3. Economic Flows, Stock Positions, and Accounting Rules

Sage De Clerck, and Tobias Wickens
Published Date:
March 2015
  • ShareShare
Show Summary Details

This chapter describes the economic flows and stock positions of the government finance statistics framework and the accounting rules used to determine all aspects of their recording.


3.1 Entries recorded in GFS are either economic flows or stock positions.1 Flows are monetary expressions of economic actions and effects of events that result in changes in economic value within a reporting period. Stock positions measure economic value at a point in time. More specifically, stock positions refer to a unit’s holdings of assets and liabilities at a specific time and the unit’s resulting net worth, equal to total assets minus total liabilities.

3.2 The flows and stock positions recorded in the GFS framework are integrated, which means that all changes in stock positions can be fully explained by the flows. In other words, the following relationship is valid for each stock position:

where S0 and S1 represent the values of a specific stock position at the beginning and end of a reporting period, respectively, and F represents the net value of all flows during the period that affected that particular stock. More generally, the value of any stock position held by a unit at a given time is the cumulative value of all flows affecting that stock position that have occurred since the unit first acquired the particular type of asset or liability.

3.3 The GFS framework includes a great diversity of economic flows. This chapter first describes several important characteristics of economic flows that underlie their classification and treatment. It then describes in a general way the accounting rules used for recording these flows and stock positions in GFS. Descriptions of specific categories of flows and stock positions and the application of the general rules to their recording are discussed in Chapters 5 through 10.

Economic Flows

3.4 Economic flows reflect the creation, transformation, exchange, transfer, or extinction of economic value; they involve changes in the volume, composition, or value of a unit’s assets, liabilities, and net worth. A flow can be a single event, such as the purchase of goods, or the cumulative value of a set of events occurring during a reporting period, such as the continuous accrual of interest expense on a government bond. All flows are classified as transactions or as other economic flows. The following sections describe these two types of economic flows.


3.5 A transaction is an economic flow that is an interaction between institutional units by mutual agreement or through the operation of the law, or an action within an institutional unit that is analytically useful to treat like a transaction, often because the unit is operating in two different capacities.2 This definition of a transaction stipulates that an interaction between institutional units occurs by mutual agreement. Mutual agreement means that there was prior knowledge and consent by the units, but it does not mean that both units entered into the transaction voluntarily. Some transactions, such as the payment of taxes, are imposed by force of law. Although individual units are not free to determine the amount of taxes they pay, there is collective recognition and acceptance by the community of the obligation to pay taxes. Thus, payments of taxes are considered transactions despite being compulsory. Similarly, the actions necessary to comply with judicial or administrative decisions may not be undertaken voluntarily, but they are taken with prior knowledge and consent of the parties involved.

3.6 The treatment of some activities in GFS takes a different perspective from the treatment of the same activities in the 2008 SNA. GFS is focused on the impact of economic events on the finances of government. In contrast, the 2008 SNA is focused on measuring production, consumption, distribution of income, and investment. Appendix 7 contains a complete description of the implications of these different perspectives. Despite different treatments of some activities, both frameworks include all flows that change stock positions so that all changes in the balance sheet can be explained by the flows.

3.7 Transactions may take on many different forms. In GFS, all transactions are classified according to their economic nature, while transactions in expenditure are also classified according to their functions (see Chapters 5, 6, 8, and the annex to Chapter 6). To give more precision to the classification of transactions, the characteristics of transactions have to be systematically described.

Monetary transactions

3.8 A monetary transaction is one in which one institutional unit makes a payment (receives a payment) or incurs a liability (acquires an asset) to (from) another institutional unit stated in units of currency. In GFS, all flows are recorded in monetary terms, but the distinguishing characteristic of a monetary transaction is that the parties to the transaction express their agreement in monetary terms. For example, goods or services are usually purchased or sold at a given number of units of currency per unit of the good or service, social security benefits are often payable in fixed amounts of currency, and taxes receivable are measured and payable in units of currency. All monetary transactions are interactions between two institutional units, recorded as either an exchange or a transfer.

3.9 An exchange is a transaction in which one unit provides a good, service, asset, or labor to a second unit and receives a good, service, asset, or labor of the same value in return.3 Compensation of employees, purchases of goods and services, the incurrence of interest expense, and the sale of an office building are all exchanges.

3.10 A transfer is a transaction in which one institutional unit provides a good, service, or asset to another unit without receiving from the latter any good, service, or asset in return as a direct counterpart. This kind of transaction is also referred to as being unrequited, a “something for nothing” transaction, or a transaction without a quid pro quo. Transfers can also arise where the value provided in return for an item is not economically significant or is much below its value. Typically, general government units engage in a large number of transfers, which may be compulsory or voluntary. Taxes and most social security contributions are compulsory transfers imposed by government units on other units. Subsidies, grants, and social assistance benefits are transfers from general government units to other units. Public corporations are, to a lesser extent, involved in transfers—they may receive subsidies or capital transfers from government and may also be involved in transfers payable resulting from their quasi-fiscal activities.

3.11 Some transactions appear to be exchanges but are actually combinations of an exchange and a transfer. In such cases, the actual transaction should be partitioned and recorded as two transactions, one that is only an exchange and one that is only a transfer. For example, a general government unit might sell an asset at a price that is clearly less than the market value of the asset, or may buy an asset at a price that is clearly above the market value of the asset. The transaction should be divided into an exchange at the asset’s market value and a transfer equal in value to the difference between the actual transaction value and the market value (see paragraph 3.107) of the asset.4

3.12 Exchange transactions do not include entitlements to collective services or benefits—these are considered transfers. The amount of collective service or benefit that may eventually be receivable by an individual unit is not proportional to the amount payable. Taxes and nonlife insurance premiums are examples of such transactions classified as transfers due to the collective nature of the benefits (see paragraphs 5.23 and 5.149, respectively).

3.13 Taxes are treated as transfers even though the units making these payments may receive some benefits from services provided by the government unit receiving the taxes. For example, in principle, no one can be excluded from sharing in the benefits provided by collective services such as public safety. In addition, a taxpayer may even be able to consume certain individual services provided by government units. However, it is usually not possible to identify a direct link between the tax payments and the benefits received by individual units. Moreover, the value of the services received by a unit usually bears no relation to the amount of the taxes payable by the same unit.

3.14 Nonlife insurance premiums and claims are also treated as transfers in GFS.5 This type of premium entitles the units making the payment to benefits only if one of the events specified in the insurance contract occurs. That is, one unit pays a second unit for accepting the risk that a specified event may occur to the first unit. These transactions are considered transfers6 because in the nature of the insurance business, they distribute income between policyholders to those who claim, as opposed to all policyholders who contribute. There is uncertainty whether the contributing unit will receive any benefits and, if it does receive benefits, they may bear no relation to the amount of the premiums previously paid. Nonlife insurance includes social security schemes and employer social insurance schemes for government employees that do not provide retirement benefits. Thus, social security contributions receivable and social security benefits payable by government units, which are not for employment-related pensions, are treated as transfers in GFS.

3.15 Transfers may be either current or capital. To distinguish current transfers from capital transfers, it is preferable to focus on the special characteristics of capital transfers.

3.16 Capital transfers are transfers in which the ownership of an asset (other than cash or inventories) changes from one party to another, or that oblige one or both parties to acquire or dispose of an asset (other than cash or inventories), or where a liability is forgiven by the creditor. Cash transfers involving disposals of noncash assets (other than inventories) or acquisition of noncash assets (other than inventories) are also capital transfers. A capital transfer results in a commensurate change in the stock position of assets of one or both parties to the transaction. Capital transfers are typically large and infrequent, but capital transfers cannot be defined in terms of size or frequency. A transfer in kind without a charge is a capital transfer when it consists of: the transfer of ownership of a nonfinancial asset (other than inventories); and the forgiveness of a liability by a creditor when no corresponding value is received in return. Major nonrecurrent payments in compensation for accumulated losses or extensive damages or serious injuries not covered by insurance policies are also capital transfers. A transfer of cash is a capital transfer when it is linked to, or conditional on, the acquisition or disposal of an asset by one or both parties to the transaction.

3.17 Current transfers consist of all transfers that are not capital transfers. Current transfers directly affect the level of disposable income and influence the consumption of goods or services. That is, current transfers reduce the income and consumption possibilities of the donor and increase the income and consumption possibilities of the recipient. For example, social benefits, subsidies, and food aid are current transfers.

3.18 It is possible that some cash transfers may be regarded as capital by one party to the transaction and as current by the other party. So that a donor and a recipient do not treat the same transaction differently, a transfer should be classified as capital for both parties even if it involves the acquisition or disposal of an asset, or assets, by only one of the parties. When there is doubt about whether a transfer should be treated as current or capital, it should be treated as a current transfer.

Nonmonetary transactions

3.19 Nonmonetary transactions are transactions that are not initially stated in units of currency. These include all transactions that do not involve any cash flows, such as barter, in-kind transactions, and certain internal transactions. They must be assigned a monetary value as GFS record flows and stock positions expressed in monetary terms. The entries therefore represent values that are indirectly measured or otherwise estimated. The values assigned to nonmonetary transactions have a different economic implication than do monetary payments of the same amount, as they are not freely disposable sums of money. Nevertheless, to have a comprehensive and integrated set of accounts, it is necessary to assign the best estimate of market values to the items involved in nonmonetary transactions.

3.20 Nonmonetary transactions can be either two-party transactions or actions within an institutional unit that are used to construct an internal transaction.

Two-party nonmonetary transactions

3.21 These nonmonetary transactions can be exchanges or transfers. Barter, remuneration in kind, and other payments in kind are nonmonetary exchanges. Transfers in kind are nonmonetary transfers.

3.22 In a barter transaction, two units exchange goods, services, or assets other than cash of equal value. For example, a government unit may agree to trade a parcel of land in an industrial area to a private corporation for a different parcel of land that the government will use as a national park. Between nations, governments may trade strategic natural resources for another kind of product or service.

3.23 Remuneration in kind occurs when an employee is compensated with goods, services, or assets other than money. Types of compensation that employers commonly provide without charge or at reduced prices to their employees include meals and drinks, uniforms, housing services, transportation services, and child care services (see paragraph 6.17–6.18).

3.24 Payments in kind other than remuneration in kind occur when any of a wide variety of payments is made in the form of goods and services rather than money. A payment to settle a liability can be made in the form of goods, services, or noncash assets rather than money. For example, a government unit may agree to settle a claim for past-due taxes if the taxpayer transfers ownership of land or fixed assets to the government, or inheritance taxes may be payable by making donations of paintings or other valuables to government.

3.25 Transfers in kind may be used to increase efficiency, or to insure that the intended goods and services are consumed. For example, aid after a natural disaster may be delivered faster and be more effective if it is provided in the form of medicine, food, and shelter instead of money. Also, a general government unit might provide medical and educational services in kind to ensure that the need for those services is met.

Internal transactions

3.26 Internal or intra-unit transactions take place when a single unit acts in two different capacities, and it is analytically useful to record this act as a transaction. The choice of which internal actions to treat as transactions is subjective to the purpose of recording these actions. GFS follows the 2008 SNA by treating consumption of fixed capital as an internal transaction to allow the calculation of the operating costs of government. Similarly, the transfer of materials and supplies from inventories to use of goods and services and other internal changes in inventories are treated as internal transactions (see paragraph 8.46–8.47).7

Rearrangements of some transactions

3.27 Some transactions are not recorded in the form in which they appear to take place. Instead, they are modified in macroeconomic statistics to bring out their underlying economic relationships more clearly. There are three kinds of rearrangements employed in GFS: rerouting, partitioning, and reassignment.


3.28 Rerouting records a transaction as taking place through channels that differ from the actual ones, or as taking place in an economic sense when no actual transactions take place. Rerouting is often required when a unit that is a party to a transaction does not appear in the actual accounting records because of administrative arrangements. Two kinds of rerouting occur:

  • In the first kind of rerouting, a direct transaction between unit A and unit C is recorded as taking place indirectly through a third unit B. For example, if government employees are enrolled in a social security or retirement scheme, accounting records may show the government unit making contributions directly to the social security fund or retirement scheme on behalf of its employees. However, these contributions are part of the compensation of employees and should be recorded as being paid to the employee. In such a case, it is necessary to reroute the payments so that the government is seen as paying the employees, who then are deemed to make payments of the same amount to the social security or retirement scheme (see paragraph 6.19). As a result of the rerouting, these contributions are included as part of the labor cost of government. Rerouting may also be necessary when recording the distribution of profits of fiscal monopolies (see paragraph 5.68).

  • In the second kind of rerouting, a transaction of one kind from unit A to unit B is recorded with a matching transaction of a different kind from unit B to unit A. For example, when a nonresident special purpose entity (SPE) of government borrows abroad for fiscal purposes, transactions should be imputed in the accounts of both the government and the nonresident SPE as if the SPE has extended a loan to government and government has invested the corresponding amount in the SPE (see paragraphs 2.136–2.139). This rearrangement of the transactions reflects government’s involvement in the nonresident SPE, which would otherwise not be captured in government accounts.


3.29 Partitioning records a transaction that is a single transaction from the perspective of the parties involved as two or more differently classified transactions. For example, when a general government unit acquires an asset below or above its current market price, the division of the actual transaction into an exchange and a transfer is an example of partitioning (see paragraph 3.11).


3.30 Reassignment records a transaction arranged by a third party on behalf of others as taking place directly by the two principal parties involved. Reassignment is required when one unit arranges for a transaction to be carried out between two other units, generally in return for a fee from one or both parties to the transaction. In this case, one unit acts as an agent for another unit. In such a case, the transaction is recorded exclusively in the accounts of the two parties engaging in the transaction and not in the accounts of the third party facilitating the transaction. The accounts of the agent show only the fee charged for the facilitation services rendered. For example, reassignment may occur when one government unit collects taxes and then transfers some or all of the taxes to another government unit. In some arrangements of this nature, the collecting unit retains a small portion of the tax collected in return for its collection efforts. The amount retained is treated as the sale of a service by the collecting unit, while the total amount of taxes collected is shown as revenue for the beneficiary government unit. For guidelines on the reassignment or attribution of taxes to collecting or beneficiary governments, see paragraphs 5.33–5.38.

Other Economic Flows

3.31 Other economic flows are changes in the volume or value of assets or liabilities that do not result from transactions. These other economic flows are not transactions because they do not meet one or more of the characteristics of transactions. For example, the institutional units involved may not be acting by mutual agreement, as in the case of an uncompensated seizure of assets, or changes due to natural events, such as an earthquake or a flood. Alternatively, the value of an asset expressed in foreign currency may change as a result of exchange rate changes, or the value of an asset may change due to the passing of time.

3.32 There are two major categories of other economic flows, described as holding gains and losses, and other changes in the volume of assets and liabilities.8

Holding gains and losses

3.33 A holding gain or loss9 is a change in the monetary value of an asset or liability resulting from changes in the level and structure of prices, excluding qualitative or quantitative changes in the asset or liability. Holding gains and losses on assets and liabilities include changes resulting from exchange rate movements. In concept, holding gains and losses are continuously recorded as market prices change.

3.34 A holding gain or loss accrues continuously, purely as a result of holding an asset or liability over time without transforming it in any way. Holding gains/losses can apply to virtually any type of asset or liability, and they may accrue on an asset held for any length of time during the reporting period. (See paragraphs 10.05–10.45 for a complete discussion.)

Other changes in the volume of assets/ liabilities

3.35 Other changes in the volume of assets are any changes in the value of an asset or liability that do not result from a transaction or a holding gain/loss. Other changes in the volume of assets cover a wide variety of specific events. These events are divided into three main categories:10

  • The first category consists of events that involve the appearance or disappearance of economic assets other than by transactions. In other words, certain assets and liabilities enter and leave the GFS balance sheet through events other than by transactions. (See paragraphs 10.48–10.58 for a complete discussion.)

  • The second group consists of the effects of external events—exceptional and unexpected—on the economic benefits derivable from assets and corresponding liabilities. (See paragraphs 10.59–10.75 for a complete discussion.)

  • The final group is made up of changes in classifications. (See paragraphs 10.76–10.84 for a complete discussion.)

Stock Positions

3.36 A stock position is the total holdings of assets and/or liabilities at a point in time. Stock positions are recorded in the balance sheet of the GFS framework (see Chapter 7). The integrated GFS framework shows stock positions at the beginning and end of a reporting period. Stock positions at these two points in time are connected by flows during that period because changes in positions are caused by transactions and other economic flows. In order to discuss stock positions, it is necessary to determine the asset boundary in macroeconomic statistics from which the definition of assets and liabilities is derived. The coverage of assets in GFS is limited to economic assets from which economic benefits may flow to the owners.

Economic Benefits

3.37 Economic benefits arise from owning and using economic assets. The economic benefits of ownership usually include the right to use, rent out, or otherwise generate income, or to sell the asset. Different kinds of economic benefits that may be derived from an asset include:

  • The ability to use assets, such as buildings or machinery, in production

  • The generation of services (e.g., renting out produced assets to another entity)

  • The generation of property income (e.g., interest and dividends received by the owners of financial assets)

  • The potential to sell and thus realize holding gains.


3.38 Two types of ownership can be distinguished in macroeconomic statistics: legal ownership and economic ownership. The legal owner of resources such as goods and services, natural resources, financial assets, and liabilities is the institutional unit entitled by law and sustainable under the law to claim the benefits associated with the resource. Sometimes government may claim legal ownership of a resource on behalf of the community at large. To be recognized in the GFS framework, a resource must have a legal owner, on either an individual or a collective basis.

3.39 The economic owner of resources such as goods and services, natural resources, financial assets, and liabilities is the institutional unit entitled to claim the benefits associated with the use of these resources by virtue of accepting the associated risks. In most cases, the economic owner and the legal owner of a resource are the same. Where they are not, the legal owner has passed responsibility for the risk involved in using the resource in an economic activity to the economic owner as well as associated benefits. In return, the legal owner accepts another package of risks and benefits from the economic owner. In GFS, when the expression “ownership” or “owner” is used and the legal and economic owners are different, the reference should generally be understood to be to the economic owner. Appendix 4 discusses a number of cases where legal and economic ownership are different.

3.40 Sometimes government may claim legal ownership of a resource on behalf of the community at large, such as territorial waters. If so, the benefits also accrue to the government on behalf of the community at large. Thus, government is both the legal and economic owner of these resources. However, governments may share the benefits with other entities but, by virtue of accepting the majority of the risks, become the economic owner of a resource. For example, in the case of public-private partnerships, economic ownership can be vested with government when government accepts the majority of the risks (see paragraphs A4.58–A4.65).

3.41 The benefits inherent in financial assets and liabilities are seldom transferred from a legal owner to an economic owner in exactly the same state. They are usually transformed to new forms of financial assets and liabilities by the intermediation of a financial institution that assumes some of the risk and benefits while passing the financial instrument on to other units.

Definition of Assets and Liabilities

3.42 An asset is a store of value representing a benefit or series of benefits accruing to the economic owner by holding or using the resource over a period of time. It is a means of carrying forward value from one reporting period to another.

3.43 Only economic assets are recorded in the macroeconomic statistical systems (i.e., included within the asset boundary) and they appear in the balance sheet of the unit that is the economic owner of the asset. Economic assets are resources over which ownership rights are enforced and from which economic benefits may flow to the owners. Personal attributes such as reputation or skill, which are sometimes described as an asset, are not recognized as such in GFS because they are not economic assets as defined earlier. A distinction is made between nonfinancial and financial assets. All financial assets have liabilities as counterparts, except for gold bullion held as a reserve asset, which, by convention, is a financial asset (7.128).

Financial assets and liabilities

3.44 A particularly important mechanism in the economy is the device whereby one economic unit exchanges a particular set of benefits with another economic unit for future payments. From this a financial claim, and hence a liability, can be defined. There are no nonfinancial liabilities recognized in the GFS framework, and thus the term liability necessarily refers to a liability that is financial in nature.11

3.45 A liability is established when one unit (the debtor) is obliged, under specific circumstances, to provide funds or other resources to another unit (the creditor). Normally, a liability is established through a legally binding contract that specifies the terms and conditions of the payment(s) to be made and payment according to the contract is unconditional. These are typically established through the provision of economic value by one institutional unit, the creditor, to another, the debtor, normally under a contractual arrangement. Liabilities can also be created by the force of law, and by events that require future transfer payments. In many cases, liabilities (and their corresponding financial claims) are explicitly identified by formal documents expressing the debtor-creditor relationship. In other cases, liabilities are imputed to reflect the underlying economic reality of a transaction, such as the creation of a notional loan when an asset is acquired under a financial lease.

3.46 Liabilities created by the force of law include those arising from taxes, penalties (including penalties arising from commercial contracts), and judicial awards at the time they are imposed. Liabilities established by events that require future transfer payments include claims on nonlife insurance companies, claims for damages not involving nonlife insurance companies, and claims arising from winnings from lottery and gambling activities.

3.47 Whenever a liability exists, the creditor has a corresponding financial claim on the debtor. A financial claim is an asset that typically entitles the owner of the asset (the creditor) to receive funds or other resources from another unit, under the terms of a liability. Like liabilities, financial claims are unconditional. A financial claim provides benefits to the creditor, such as by acting as a store of value, or by generating interest, other property income, or holding gains. Financial claims consist of equity and investment fund shares, debt instruments, financial derivatives and employee stock options, and monetary gold in the form of unallocated gold accounts (see paragraphs 7.15, 7.127, and 7.139).

3.48 Financial assets consist of financial claims and gold bullion held by monetary authorities as a reserve asset. For a complete discussion of financial assets and liabilities, see paragraphs 7.118–7.227.

3.49 This Manual follows the 2008 SNA by not treating guarantees other than derivatives and the provision for calls under standardized guarantee schemes as financial assets or liabilities. However, it is recommended to report these guarantees as memorandum items to the balance sheet. (See paragraphs 4.48 and 7.251–7.261.)

Nonfinancial assets

3.50 Nonfinancial assets are economic assets other than financial assets. Nonfinancial assets are further subdivided into those that are produced (fixed assets, inventories, and valuables) and those that are nonproduced (land, mineral and energy resources, other naturally occurring assets, and intangible nonproduced assets). For a complete description of the nature of nonfinancial assets, see paragraphs 7.34–7.117.

Accounting Rules

3.51 All entries in GFS have to be measured in monetary terms. In some cases, the amounts entered are the actual payments that form part of flows, and in other cases the amounts entered are estimated by reference to monetary values. Money is thus the unit of account in which all stocks and flows are recorded.

3.52 In principle, a reporting period can cover any period of time. Periods that are too short have the disadvantage that statistical data are influenced by incidental factors, while long periods may not adequately portray changes in the economy in a timely manner. Merely seasonal effects can be avoided by having the reporting period cover a whole cycle of regularly recurrent economic phenomena. In general, calendar years, financial years, and quarters are well suited for drawing up a complete set of GFS for the consolidated general government or public sectors, while monthly data with the broadest institutional coverage possible provide a good high-frequency indicator of fiscal performance. Country-specific circumstances will influence the coverage, frequency, and periodicity of fiscal reporting. However, in deciding these, data reporting guidelines and standards, such as the General Data Dissemination System (GDDS), Special Data Dissemination Standard (SDDS), SDDS Plus, and Code on Good Practices on Fiscal Transparency, should also be considered.12

3.53 The GFS framework is well suited to cover all economic activities in such a way that it is possible to compile GFS statements for individual units, groups of units, or all units in the general government or public sectors. To permit this, the accounting rules for recording flows and stock positions in the GFS framework are designed to ensure consistency in the data generated, and to conform to accepted standards for compiling other macroeconomic statistics. With the exception of consolidation, as noted later in this chapter, the accounting rules of the GFS framework are the same as those of the 2008 SNA (see Appendix 7). There are also many similarities between the rules used in the GFS framework and those applied by businesses and governments in their financial statements.13 The following sections describe the type of accounting system used, the accounting rules governing topics such as the time of recording, and the valuation of flows and stock positions.

Type of Accounting System

3.54 The recording of economic events underlying GFS derives from general bookkeeping principles. Double-entry recording is used for recording all flows. In a double-entry system, each transaction gives rise to at least two equal-value entries, traditionally referred to as a credit entry and a debit entry. This principle ensures that the total of all credit entries and that of all debit entries for all transactions are equal, thus permitting a check on consistency of the GFS accounts for a unit, subsector, or sector. Other economic flows also lead to debit and credit entries. These flows have their corresponding entries directly in changes in net worth. As a result, double-entry recording ensures the fundamental identity of a balance sheet—that is, the total value of assets equals the total value of liabilities plus net worth.

3.55 A debit entry is an increase in an asset, a decrease in a liability, or a decrease in net worth. A credit entry is a decrease in an asset, an increase in a liability, or an increase in net worth. Revenue entries result in an increase in assets or decrease in liabilities, which ultimately increase net worth; therefore, revenue entries are recorded as credits. Conversely, expense entries result in a decrease in assets or increase in liabilities, which ultimately decrease net worth; therefore, expense entries are recorded as debits. Other economic flows can increase or decrease assets and liabilities, thereby directly impacting net worth. In the case of the reclassification of assets or liabilities, a change occurs in the stock positions of two categories of assets or liabilities with no impact on net worth (e.g., an increase in one category of asset is paired with a decrease in another category of asset).

3.56 A balance sheet is a statement of the values of the stock positions of assets owned and of the liabilities owed by an institutional unit or group of units, drawn up in respect of a particular point in time. The fundamental identity of the balance sheet and of accounting in general is that the total value of the assets always equals the total value of the liabilities plus net worth. Use of the double-entry recording ensures that this identity is maintained. There are several possible combinations of debits and credits affecting assets, liabilities, and net worth. For example, the purchase of a service by a general government unit with payment to be made in 30 days would be recorded on an accrual basis as an expense (debit) and an increase in the liability, other accounts payable (credit). Thus, net worth, through the expense, decreases by the same amount that liabilities increase, and assets are not affected. The subsequent payment at the end of the 30 days would be recorded on an accrual basis as a decrease in currency and deposits (credit) and a decrease in other accounts payable (debit). In this case, both assets and liabilities decrease by the same amount and net worth is unaffected.

Time of Recording Flows

3.57 Once a flow has been identified, the time at which it occurred must be determined so that the results of all flows within a given reporting period can be reported. Although this section is concerned with the time assigned to flows, the integrated nature of the GFS framework means that the stock positions recorded on the balance sheet are also influenced by the timing of flows.

3.58 One of the problems in determining the timing of transactions is the frequent existence of a long period between the initiation of an action and its final completion. For instance, many purchases of goods commence with the signing of a contract between a seller and a buyer, followed by the initiation and completion of production of the item ordered, shipment from the seller’s location, arrival at the buyer’s location, preparation and mailing the invoice, receipt of the invoice, approval of payment, the beginning of interest accruing on a late payment or the expiration of a discount for prompt payment, signing a check for payment, mailing of the check by the buyer, receipt of the check by the seller, deposit of the check in the seller’s bank, and finally the check is paid by the buyer’s bank. Even then, the transaction may not be complete as there may be rights of return or warranty claims. Each of these distinct moments is to some extent economically relevant and may result in multiple transactions being recorded in GFS, but only one time can be attributed to each transaction.

3.59 Similarly, in analyzing government expense and acquisition of nonfinancial assets, one can distinguish the day that a budget is voted upon by the legislature, the day on which the ministry of finance authorized a department to pay out specified funds, the day a particular commitment is entered into by the departments, the day deliveries take place, and finally, the day payment orders are issued and checks are paid. With regard to taxes, for example, important moments are the day or the period in which the liability arises, the moment the tax liability is definitively assessed, the day that it becomes due for payment without penalty, and the day the tax is paid or refunds are made.

3.60 In summary, when using the accrual basis of recording, transactions are recorded when economic ownership changes hands for goods, nonproduced nonfinancial assets and financial assets and liabilities, when services are provided, and for distributive transactions when the related claims arise. On the other hand, when using the cash basis of recording, flows are recorded when cash is received and disbursed. These alternative recording bases are discussed in more detail in paragraphs 3.61–3.68.

Alternative recording bases

3.61 While making entries for all successive stages discernible within the activities of institutional units may be possible, it could severely overburden compilers, so a choice has to be made. Broadly, the time of recording could be determined on four bases: the accrual basis, the commitments basis, the due-for-payment basis, and the cash basis. In practice though, many variations on these bases of recording may exist. Accounting systems may use a mixed basis of recording; for example, tax revenue may be recorded on a cash basis while other transactions are recorded on an accrual basis.

3.62 In the accrual basis of recording, flows are recorded at the time economic value is created, transformed, exchanged, transferred, or extinguished. In other words, the effects of economic events are recorded in the period in which they occur, irrespective of whether cash was received or paid, or was due to be received or paid. Nevertheless, the time at which the economic events occur is not always clear. In general, the time attributed to events is the time at which economic ownership of goods changes, services are provided, the obligation to pay taxes is created, the claim to a social benefit payment is established, or other unconditional claims are established.

3.63 If an economic event requires a subsequent cash flow, such as purchases of goods and services on an installment plan, then the length of time between the time attributed to an event with the accrual basis and the time of the cash flow is bridged by recording other accounts receivable/payable. For example, when a general government unit purchases goods on credit, it records a debit to an inventory account and a credit to other accounts payable when ownership of the goods shifts. When the cash payment is made, the general government unit records a debit to other accounts payable and a credit to currency and deposits.

3.64 All events that result in the creation, transformation, exchange, transfer, or extinguishment of economic value are recorded on an accrual basis in the GFS Statement of Operations or Statement of Other Economic Flows (see paragraphs 4.16–4.31 and 4.36–4.38). Thus, all monetary and nonmonetary transactions are included in statistics compiled on the accrual basis.

3.65 In the commitments basis of recording, flows are recorded when an institutional unit has committed itself to a transaction. This basis often applies only to purchases of assets, goods, services, and to compensation of employees. The time of recording generally is when approval for the purchase was issued, which results in funds being earmarked for a specific transaction. Flows for which the commitments basis is not applicable, such as revenue, must be recorded using one of the other three bases of recording. Inkind transactions may or may not be recorded.

3.66 In the due-for-payment basis of recording, flows that give rise to cash payments are recorded at the latest times they can be paid without incurring additional charges or penalties or, if sooner, when the cash payment is made.14 The period of time (if any) between the moment a payment becomes due and the moment it is actually made is bridged by recording other accounts receivable/payable, just as with the accrual basis. If a payment is made before it is due, then no receivable is necessary. Although due-for-payment recording furnishes a more comprehensive description of monetary flows than cash accounting, recording is limited to monetary flows and therefore does not capture all economic events.

3.67 In the cash basis of recording, flows are recorded when cash is received or disbursed. The cash basis of recording provides analytically useful information on the liquidity position of government, which allows for liquidity management. All events resulting in a cash flow are recorded in the GFS Statement of Sources and Uses of Cash (see paragraphs 4.32–4.35). Nonmonetary flows are not recorded, since no cash flows are involved in these transactions. Therefore, the cash basis of recording does not fully record all economic activity and resource flows.

3.68 Cash flows determine the government’s ability to pay its bills and, by influencing the liquidity of the community, activate or validate the demand for goods and services in the rest of the economy. Traditionally, payments data form the basis for most government accounting systems and often represent the most readily available estimates generated by the administrative accounting system. For instance, data on payments for the delivery of goods and services may frequently be more readily available than data on the time of delivery. But while statistics based on cash flows are of analytical value in their own right they do not fully meet the government’s reporting needs. Therefore, the statistics on the accrual basis should be seen as necessary for the proper recording of all governments’ operations.

Using the accrual basis of recording in the Statements of Operations and Other Economic Flows, and Balance Sheet of the GFS framework

3.69 Using the accrual basis of reporting provides the most comprehensive information because all economic events and resource flows are recorded, including internal transactions, in-kind transactions, and other economic flows. Moreover, it is only this comprehensive recording framework that permits the full integration of all flows with stock positions in the balance sheet. In general, accounts using the commitment, due-for-payment, or cash basis are restricted to monetary transactions.

3.70 The integrated framework of GFS uses the accrual basis, primarily because the time of recording matches the time of the actual resource flows and economic events. As a result, the accrual basis provides the best estimate of the macroeconomic impact of government fiscal policy. With the cash basis, the time of recording may diverge significantly from the time of the economic activities and the transactions to which they relate. For example, the interest accrued on a zero-coupon bond would not be recorded until the bond matures, which could be many years after the expense was incurred. The due-for-payment basis will frequently record transactions after the resource flows have taken place, although the very long delays permitted by the cash basis would, in most cases, be reduced. The timing of the commitments basis will precede the actual resource flow.

3.71 The accrual basis of recording also fully captures all amounts receivable or payable in arrears. Arrears are defined as amounts that are both unpaid and past the due date for payment. Because the due-for-payment date is always the same or later than the date attributed to a flow under the accrual basis, all arrears will be included in statistics compiled using the accrual basis. Without supplementary information, however, it may be difficult to estimate the share of total accounts payable that is in arrears as opposed to the share that exists because of normal payment delays. By definition, the due-for-payment basis will clearly show the arrears arising from purchases on installment plans, but arrears from the failure to repay debt obligations as scheduled, such as loans and debt securities, will not always be apparent without supplemental information. With the commitments basis, the availability of information on arrears will be the same as with the accrual basis. With the cash basis, there is no impact on the accounts when a unit does not pay for purchases made or comply with the terms for the repayment of debt. Thus, there will be no information on arrears unless special compilation effort is made.

3.72 In accrual accounting systems, provision is usually made for a separate statement of cash flows to be prepared as part of the comprehensive suite of statements—implementing accruals accounting systems therefore usually maintains some information on cash flows. Managing liquidity is crucial for the operation of any unit, and information on cash flows helps meet this need. Moreover, it may be difficult to assess solvency and future cash flows without an accrual-based system because information on arrears and on other accounts payable/receivable, such as trade credits and advances, is missing.

3.73 Accounts using the due-for-payment, commitments, or cash basis normally do not differentiate between the time of payment, acquisition, and use of resources. With the accrual basis, acquisitions of nonfinancial assets are recorded separately and the expense of using those assets in operating activities is matched with the period of their use through the consumption of fixed capital.

3.74 Additionally, the other major macroeconomic statistical frameworks (national accounts, balance of payments, and monetary and financial statistics) use the accrual basis. Thus, the consistency of statistics from the different systems is facilitated greatly by sharing the use of the accrual basis for all the macro-economic frameworks.

3.75 Despite the advantages of the accrual basis of recording, its implementation is likely to be more complex than the other bases of recording and may require additional estimates. For example, it may be difficult for a government unit to know the full amount of tax revenue to which it is entitled because these amounts may depend on transactions and other events to which the government is not a party.

Application of the accrual principles

3.76 As a general rule, a flow is recorded under the accrual basis of recording when economic ownership changes or another economic event has occurred. More specific guidelines for the application of the accrual recording basis are described in the following paragraphs.

Time of recording and measurement of taxes and other compulsory transfers

3.77 The general principle is that taxes and other compulsory transfers should be recorded when the underlying activities, transactions, or other events occur that give rise to the liability to pay (i.e., the moment when it creates the government’s unconditional claim to the taxes or other payments) (see paragraphs 5.10–5.20). This time is not necessarily the time at which the event being taxed occurred. For example, the obligation to pay tax on capital gains normally occurs when an asset is sold, not when the asset’s value appreciated.

3.78 Estimating the revenue from taxes and compulsory social insurance contributions must take many uncertainties into account. The primary uncertainty is that the government unit receiving the revenue is usually not a party to the transaction or other event that creates the obligation to pay the taxes or social insurance contributions. Consequently, many of these transactions and events permanently escape the attention of the tax authorities. The amount of revenue from taxes and social insurance contributions should exclude the amounts that possibly could have been received from such unreported events, had the government learned about them. In other words, only those taxes and social insurance contributions that are evidenced by tax assessments and declarations, customs declarations, and similar documents are considered to create revenue for government units.

3.79 In addition, it is typical that some of the taxes and social insurance contributions that have been assessed will never be collected. Uncollectable taxes include amounts deemed uncollectable due to non-compliance with tax laws or insolvency of taxpayers. Taxes should also exclude contested tax assessments, which are treated as contingencies. It would be inappropriate to accrue revenue for an amount that the government unit does not realistically expect to collect. Thus, the difference between assessments and expected collections represents a claim that has no real value and should not be recorded as revenue (see paragraph 5.20). The amount of taxes and social insurance contributions that is recorded as revenue should be the amount that is realistically expected to be collected. The actual collection, however, may be in a later period, possibly much later.

3.80 To ensure that the amount of taxes and social contributions recorded on an accrual basis is equivalent to the corresponding amounts actually received over a reasonable amount of time, the following possibilities for the accrual recording of taxes could be considered:

  • Amounts to be recorded are assessed amounts adjusted by coefficients reflecting the assessments not likely to be collected. The coefficients are estimated on the basis of past experience and current expectations in respect of assessed amounts never collected.

  • Cash amounts are recorded in the accounts, but they are time-adjusted so that they are attributed to the period when the activity took place to generate the liability.

3.81 If taxes are imposed on specific transactions or events, they are recorded at the time the underlying transaction or event occurs, even though these times may not coincide with the actual payment of the tax to the government. This implies that taxes on products and imports are recorded at the time the products in question are produced, imported, or sold, depending on the basis of the taxation. Examples include sales taxes, value-added taxes, import duties, and estate and gift taxes.

3.82 In principle, income taxes and social contributions based on income should be attributed to the period in which the income is earned, even though there may be a significant delay between the end of the reporting period and the time at which it is feasible to determine the actual liability. In practice, however, some flexibility is permitted. In particular, as a practical deviation from the general principle, income taxes deducted at source, such as pay-as-you-earn taxes and regular prepayments of income taxes, may be recorded in the periods in which they are paid, and any final tax liability on income may be recorded in the period in which it is determined.

3.83 Income taxes are normally imposed on the income earned during an entire year. If monthly or quarterly statistics are compiled in the absence of monthly accrual administrative records, indicators of seasonal activity or other appropriate indicators may be utilized to allocate the annual totals.

3.84 Taxes on the ownership of specific types of property may be based on the value of the property at a particular time, but are deemed to accrue continuously over the entire year or the portion of the year that the property was owned, if less than the entire year. Similarly, taxes on the use of goods or the permission to use goods or perform activities usually relate to a specific time period, such as a license to operate a business during a specific period.

3.85 Some compulsory transfers, such as fines, penalties, and property forfeitures, are determined at a specific time. These transfers are recorded when the government has an unconditional legal claim to the funds or property, which usually is when a court renders judgment or an administrative ruling is published. If such judgment or ruling is subject to further appeal, then the time of recording is when the appeal is resolved.

3.86 Determining the time of recording for grants and other voluntary transfers is influenced by a wide variety of eligibility conditions that have varying legal powers. In some cases, a potential grant recipient has a legal claim when it has satisfied certain conditions, such as the prior incurrence of expenses for a specific purpose or the passage of legislation. These transfers are recorded by recipient and donor when all conditions are satisfied. In other cases, the grant recipient never has a claim on the donor, and the transfer should be attributed to the time at which the cash payment is made, or when the goods or services are delivered (see paragraph 5.105).

Time of recording dividends

3.87 Dividends and withdrawals of income from quasi-corporations are distributive transactions for which the time of recording depends on the unit’s decision regarding when to distribute income (see paragraphs 5.111–5.119). The level of dividends is not unambiguously attributable to earnings of a particular period, and dividends are to be recorded as of the moment the associated share starts to be quoted “ex-dividend.” Withdrawals of income from quasi-corporations are recorded on the date the payment actually occurs.

Time of recording transactions in goods, services, and nonfinancial assets

3.88 The time of recording transactions (including by barter, payment in kind, or transfer in kind)15 in goods and nonfinancial assets is, in principle, when economic ownership changes, which depends on the provisions in the sales contract (see paragraphs 8.13–8.17). When change of ownership is not obvious, the time of recording by the transaction partners16 may be a good indication and, failing that, the moment when there is a change in physical possession or control. For example, a change of ownership is imputed to have taken place under a financial lease when the lessee takes control of the asset.

3.89 Transactions in services normally should be recorded when the services are provided. If a service, such as transportation, is provided at a specific time, then the transaction is recorded at that time. Some services are supplied or take place on a continuous basis. For example, insurance, and rental of housing services are continuous flows and should be recorded continuously for as long as they are being provided. More practically, the value of the services attributed to a reporting period is based on the quantity supplied during the period rather than the payments required.

3.90 Several other transactions also relate to flows that take place continuously or over extended periods. For example, operating leases and consumption of fixed capital accrue continuously over the whole period a fixed asset is used and interest17 accrues continuously over the period that the financial claim exists. These flows are recorded as being provided continuously over the whole period the contract lasts or the asset is available for use.

3.91 Inventories may be materials and supplies held as input for producing goods and services, work-in-progress, or finished goods held for resale or distribution. Additions to inventories are recorded when products are purchased, produced, or otherwise acquired. Withdrawals from inventories are recorded when products are sold, used up in production, or otherwise relinquished. Additions to work-in-progress inventories are recorded continuously as work proceeds. When production is completed, the goods valued at costs accumulated to that point are transferred to inventories of finished goods.

3.92Use of goods and services is recorded when the good or service enters the process of production, as distinct from the time it was acquired. For goods, this time may be quite different from the time they were acquired. In the meantime, they are classified as inventories.

Time of recording transactions in financial assets and liabilities

3.93 Transactions in many types of financial assets and liabilities, such as debt securities, loans, currency, and deposits, are recorded when economic ownership changes (see paragraphs 9.13–9.16). This date may be specified according to a contract to ensure matching entries in the books of both parties. If no precise date is fixed, the date on which the creditor receives payment, or some other financial claim, is the determining factor. For example, loan drawings are entered in the accounts when actual disbursements are made and financial claims are established, which is not necessarily when an agreement is signed. On practical grounds, public sector liabilities may have to take account of the time of recording from the viewpoint of the public sector unit.

3.94 In some cases, the parties to a transaction may perceive ownership to change on different dates because they acquire the documents evidencing the transaction at different times. This variation usually is caused by the process of clearing, or the time checks are in the mail. The amounts involved in such a “float” may be substantial in the case of transferable deposits and other accounts receivable or payable. If there is disagreement on a transaction between two units, the date on which the transaction is fully completed, which is the day the creditor regards change of ownership to have taken place, is the date of record; this date could be when the creditor receives payment or some other financial claim.

3.95 Some financial claims or liabilities, in particular the various types of other accounts payable and receivable, such as trade credit and advances, general accounts payable, and wages payable, are the result of a nonfinancial transaction and are not otherwise evidenced. In these cases, the financial claim is created when the counterpart transaction (such as the purchase of a good on credit or provision of labor) occurs.

3.96 For securities, the transaction date (i.e., the time of the change in ownership of the securities) may precede the settlement date (i.e., the time of the delivery of the securities). Both parties should record the transactions at the time ownership changes, not when the underlying financial asset is delivered. Any significant difference between transaction and settlement dates gives rise to other accounts payable or receivable. In practice, when the delay between the transaction and settlement is short, the time of settlement may be considered an acceptable proxy.

3.97 According to the accrual basis of recording, repayments of debts are recorded when they are extinguished (such as when they are paid, or rescheduled, or forgiven by the creditor). When arrears occur, no transactions should be imputed, but the arrears should continue to be included in the same instrument until the liability is extinguished. If the contract provided for a change in the characteristics of a financial instrument when it goes into arrears, this change should be recorded as a reclassification in the other changes in volume of the financial assets and liabilities account. The reclassification applies to situations where the original contract remains, but the terms within it change (e.g., interest rates, repayment periods, etc.).18 If a new contract is negotiated or the nature of the instrument changes from one instrument category to another (e.g., from bonds to equity), transactions should be recorded to reflect the redemption of the old instrument and to create a new instrument.

Time of recording other economic flows

3.98 Other changes in the volume of assets are usually discrete events that occur or accrue at particular moments or within fairly short periods of time (see paragraphs 10.46–10.47). For example, the destruction of an asset by fire happens at a specific time, and the impact of a natural disaster can be allocated to a specific period.

3.99 Changes in prices often have a more continuous character, particularly in respect of assets for which active markets exist. In practice, holding gains or losses will be computed between two points in time. The starting point will be the moment at which:

  • The reporting period begins

  • Ownership is acquired from other units (through purchase or a transaction in kind)

  • An asset is produced.

The end point in time will be the moment at which:

  • The reporting period ends

  • Ownership of an asset is relinquished (through a sale or a transaction in kind)

  • An asset is consumed in the production process.

3.100 Holding gains and losses are not calculated over a period beginning the moment two units agree to a mutual exchange of assets. Instead, the calculation of the holding gains and losses starts when economic ownership of the assets is acquired. The signing of the contract fixes the market price for the transaction. A unit can incur holding gains and losses only on the assets or liabilities over which it has economic ownership. This implies that during the period between the signing of a contract and the date on which the first party delivers, the second party cannot incur any price risks on this contract; the second party neither owns the assets to be delivered nor owns a claim on the first party to be recorded in the financial accounts.19

3.101 Other changes in volume, including reclassifications, are recorded as these changes occur. An integrated stock-flow framework requires that both the removal of an existing asset or liability from the original category and its inclusion in the new category are recorded at the same time.

3.102 Reclassifications should be recorded when the change in the nature of the asset, liability, or entity occurs. Although one might be tempted to stockpile major reclassifications for a number of years and enter them as one block at the end of this period, this procedure does not conform to the accrual principles of GFS, which aim for correct estimates at all times. Keeping records of reclassifications makes it possible to reconstruct supplementary time series based on the situation before the reclassification, if needed.

Using the cash basis of recording in the Statement of Sources and Uses of Cash

3.103 GFS includes a Statement of Sources and Uses of Cash. For this statement, statistics on monetary flows should be based on transactions as close to the payment/receipt stage as possible. These statistics, based on cash payments/receipts, measure in aggregate the government’s impact on the liquidity conditions in the economy (see paragraph 3.67). Although these cash-based data lack an integration of flows with stock positions, they are complementary to the accrual data and form an integral part of the complete GFS framework.

3.104 For expense and the acquisition of nonfinancial assets, data at the stage of payment by cash disbursed, or checks or warrants issued, represent the most desirable basis on which to record the cash-based data.20 For revenue, data representing the tax payments received by government, net of refunds paid out during the period covered should be reported. These data will include taxes paid after the original assessment, taxes paid or refunds deducted from taxes after subsequent assessments, and taxes paid or refunds deducted after any subsequent reopening of the accounts. In the reporting of tax revenue, the use of payment basis data is often the first best estimate for a cash statement.

3.105 In the case of government borrowing, the cash basis of recording will report borrowing when funds are received by government, or when lenders pay government suppliers on behalf of government. Government lending should be recorded when the government makes payment, or when funds are provided to a borrower.

3.106 However, payments data must be adjusted to an accrual basis to permit the measurement of production, income, consumption, capital accumulation, and finance in the national accounts. To reconcile data extracted from cash-based data and data maintained on an accrual basis, the cash flows must be adjusted for accrued revenue not yet received and accrued expense not yet paid, respectively.


General rule

3.107 All flows and stock positions should be measured at market prices. Market prices refer to current exchange value—that is, the value at which goods, services, labor, or assets are exchanged or else could be exchanged for cash (currency or transferable deposits). Flows recorded in the Statement of Operations should be valued at the market prices at which these flows take place, while flows recorded in the Statement of Sources and Uses of Cash should be valued at the monetary value of the cash flows. Stock positions should be valued at the market prices prevailing on the balance sheet date. Valuation of specific types of flows and stock positions are discussed in further detail in the remainder of this section.

Valuation of transactions

3.108 Market prices for transactions are defined as amounts of money that willing buyers pay to acquire something from willing sellers; the exchanges are made between independent parties and on the basis of commercial considerations only, sometimes called “at arm’s length.” Thus, according to this strict definition, a market price refers only to the price for one specific exchange under the stated conditions. A second exchange of an identical unit, even under circumstances that are almost exactly the same, could result in a different market price. A market price defined in this way is to be clearly distinguished from a price quoted in the market, a world market price, a going price, a fair market price, or any price that is intended to express the generality of prices for a class of supposedly identical exchanges rather than the price actually applying to a specific exchange. Furthermore, a market price should not necessarily be construed as equivalent to a free market price; that is, a market transaction should not be interpreted as occurring exclusively in a purely competitive market situation. In fact, a market transaction could take place in a monopolistic, monopsonistic, or any other market structure. Indeed, the market may be so narrow that it consists of a sole transaction of its kind between independent parties.

3.109 When a price is agreed to by both parties in advance of a transaction taking place, this agreed or contractual price is the market price for that transaction regardless of the prices that prevail when the transaction takes place.

3.110 Actual exchange values, expressed in monetary terms, are presumed to be the market prices in most cases. Paragraph 3.122 describes those cases where actual exchange values do not represent market prices. Transactions that involve dumping and discounting represent market prices. Transaction prices for goods and services are inclusive of appropriate taxes and subsidies. A market price is the price payable by the buyer after taking into account any rebates, refunds, adjustments, etc., from the seller.

3.111 Transactions in financial assets and liabilities are recorded at the prices at which they are acquired or disposed of. Transactions in financial assets and liabilities should be recorded exclusive of any service charges, commissions, fees, taxes, and similar payments for services that would be necessary to require the asset or incur the liability. These costs of ownership transfers are excluded regardless of whether these are charged explicitly, included in the purchaser’s price, or deducted from the seller’s proceeds. This is because both debtors and creditors should record the same amount for the same financial instrument. The commissions, fees, and/or taxes should be recorded separately from the transaction in the financial asset and liability, under appropriate categories of revenue or expense. The valuation of financial instruments, which excludes commission charges, differs from the valuation of nonfinancial assets (excluding land), which includes any costs of ownership transfer. Costs of ownership transfer on land are included in the value of land improvements (see paragraphs 8.6–8.8).

3.112 When market prices for transactions are not observable, such as for some barter or transfers-in-kind transactions, valuation according to market price-equivalents provides an approximation to market prices. In such cases, market prices of the same or similar items, when such prices exist, will provide a good basis for applying the principle of market prices. Generally, market prices should be taken from the markets where the same or similar items are currently traded in sufficient numbers and in similar circumstances. If there is no appropriate market in which a particular good or service is currently traded, the valuation of a transaction involving that good or service may be derived from the market prices of similar goods and services by making adjustments for quality and other differences.

Valuation of stock positions

3.113 Stock positions should be valued at market value—that is, as if they were acquired in market transactions on the balance sheet reporting date (reference date). Market prices are readily available for assets and liabilities that are traded in active markets, most commonly certain financial assets and their corresponding liabilities. Market values of other assets and liabilities need to be estimated in a manner similar to nonmonetary flows, as described in paragraphs 3.118–3.125 and 7.20–7.33.

3.114 Valuation according to market-value equivalent is needed for valuing assets and liabilities that are not traded in markets or are traded only infrequently. For these assets and liabilities, it will be necessary to estimate values that, in effect, approximate market prices (see paragraph 3.125).21

3.115 It may also be analytically useful and appropriate, in some circumstances, to use alternative valuation methods and to compare these with market values. Market values, fair values, and nominal values should be distinguished from such notions as amortized values, face values, book values, and historic cost.

  • Fair value is a market-equivalent value defined as the amount for which an asset could be exchanged, or a liability settled, between knowledgeable, willing parties in an arm’s-length transaction. It thus represents an estimate of what could be obtained if the owner sold the asset or the debtor settled the liability.

  • Nominal value at any moment in time is the amount that the debtor owes to the creditor. It reflects the value of the instrument at creation and subsequent economic flows, such as transactions, exchange rate and other valuation changes other than market price changes, and other volume changes. For financial instruments other than debt securities, equity, and financial derivatives, the lack of generally available market values means that these values are estimated by using the nominal value.

  • The amortized value of a loan reflects the gradual elimination of the liability by regular payments over a specified period of time. On the date of each scheduled payment, the amortized value is the same as the nominal value, but it may differ from the nominal value on other dates because the nominal value includes interest that has accrued.

  • The face value of a debt instrument is the undiscounted amount of principal to be repaid at (or before) maturity.22 The use of face value as a proxy for nominal value in measuring the gross debt position can result in an inconsistent approach across all instruments and is not recommended. For example, the face value of deep-discounted bonds and zero-coupon bonds includes interest not yet accrued, which runs counter to accrual principles.

  • Written-down replacement cost is the current acquisition price of an equivalent new asset minus the accumulated consumption of fixed capital, amortization, or depletion.

  • Book value generally refers to the value recorded in the entities’ records. Book values may have different meanings because their values are influenced by accounting standards, rules, and policies, as well as the timing of acquisition, company takeovers, frequency of revaluations, and tax and other regulations.

  • Historic cost, in its strict sense, reflects the cost at the time of acquisition, but sometimes it may also reflect occasional revaluations.

3.116 The valuation of assets and liabilities based on accounting standards may not fully reflect the market prices of the assets and liabilities. In such cases, the source data for GFS should be adjusted to reflect, as closely as possible, the market value of the assets and liabilities.23

3.117 Some financial assets and liabilities, such as bonds, have a nominal value, face value, and market value, and for some purposes, supplementary data on the nominal and face values of stock positions may be useful.24 However, transactions in these assets and liabilities should be valued at the prices actually paid. Similarly, to attain integration between stock positions and flows, the stock positions of debt securities should be valued at their market value when recorded on the balance sheet.

Valuation adjustments in special cases

3.118 When a unit sells an item and does not expect to receive payment, or the corresponding payment is not due for an unusually long time,25 the value of the principal (recorded in other accounts payable/ receivable) is reduced by an amount that reflects the time to maturity using an appropriate discount rate, such as the contractual rate for similar debt instruments. If payment is not due for an unusually long period of time, this reduction is by partitioning the market price of the item purchased, which equals the reduced principal amount, and accrued interest, the assumption being that the amount to be paid includes an allowance for interest. If payment is not expected for an unusually long period of time, such as due to the circumstances of the debtor, a reduction in the principal to be paid is recorded through a valuation change in other accounts payable/receivable, with interest accruing on the reduced principal amount, reflecting the time delay in payment. In both the circumstances described in this paragraph, interest should accrue until payment is made, at the rate used to discount the principal.

3.119 Flows and stock positions expressed in a foreign currency are converted to their value in the domestic currency at the rate prevailing at the moment they are entered in the accounts—that is, the moment the transactions or other flow takes place, and stock positions are converted at the rate prevailing on the balance sheet date. The midpoint between the buying and selling spot rates should be used so that any service charge is excluded. When a multiple exchange rate system is in operation, the valuation should be based on the rate applicable to the type of asset in question. The valuation in the domestic currency of a purchase or sale on credit expressed in a foreign currency may differ from the value in domestic currency of the subsequent cash payment because the exchange rate changed in the interim. Both transactions should be valued at their market values as of the dates they actually occurred, and a holding gain or loss resulting from the change in the exchange rate should be recorded for the period or periods in which it occurs.

3.120 For some transactions in goods, contracts establish a quotation period often months after the goods have changed hands. In such cases, market value at the time of the change of ownership of the goods should be initially estimated, and revised with the actual market value, when known. Market value is given by the contract price even if it is unknown at the time of change of ownership.

3.121 Transfers in kind should be valued at the market prices that would have been receivable if the resources had been sold in the market. In the absence of a market price, the donor’s view of the imputed value of the transaction will often be quite different from that of the recipient. The suggested rule of thumb is to use the value assigned by the donor as the basis of recording.

3.122 In some cases, actual exchange values may not represent market prices. Examples are transactions involving transfer prices between affiliated units, manipulative agreements with third parties, and certain noncommercial transactions. Prices may be under- or overinvoiced, in which case an assessment of a market-equivalent price needs to be made. Although, conceptually, adjustment should be made when actual exchange values do not represent market prices, this may not be practical in many cases. In some cases, transfer pricing may be motivated by income distribution or equity buildups or withdrawals. Replacing book values with market-value equivalents is desirable in principle, when the distortions are large and when the availability of data (such as adjustments by customs or tax officials or from partner economies) makes it feasible to do so. Selection of the best market-value equivalents to replace book values is an exercise calling for cautious and informed judgment. In many cases, compilers may have no choice other than to accept valuations based on explicit costs incurred in production or any other values assigned by the unit.

3.123 While some nonmarket transactions, such as grants in kind, have no market price, other nonmarket transactions may take place at implied prices that include some element of grant or concession so that those prices also are not market prices (see paragraphs 3.10–3.11). Examples of such transactions could include negotiated exchanges of goods between governments and governments’ concessional lending. While there is no precise definition of concessional loans, it is generally accepted that they occur when units lend to other units and the contractual interest rate is intentionally set below the market interest rate that would otherwise apply. The degree of concessionality can be enhanced with grace periods (see paragraph 6.69), frequencies of payments, and a maturity period favorable to the debtor. Since the terms of a concessional loan are more favorable to the debtor than market conditions would otherwise permit, concessional loans effectively include a transfer from the creditor to the debtor. However, except for the case of concessional lending to government employees (see paragraph 6.17 and Chapter 6, footnote 11) and concessional lending by central banks (see Box 6.2), the means of incorporating the impact of concessional lending into GFS have not been fully developed. Accordingly, until the appropriate treatment of concessional debt is resolved, information on concessional debt should be provided as supplementary information (see paragraph 7.246).

3.124 Where a single amount payable/receivable refers to more than one transaction category, the individual flows should be partitioned and recorded separately (see paragraph 3.29). In such a case, the total value of the individual transactions after partitioning must equal the market value of the exchange that actually occurred.

3.125 The value of flows that are not already expressed at their market value, such as barter transactions, must be estimated. In addition, market values for many stock positions will not be readily available and must be estimated. The following list suggests several estimation possibilities. The choice of which method to use in a given circumstance depends on the information available.

  • It may be possible to estimate the values of transactions based on values taken from markets in which similar transactions take place under similar conditions. The value of certain stock positions, primarily financial assets, may also be estimated using market transactions involving similar assets that take place at the end of the reporting period.

  • Flows and stock positions involving existing fixed assets can be valued using the market price for similar new goods, properly adjusted for consumption of fixed capital and other events that may have occurred since they were produced.

  • If there is no appropriate market in which a particular good or service is currently traded, the valuation of a flow involving that good or service may be derivable from the market prices of similar goods and services by making adjustments for quality and other differences.

  • The value of flows and stock positions of assets may be estimated on the basis of the historic or acquisition cost of the item, adjusted for all changes that have occurred since it was purchased or produced, such as consumption of fixed capital, holding gains or losses, depletion, exhaustion, degradation, unforeseen obsolescence, and exceptional losses.26

  • Goods and services can be valued by the amount that it would cost to produce them in the current reporting period. For market producers, the market value of a nonfinancial asset valued in this way should include a mark-up that reflects the net operating surplus attributable to the producer. For nonmarket goods and services produced by general government units or nonprofit institutions serving households (NPISHs), however, no allowance should be made for any net operating surplus in the calculation of the market price.

  • Assets can be valued at the discounted present value of their expected future returns. This method is particularly prominent for a number of financial assets, natural assets, and intangible assets. For some financial assets, the present market value is established by discounting future payments or receipts to the present, using the market interest rate. In principle, therefore, if a reasonably robust estimate of the stream of future earnings to come from an asset can be made, along with a suitable discount rate, this method allows an estimate of the present value. However, it may be difficult to determine the future earnings with the appropriate degree of certainty, given that assumptions are also needed about the asset’s life span and the discount factor to be applied. Because of these uncertainties, the other possible sources of valuation described in the preceding paragraphs should be exhausted before resorting to this method.

Valuation of other economic flows

3.126 Apart from transactions, the change in the value of assets and liabilities between two end-periods also results from holding gains and losses, and other changes in volume of assets and liabilities. The valuation of these other economic flows is discussed in the remainder of this section.

Holding gains and losses

3.127 Holding gains and losses accrue continuously and apply to both nonfinancial and financial assets and liabilities. Since all financial assets, except gold bullion, are matched by liabilities either within the domestic economy or with the rest of the world, it is important that holding gains/losses are recorded symmetrically. A holding gain occurs when an asset increases in value or a liability decreases in value; a holding loss occurs when an asset decreases in value or a liability increases in value. Holding gains and losses during a reporting period are shown separately for assets and liabilities. In practice, the value of holding gains and losses is calculated for each asset and liability between two points in time: the beginning of the period or when the asset or liability is acquired or incurred, and the end of the period or when the asset or liability is sold or extinguished.

Other changes in the volume of assets

3.128 In order to determine the valuation of the other changes in the volume of nonfinancial assets, it is usually necessary to determine the market value of the asset before and after the economic event, such as its appearance, disappearance, catastrophic loss, or reclassification (see paragraphs 10.46–10.84). The value of the other change in volume is calculated as the difference in the market value of the asset immediately before and after the event.

3.129 Other changes in the volume of financial assets and liabilities are recorded at the market or market-equivalent prices of similar instruments. For writing-off of marketable financial instruments that are valued at their market values, the value recorded in the other changes in the volume of asset account should correspond to their market values prior to being written off. For nonmarketable financial instruments that are recorded at nominal values, the value recorded in the other changes in the volume of assets account should correspond to their nominal value prior to being written off. For all reclassifications of assets and liabilities, values of both the new and old instruments should be the same.


Unit of account

3.130 The compilation of GFS, particularly transactions and stock positions with nonresidents, is complicated by the fact that the values may be expressed initially in a variety of currencies or perhaps in other standards of value, such as Special Drawing Rights (SDRs). The conversion of these transactions and stock positions expressed in another currency, or a commodity into a reference unit of account, is a requisite for the construction of consistent and analytically meaningful statistics. If financial assets or liabilities are in foreign currency units, data in a single currency unit are needed for compiling meaningful statistics.

3.131 From the perspective of the national compiler, the domestic currency unit is the obvious choice for measuring transactions and stock positions. Denominating government finance transactions and stock positions in such a way is compatible with the national accounts and most of the economy’s other economic statistics. Where a foreign currency is used to settle domestic transactions, such as with “dollarized” economies, this foreign currency may be the relevant currency unit for the compilation of GFS.

Currency conversion

3.132 The most appropriate exchange rate to be used for conversion of transactions and stock positions is the market (spot) rate prevailing on the transaction or balance sheet date. The midpoint between buying and selling rates should be used.

3.133 For debt transactions, in principle, the actual exchange rate applicable to each transaction should be used for currency conversion. The use of a daily average exchange rate for transactions usually provides a good approximation. If daily rates cannot be applied, average rates for the shortest period should be used. Some transactions occur on a continuous basis, such as the accrual of interest over a period of time. For such flows, therefore, an average exchange rate for the period in which the flows occur should be used for currency conversion. More details on currency conversion are given in the BPM6, paragraphs 3.104–3.108.

Domestic and foreign currency

3.134 For an economy, a domestic currency is distinguished from foreign currency. Domestic currency is that which is legal tender in the economy and issued by the monetary authority for that economy—that is, either that of an individual economy or, in a currency union, that of the common currency area to which the economy belongs. All other currencies are foreign currencies.

3.135 Under this definition, an economy that uses as its legal tender a currency issued by a monetary authority of another economy—such as U.S. dollars—or of a common currency area to which it does not belong should classify the currency as a foreign currency, even if domestic transactions are settled in this currency. Unallocated gold accounts and other unallocated accounts in precious metals giving title to claim the delivery of gold or precious metal are treated as denominated in foreign currency.

3.136 SDRs are considered to be foreign currency in all cases, including for the economies that issue the currencies in the SDR basket. Any other currency units issued by an international organization, except in the context of a currency union, are considered foreign currency.

Currency of denomination and currency of settlement

3.137 For debt statistics, a distinction should be made between the currency of denomination and the currency of settlement. The currency of denomination is determined by the currency in which the value of flows and stock positions is fixed as specified in the contract between the parties. Accordingly, all cash flows are determined using the currency of denomination and, if necessary, converted to the domestic currency or another unit of account for the purpose of settlement or compilation of accounts. The currency of denomination is important for distinguishing transaction values and holding gains and losses.

3.138 The currency of settlement may be different from the currency of denomination. Using a currency in settlement that is different from the currency of denomination simply means that a currency conversion is involved each time a settlement occurs. The currency of settlement is important for international liquidity and measurement of potential foreign exchange drains.

3.139 A financial instrument may be settled in domestic currency with both the amount to be paid at maturity and all periodic payments (such as coupons) linked (or indexed) to a foreign currency. In this instance, the currency of denomination is the foreign currency. Some instruments are denominated in more than one currency. However, if the amounts payable are linked to one specific currency, then the liability should be attributed to that currency.

Derived Measures

3.140 Derived measures consist of aggregates and balancing items. They are important analytic tools that summarize the values of selected flows or stock positions that have been individually recorded in the GFS framework. These derived measures are the sum or the balance of two or more flows or stock positions.

3.141 Aggregates are summations of individual entries and elements in a class of flows or stock positions. They allow for these data to be arranged in a manageable and analytically useful way. For example, tax revenue is the sum of all flows that are classified as taxes, and data for social security funds are the aggregations of the data for all institutional units in the economy that are classified as social security funds. Aggregates and classifications are closely linked in that classifications are designed to produce the aggregates thought to be most useful. Conceptually, the value for each aggregate is the sum of the values for all items in the relevant category. However, estimates of some aggregates may be needed due to deficiencies in source data, such as missing information on individual transactions, other economic flows, and asset and liability positions that may be incomplete or even nonexistent.

3.142 Balancing items are economic constructs obtained by subtracting one aggregate from a second aggregate. For example, the net operating balance is obtained by subtracting the total expense aggregate from the total revenue aggregate. Net worth is the balancing item equal to total assets minus total liabilities (see Chapter 4).

Netting of Flows and Stock Positions

3.143 It is feasible to present many categories of flows and stock positions on a gross or net basis. An item presented on a net basis is calculated as the sum of one set of flows or stock positions minus the sum of a second set of a similar kind. For example, total tax revenue could be presented on a gross basis as the total amount of all taxes accrued, or on a net basis as the gross amount minus tax refunds. Similarly, interest can be presented on a gross basis as interest revenue and interest expense, respectively, while it is feasible to calculate the net interest. The choice depends on the category of flows or stock positions, the nature of the items that might be subtracted to obtain the net value, and the analytic utility of the gross and net values. The choices for gross and net presentations as used in the GFS framework are discussed in paragraphs 3.144–3.151.

3.144 In GFS, revenue categories are presented gross of expense categories for the same or related category and likewise for expense categories. In particular, interest revenue and interest expense are presented gross rather than as net interest expense or net interest revenue. Similarly, social benefits and social contributions, grant revenue and expense, and rent revenue and expense are presented gross. Also, sales of goods and services are presented gross of the expenses incurred in their production.

3.145 In the case of the correction of erroneous or unauthorized transactions, revenue categories are presented net of refunds of the relevant revenue, and expense categories are presented net of inflows from the recovery of the expense. For example, refunds of income taxes may be paid when the amount of taxes withheld or otherwise paid in advance of the final determination exceeds the actual tax due. Such refunds are recorded as a reduction in tax revenue. Similarly, if social benefits that were paid in error are recovered, then such recoveries are recorded as a reduction in expense.

3.146 Acquisitions and disposals of nonfinancial assets other than inventories are presented gross. For example, acquisitions of land are presented separately from disposals of land. For analytic presentations, the net acquisition of each category of nonfinancial asset may be preferable and can be derived easily.

3.147 Netting is implicit in the presentation of some specific transactions categories in GFS—for example, changes in inventories. Changes in each type of inventories are presented net rather than tracking daily additions and withdrawals. That is, the change in materials and supplies is presented in the GFS framework as the net value of additions minus withdrawals. Nonetheless, full inventory accounting could allow for the gross recording of all the movements in inventories in the underlying administrative records. Similarly, tax revenue is presented net of nonpayable tax credits (see paragraphs 5.29–5.32).

3.148 Acquisitions and disposals of each category of financial assets/liabilities are also presented net in the GFS framework, to reflect the nature of the financial flows. For example, only the net change in the holding of assets related to currency and deposits is presented, not gross receipts and disbursements. Similarly, additions to liabilities in the form of loans are presented net of repayments. However, for analytical and administrative reasons, it may be useful to develop source data on the gross acquisitions and the gross disposals of each financial instrument as separate data categories.

3.149 Other economic flows are presented net. That is, the net holding gain for each asset and liability is presented, not gross holding gains and gross holding losses. Similarly, other changes in the volume of assets and liabilities are presented net, rather than recording increases and decreases in volume changes on a gross basis.

3.150 Stock positions held for the same type of financial instrument, both financial assets or liabilities, are presented gross. For example, a unit’s holding of debt securities as financial assets is presented separately from its liabilities for debt securities issued.

3.151 In the GFS framework the terms “gross” and “net” are used in a very specific manner. Apart from the balancing items net worth, net operating balance, and net lending/net borrowing, the GFS classifications employ the word “gross” and “net” to indicate the value of the operating balance and investment in nonfinancial assets before or after deduction of consumption of fixed capital. The framework also uses the term “net” to indicate that the net acquisition of financial assets represents both acquisition and disposal of assets, while the net incurrence of liabilities represents both incurrence and repayment of liabilities.


3.152 A consolidated set of accounts for a group of units, subsectors, or sectors, is produced by, first, an aggregation of all flows and stock positions within the GFS analytical framework, followed by the elimination, in principle, of all flows and stock positions that represent relationships among the units or entities being consolidated. In other words, consolidation eliminates the double-count because a flow or stock position of one unit is paired with the corresponding flow or stock position recorded for the second unit with which it is being consolidated, and both flows and/or stock positions are eliminated. For example, if one general government unit owns a bond issued by a second general government unit, and data for the two units are being consolidated, then the stock positions of bonds held as assets and liabilities of the consolidated unit are reported as zero (i.e., as if the bond position between them did not exist). At the same time, the interest related to this bond is consolidated, so that the interest revenue and expense of the consolidated account exclude the interest paid by the debtor general government unit to the creditor general government unit. Similarly, sales of goods and services between consolidated units are also eliminated.27


3.153 Consolidation is a method of presenting statistics for a set of units (or entities) as if they constituted a single unit. In the GFS framework, the data presented for a group of units are consolidated. In particular, statistics for the general government sector and each of its subsectors are presented on a consolidated basis. When units of the public sector are included in a presentation, the data for public corporations should be presented in two ways: as separate subsectors for the financial public corporations and for the nonfinancial public corporations; and together with general government units for the consolidated public sector. In both cases, the statistics should be presented on a consolidated basis within each group.

3.154 When compiling general government or public sector statistics, two types of consolidation may be necessary—namely, intrasectoral consolidation and intersectoral consolidation.

3.155 Intrasectoral consolidation is consolidation within a particular subsector to produce consolidated statistics for that particular subsector—for example, within the central government subsector or within the public nonfinancial corporations subsector. This consolidation may be required at two stages. A single institutional unit may require consolidation when the unit has multiple funds and accounts to carry out its operations and there are flows and stock positions among those funds. For example, a country may have a core central government institutional unit that has one or more departmental accounts, as well as special funds and accounts established for specific purposes. There are often flows and stock positions held between these accounts and funds that are recorded on a gross basis in the respective accounts. Failure to eliminate these transfers would yield aggregates that result from the accounting device, and not from interaction with units outside of the central government.

3.156 Intersectoral consolidation is consolidation between subsectors of the public sector to produce consolidated statistics for a particular grouping of public sector units—for example, consolidation between central, state, and local governments, and between general government and public nonfinancial corporations.

3.157 Intrasectoral consolidation is always done before intersectoral consolidation—for example, where more than one central government social security fund exists, the data for all social security funds should be consolidated before the consolidated social security data is presented as a subsector of the central government. Subsequently, data for all the subsectors of the central government will be subject to intersectoral consolidation to produce data for the consolidated central government.

Reasons for consolidation

3.158 The main reason for consolidation lies in the analytical usefulness of the consolidated statistics: consolidation eliminates the distorting effects on aggregates of differing administrative arrangements across countries or over time. The main impact of consolidation on the statistics is on the magnitude of aggregates. To relate government aggregates to the economy as a whole (e.g., revenue, expense, or debt to gross domestic product (GDP) ratios), it is better to eliminate the internal movement of economic value and include only those flows and stock positions that actually cross the boundaries with other sectors or nonresidents. The same arguments apply to why public corporations and public sector statistics should be consolidated.

3.159 By eliminating all reciprocal flows and stock positions among the units being consolidated, consolidation has the effect of measuring only flows or stocks of the consolidated units versus units outside the boundary. Consolidation excludes the economic interaction within the grouping of institutional units, and presents only those flows or stocks that involve interactions with all other institutional units in the economy, and the rest of the world.

3.160 Consolidation avoids double-counting of flows or stock positions among a grouping of institutional units, thus producing statistics that exclude these internal flows or stock positions. It is the avoidance of double-counting that produces the increased analytical usefulness of consolidated statistics in cases where it makes sense to view the consolidated group as acting as if it were a single entity.

Conceptual guidelines

3.161 Conceptually, consolidation entails the elimination of all intra- and intergovernmental flows and all debtor-creditor relationships among the units or entities that are combined. Consolidation requires a review of the accounts to be consolidated to identify inter- and intrasectoral flows and stock positions. The goal is to eliminate, in a consistent manner, flows and stock positions that will have a significant effect on the final derived measures. However, two types of transactions that appear to take place between two government units are never consolidated because they are rerouted in GFS (see paragraph 3.28):

  • Employer social contributions, whether paid to social security or government pension funds, are treated as being payable to the employee in the household sector as part of compensation and then payable by the employee to the social security scheme.

  • Taxes withheld by government units from the compensation of their employees, such as pay-as-you-earn taxes, and paid to other government units should be treated as being paid directly by the employees. The government employer is simply the collecting agent for another government unit, and is acting on behalf of the employees in the household sector.

3.162 Consolidation covers a range of categories of flows that may vary greatly in importance. The major transactions, in likely order of importance, cover:

  • Grants (current and capital) among general government units or entities

  • Interest income/expense

  • Taxes paid by one government unit or entity to another (except those taxes withheld on behalf of the household sector)

  • Purchases/sales of goods and services

  • Acquisitions/disposals of nonfinancial assets.

3.163 The following major transactions, other economic flows, and stock positions in financial assets and liabilities, in likely order of importance, should be consolidated:

  • Loans

  • Debt securities

  • Other accounts receivable/payable.

3.164 For the public sector, in addition to the foregoing financial instruments, the following flows and stock positions should also be eliminated—in principle—in both intra- and intersectoral consolidation:

  • Equity and investment fund shares

  • Currency and deposits

  • Insurance, pension, and standardized guarantee schemes.

Implementing consolidation

3.165 This Manual recommends that, based on Table 3.1, counterparty flows and stock position information be identified that will be eliminated in consolidation. But practicality should be kept in mind: the resources devoted to consolidation and the level of detail applied in consolidation should be in direct proportion to their fiscal significance. Suggestions for the sequence of analysis are:

  • Begin all consolidation exercises with an analysis of the accounts involved to determine whether there are flows or stock positions internal to the unit(s) to be consolidated. This will depend on knowledge of the relationships among the units involved. Do some of the units incur expense or receive revenue from the other units? Do some units extend loans to the other units? Do they buy debt securities issued by the other units? Do they have currency and deposits held by the other units?

  • Once these relationships are established, compilers must determine whether the intra- and/or intersectoral flows and stock positions can be measured or estimated, and whether the amounts will be significant in terms of analytical importance.

  • If the amounts are likely to be significant, are they large enough to justify the effort to collect the data and other information for consolidation purposes? The effort and cost to identify an amount to be consolidated should be directly proportional to the expected amount and its impact on the aggregates.

  • The “one-side” rule of thumb is commonly used. That is, if there is convincing evidence from the one institutional unit that a flow or stock position exists, it should be imputed to the counterparty. The imputation should be recorded even if there is no record of the flow or stock position in the counterparty’s accounts. When such an adjustment is made in the data for a unit where the flow or stock position cannot be directly identified, it will be necessary to ensure that the records for that unit are properly modified.

  • For flows and stock positions in financial assets and liabilities, normally the creditor can be expected to maintain the most reliable records. For loans, the creditor unit usually maintains the most complete records, but, with the international emphasis on proper debt recording, the debtor unit may be equally reliable. For debt securities, especially bearer instruments, only the creditor may have the information needed for consolidation. For example, when a central government issues bearer securities, some of which are acquired by public corporations, the central government may have no direct information on who is holding the securities, especially if they can be acquired on secondary markets. It is therefore necessary to rely on the creditor’s records.

  • Sometimes discrepancies exist between data for two units that are being consolidated. There are many reasons for such discrepancies, such as coverage, time of recording, valuation, and classification. Resolving these discrepancies will promote proper consolidation and improve the overall quality of GFS compiled. However, where a discrepancy cannot be resolved, decisions need to be made about which unit or group of units has the most reliable source data. Generally, the higher level of government is considered to have more reliable records than the lower levels of government.

  • To create consistency with other macroeconomic datasets, components of the data for the public sector should be presented in such a way that they show the data before consolidation and after consolidation. This will allow for the unconsolidated data to be consistent with the data required in the national accounts and other datasets that are presented before consolidation (see paragraph 3.168).

Table 3.1Detailed Classification of Counterparty Information
General government
Central government
State governments
Local governments
Social security funds2
Private corporations
Private nonfinancial corporations
Private financial corporations
Public corporations
Public nonfinancial corporations
Public financial corporations

Further breakdown/“of which” lines could allow for the identification of subsectors and individual units.

Social security funds are presented as a subsector only if their data are excluded from the data of the government level at which they are organized (see paragraph 2.78).

Further breakdown/“of which” lines could allow for the identification of subsectors and individual units.

Social security funds are presented as a subsector only if their data are excluded from the data of the government level at which they are organized (see paragraph 2.78).

3.166 Consolidation does not affect balancing items. In other words, the balancing items that are produced by simple aggregation are the same as those produced by consolidation. This is a result of the symmetry of the consolidation process, wherein the two sides of the consolidation adjustment fall within the same section of the analytical framework. When consolidated data produce different balancing items from the unconsolidated data, recording errors have been made. Therefore, when intra- or intersectoral flows and stock positions to be consolidated are not measured in the same amount by the units or subsectors involved, a consolidation method must be chosen that does not affect balancing items (see paragraph 3.165).

Consolidation in other datasets

2008 System of National Accounts

3.167 The 2008 SNA recommends, as a matter of principle, that statistics of institutional units should not be consolidated in the national accounts, but that consolidated accounts may be compiled for complementary presentations and analyses. Even then, transactions appearing in different accounts of the national accounts are never consolidated. The difference between the 2008 SNA and this Manual reflects the different uses of the statistics. The GFS framework is designed to produce statistics suitable for use in the analysis of the net relations between government and the rest of the economy. In particular, assessing the overall impact of government operations on the total economy or the sustainability of government operations is more effective when the measure of government operations is a set of consolidated statistics rather than unconsolidated statistics. The GFS framework also is not intended to produce a measure of production. The 2008 SNA, on the other hand, serves a range of other uses, including a comprehensive measure of production and relations among all the sectors of the economy.

Financial statements

3.168 In financial statements, compiled in accordance with accounting standards, accounting entries are often presented on a consolidated basis for the reporting entity and all of its controlled entities. This is done without regard to whether the controlled entities are general government units or public corporations, as those terms are used in this Manual, or whether the controlled entities are residents or nonresidents. This use of consolidation attempts to portray the operations and financial position of a parent and its subsidiaries as though the group of enterprises were a single unit. For example, a financial report for a state government unit would include all public corporations controlled by that government unit but would not include the statistics of any other state governments. In contrast, the consolidated statistics of the state government subsector in the GFS framework would include all state government units of the country, but would exclude all public corporations owned or controlled by those state governments.

In macroeconomic statistics, the term “flows” will often be used as a short form for “economic flows,” and “stocks” will often be used as a short form for “stock positions.”

For example, consumption of fixed capital (23) and the use of inventories in the production of goods and services (22) (see paragraphs 6.27 and 6.53, respectively).

The term “provides a good, service, asset …” is meant to include one unit allowing a second unit to use an asset owned by the first unit or a change in the ownership of an asset. Interest and other property income transactions are exchanges because they are receivable by a unit in return for putting assets at the disposal of another unit.

See paragraph 3.29 for a general statement on partitioning of transactions.

The prepayment of a premium is the acquisition of a financial asset by the insured (see paragraph A4.76), which reduces as insurance coverage is provided.

In the 2008 SNA, nonlife insurance premiums are partitioned into a sale of a service and a transfer. In GFS, the entire premium is considered a transfer (see paragraph 5.149).

Internal transactions are described in the 2008 SNA, paragraphs 3.85–3.90. GFS does not record all the internal transactions that relate to production processes.

References to changes in the valuation or volume of assets also apply to liabilities.

Holding gains and losses are also referred to as revaluations in the 2008 SNA and in generally accepted accounting standards.

The distinctions made are only for purposes of description; the GFS framework and classification system does not allow for this breakdown.

In contrast, accounting standards recognize nonfinancial liabilities under certain conditions.

Appendix 6 presents a broad description of linkages with financial accounting standards. It is recommended that the financial statements of government entities compiled in accordance with international accounting standards for governments be harmonized to the extent possible, and where differences remain, be reconciled with the equivalent GFS statements.

Another basis of recording is when flows are recorded at the time legal controls that are necessary to authorize a payment have been completed. This basis of recording is entitled “due-and-payable.”

These transactions are excluded from a pure cash basis of recording.

To maintain symmetry in the macroeconomic statistical system, the time of recording should be the same for both parties to the transaction.

The counterpart entry to accruing interest expense is a simultaneous increase in the amount of liability outstanding. Periodic payments reduce the liability that has accrued and are not expense transactions.

Charging a penalty interest rate on arrears that was stipulated in the original contract is not in itself a reason to reclassify the debt.

For example, a sales contract to the value of 100 is agreed on day 1, when the market price of the transaction is 100, for delivery on day 5. On day 5, the price of the item prevailing on the market is 102—the buyer records a transaction of 100 and immediately revalues the item.

Cash accounting that allows back-dated transactions (known as complementary periods), may distort actual cash flows. Recording of cash transactions in such a manner should be disclosed.

International statistical manuals consider that for nonnegotiable instruments, nominal value is an appropriate proxy for market value (see paragraph 7.30). Nonetheless, the development of markets, such as for credit derivatives linked to the credit risk of individual entities, is increasing the likelihood that market prices can be estimated even for nonnegotiable instruments. As these markets extend, consideration might be given to compiling additional information on market values of nonnegotiable debt.

In some statistical databases, face value is also called nominal value. However, in GFS, nominal value is understood to be different from face value, except on the date of maturity of the instrument.

More information on the valuation rules and numerical examples are in the PSDS Guide, paragraphs 2.115–2.123, and the 2013 EDS Guide.

The PSDS Guide recommends that debt instruments should be valued at nominal value, while debt securities should be valued at market value as well.

What constitutes an unusually long time in this context will depend on the circumstances. For example, for any given time period, the higher the level of interest rates or the longer the delay in payment, the greater is the opportunity cost of delayed payment.

This estimate is also referred to as the “written-down current acquisition” value.

See PSDS Guide, Box 8.1 and Table 8.2, for examples of consolidation.

    Other Resources Citing This Publication