiscal position should be assessed in terms of both flow indicators, such as the overall balance, and stock indicators, such as the amount of government debt.
The overall fiscal balance is the most common fiscal indicator. It is the difference between total revenue (including grants) and total expenditures plus lending minus repayments. The widespread use of the overall balance reflects its links to the government’s net financing requirements and to the external current account.12
An adjusted overall fiscal balance is an overall balance excluding such items as grants or revenue from certain enclave activities (e.g., the oil sector), or certain lumpy expenditure items.
External grants are included under total revenue and grants because they do not add to debt and may finance expenditures that would otherwise not take place. But grants are often volatile, unpredictable, and outside government control. Moreover, they may not last and do not reduce domestic demand. Thus, when grants are significant, the overall balance is often reported both with and without grants. The balance excluding grants also indicates the extent of grant dependency.
Box 3.Fiscal Policy and Nonrenewable Resources
Having nonrenewable resources, such as oil, should be a blessing, but experience suggests it is often a curse. It is a blessing in that exploiting such resources relaxes the traditional obstacles to growth (foreign exchange, domestic savings, and fiscal revenues), but it is a curse in that many such countries seem to suffer from excessive dependence on these resources.
Heavy dependence on nonrenewable resource revenue complicates fiscal policy. Nonrenewable resource revenues are typically volatile and uncertain—linking spending to such revenue induces macroeconomic volatility and reduces its quality. As the resource is by definition nonrenewable, it will, sooner or later, run out—consuming all the revenue now may mean difficult fiscal consolidation later. The influx of foreign exchange can appreciate the real exchange rate, shrinking traditional tradable sectors (“Dutch disease”). And the large windfalls can foster corruption and undermine governance. That said, many, especially low-income, nonrenewable-resource-producing countries face large and pressing social and development needs.
There is no “silver bullet” to fix these problems, and no substitute for prudent macroeconomic management. But research suggests some approaches can help (Davis, Ossowski, and Fedelino, 2003).
Adjust expenditure gradually. Setting expenditure (or, more precisely, the primary balance excluding nonrenewable resource revenue as a ratio to GDP excluding the nonrenewable resource sector) in a medium-term framework will help avoid procyclicality. Countries with stronger financial positions have more leeway to increase spending in response to more permanent price rises, whereas those with weaker positions have less scope to finance downturns.
Save part of nonrenewable resource revenues. Fiscal policy should be set so that no abrupt consolidation is needed when production wanes. This typically requires saving a significant share of the nonrenewable resource revenues, especially when production is not expected to last long. Financial savings should be held offshore to avoid imparting volatility to the domestic economy and excessive real appreciation (some real appreciation may be an intrinsic part of the process of absorbing resource revenues).
Be transparent. Governments should establish a clear legal and regulatory framework for the nonrenewable resource sector and comprehensively disclose related revenue data. International and national nonrenewable resource companies should comply with international accounting standards.
Nonrenewable resource funds should be transparent and well integrated with the budget. A well-designed fund, such as Norway’s State Petroleum Fund, can help manage nonrenewable resource revenues, mainly for political economy reasons. A poorly designed fund, however, can complicate fiscal policy management, for example, by creating a dual budget and making cash management inefficient.
Consider hedging. Hedging can help governments make their nonrenewable resource revenue stream more stable and predictable, but this requires institutional capacity and stringent control.
Set domestic nonrenewable resource product prices at international levels. Subsidization is generally inefficient, poorly targeted, and often nontransparent.
Like grants, oil revenue is highly volatile and unpredictable. It is also, however, nonrenewable and consuming it reduces government wealth. Countries that are heavily dependent on oil (or on other nonrenewable resources) should focus on the non-oil balance, ideally as a ratio to non-oil GDP (Box 3). Privatization receipts are also often excluded from the overall balance for similar reasons.
Externally financed project spending may be excluded as it is outside the government’s control, typically has a large import component with consequent limited impact on domestic supply, and is automatically financed (though if financed by lending, it does increase debt). Excluding external grants, externally financed project spending and external interest payments yields the domestic balance (with specific definitions varying across countries).
The primary balance—revenue minus noninterest (primary) expenditure—is an indicator of fiscal “effort,” in that interest payments are predetermined by the size of previous deficits; the primary balance is a critical variable for debt sustainability analysis. The debt-stabilizing primary balance is the primary balance necessary to keep the debt-to-GDP ratio stable.
The operational balance (the overall balance minus the part of debt service that compensates debt holders for inflation) is often reported when there is high inflation.
The current balance, the difference between current revenue and current expenditures, indicates the extent of government savings. Targeting the current balance could help safeguard investment in times of fiscal consolidation, while leaving the net worth of the government unaffected and promoting intergenerational equity. But focusing exclusively on the current balance is risky (IMF, 2004a) for the following reasons:
Investment spending adds to aggregate demand.
In countries where financing is constrained, there is little alternative to focusing on the overall balance. Indeed, if amortizations are large, fiscal targets may have to be set in the light of total, rather than net, financing availability.
Borrowing may need to be constrained because of longer-term debt sustainability concerns.
Debt can become unsustainable if public investment projects are not of high quality. Even when they are, governments may not be able to realize the fiscal dividends of growth (i.e., public investment does not typically pay for itself).
Other uses of public funds—notably reducing tax rates or investment in human capital and operations and maintenance—may have a higher rate of return than public investment. Excluding public investment from fiscal targets would create a bias against these choices and may also discriminate against private sector involvement in infrastructure.
Focusing on the current balance may invite creative accounting, with a view to classifying current expenditure as capital.
The cyclically adjusted balance13 measures the fiscal position net of the impact of output effects on the budget (IMF, forthcoming). It is obtained by removing the cyclical component of the budget from the nominal fiscal balance. The cyclical component in turn depends on the size of the output gap and on the output elasticity of the budget (determined by the extent that individual budgetary items react to fluctuations in output, as well as by the size of the budget).
Budgetary targets are seldom framed in cyclically adjusted terms. This partly reflects the relative complexity of the techniques used to estimate output gaps and budgetary elasticities. But while the computation of cyclically adjusted balances is fraught with difficulties, a variety of measures can be undertaken to address them.14 Cyclically adjusted balances can play a useful role as a reference for policy design and implementation.
The augmented balance is the overall balance, including such exceptional outlays as the fiscal costs of bank recapitalization or enterprise restructuring not otherwise captured in expenditure. Such outlays need to be financed and may have a significant impact on aggregate demand.
Gross financing needs and sources helps focus on liquidity issues. Needs comprise the overall deficit, any other transactions that require financing, plus amortization. This is equal to the necessary financing, from which the implications of meeting this need can be drawn. For example, if the necessary financing entails a level of external market access much greater than the government had previously enjoyed, this could portend liquidity problems.
Fiscal policy analysis has traditionally focused on such flow variables as the overall balance. But flows are changes in stocks, and stocks are increasingly seen as important yardsticks for gauging fiscal policy in their own right. Reconciling flows and stocks also serves as a consistency check on the quality of fiscal data. The starting point is the public (or government) balance sheet at a given moment. It comprises, on one side, public assets, both financial (e.g., government deposits) and nonfinancial (e.g., roads), and on the other, public liabilities (e.g., government debt) and net worth (the difference between assets and liabilities).
Liabilities, the bulk of which typically consists of debt, are a commonly used stock variable. Large or growing liabilities (typically measured against a scalar, such as GDP or revenue) may signal future debt-servicing problems.
Net financial worth recognizes that liabilities may be matched by financial assets, and thus looking just at liabilities may misrepresent the government’s financial position.15 But only liquid financial assets can be used to meet liabilities coming due. For example, government equity in public corporations may not be easily sold, and withdrawing government deposits from public banks may precipitate their collapse. In addition, netting offsetting claims/liabilities within the public sector against each other may mask mismatches, such as the central government’s inability to service its debt to the social security fund.
Net worth goes further by recognizing that liabilities may also be matched, not just by financial assets, but also by nonfinancial assets. This is the most demanding stock variable to measure as it involves, for example, valuing nonfinancial assets.
The balance sheet can miss potentially critical aspects of the government’s financial position and should be supplemented with additional information. For example:
Contingent liabilities, such as loan guarantees (see the discussion in Section I on Fiscal Adjustment to Reduce Vulnerability), are frequently entered into for fiscal policy purposes, have important economic effects, and can threaten fiscal sustainability. The nominal amount, and the nature, of such liabilities should be published, as should, if possible, an estimate of their expected cost. Where guarantees are significant, budgets should limit their amount and include provisions for their expected cost.
Public-private partnerships are essentially leasing arrangements in which the private sector provides initial finance and the state retains ownership and bears certain future costs. While public-private partnerships can elicit additional financing for investment and may increase efficiency, they may also (possibly deliberately) conceal fiscal activity and expose governments to greater (often hidden) costs than direct procurement of public works. To improve their transparency and assessment, governments should include information on public-private partnerships in budget documents and in end-year financial reports.