3.12. A transaction is an economic flow that is an interaction between institutional units by mutual agreement or an action within an institutional unit that it is analytically useful to treat like a transaction, often because the unit is operating in two different capacities.

3.13. Institutional units, referred to in the definition, are the fundamental economic units of the System (see chapter IV). The following are the main attributes of institutional units that are relevant to their engaging in transactions:

• (a) They are entitled to own goods or assets in their own right, and therefore are able to exchange them;

• (b) They are able to take economic decisions and engage in economic activities for which they are held to be directly responsible and accountable at law;

• (c) They are able to incur liabilities on their own behalf, to take on other obligations or future commitments and to enter into contracts.

3.14. The definition of a transaction stipulates that an interaction between institutional units be by mutual agreement. When a transaction is undertaken by mutual agreement, the prior knowledge and consent of the institutional units is implied. This does not mean, however, that both units necessarily enter a transaction voluntarily, because some transactions are imposed by force of law, such as payments of taxes or other compulsory transfers. Although individual institutional units are not free to fix the amounts of taxes they pay, there is nevertheless collective recognition and acceptance by the community of the obligation to pay taxes. Thus, payments of taxes are considered transactions despite being compulsory.

3.15. Transactions take so many different forms that, even with these explanations, any general definition is inevitably rather imprecise. To give more precision, the various kinds of transactions have to be systematically described and classified. A first distinction is between monetary and non-monetary transactions. Other distinctions, such as between transactions with and without counterparts, are drawn within each of these kinds of transactions. Frequently the individual, identifiable transactions of everyday economic life are simply grouped together in the accounts; sometimes they are subdivided and rearranged in order to form the transaction categories of the System.

3.57. Other flows are changes in the value of assets and liabilities that do not take place in transactions. Although these entries, which are all in the other changes in the volume of assets account and the revaluation account (which make up the other changes in assets accounts), do share the characteristic that they record changes that are not the result of transactions, they cover very different kinds of changes in assets and liabilities. The first kind consists of changes due to factors such as discoveries or depletion of subsoil resources, destruction by war or other political events or destruction by natural catastrophes, all of which actually change the volume of assets. The second kind consists of changes in the value of assets, liabilities, and net worth due to changes in the level and structure of prices, which are reflected in holding gains and losses.

3.64. A balancing item is an accounting construct obtained by subtracting the total value of the entries on one side of an account from the total value for the other side. It cannot be measured independently of the other entries; as a derived entry, it reflects the application of the general accounting rules to the specific entries on the two sides of the account. It does not relate to any specific set of transactions, and so it cannot be expressed in terms of its own price or quantity units.

3.65. Balancing items are not simply devices introduced to ensure that accounts balance. They encapsulate a great deal of information and include some of the most important entries in the accounts, as can be seen by the examples of balancing items for the accounts containing flows reproduced below:

• Value added/domestic product

• Operating surplus

• Disposable income

• Saving

• Net lending/net borrowing

• Current external balance.

3.66. Stocks are a position in, or holdings of, assets and liabilities at a point in time. The System records stocks in accounts, usually referred to as balance sheets, and tables at the beginning and end of the accounting period. However, stocks are connected with flows: they result from the accumulation of prior transactions and other flows, and they are changed by transactions and other flows in the period. They result in fact from a continuum of entries and withdrawals, with some changes, in volume or in value, occurring during the time a given asset or liability is held.

3.67. Values are recorded for non-financial assets, both produced and non-produced, and for financial assets and liabilities. The coverage of assets is limited to those assets used in economic activity and that are subject to ownership rights; thus for example, consumer durables and human capital, and also natural resources that are not owned, are excluded.

3.68. Net worth, which is defined as the value of all the non-financial and financial assets owned by an institutional unit or sector less the value of all its outstanding liabilities, is the balancing item in the balance sheets. As is true for other balancing items in the System, net worth cannot be measured independently of the other entries, and it does not relate to any specific set of transactions.

3.70. The power of the SNA as an analytical tool stems largely from its ability to link numerous, very varied economic phenomena by expressing them in a single accounting unit. The System does not attempt to determine the utility of the flows and stocks which come within its scope. Rather, it measures the current exchange value of the entries in the accounts in money terms, i.e., the values at which goods and other assets, services, labour or the provision of capital are in fact exchanged or else could be exchanged for cash (currency or transferable deposits).

3.71. When institutional units exchange these items with other institutional units for cash, the values required by the System are directly available. These transactions are recorded at the actual exchange value agreed upon by the two parties. Furthermore, of course, the values for all flows and stocks that concern cash holdings and liabilities are directly known.

3.72. In respect of all remaining flows and stocks, no actual exchange values are at hand, so their values must be assessed indirectly. These values should be taken from markets in which the same or similar items are traded currently in sufficient numbers and in similar circumstances against cash. The selection of the appropriate reference markets requires that attention be paid to differences between wholesale and retail markets, regional divergences, etc. Non-monetary transactions in existing goods often can be valued at the market price for similar new goods, if properly adjusted for consumption of fixed capital and other elements, such as unanticipated damage, which may have accrued to the asset since the time it was produced.

3.73. If there is no appropriate market from which the value of a particular non-monetary flow or stock item can be taken by analogy, as a second best, its valuation could be derived from prices that are established in less closely related markets. Ultimately, some goods and services can only be valued by the amount that it would cost to produce them currently. Market and own-account goods and services valued in this way should include a mark-up that reflects the net operating surplus or mixed income attributable to the producer. For non-market goods and services produced by government units or NPISHs, however, no allowance should be made for any net operating surplus.

3.74. Sometimes it is necessary to value stocks at their estimated written down current acquisition values or production costs. The write-down should then include all changes which have occurred to the item since it was purchased or produced (such as consumption of fixed capital, partial depletion, exhaustion, degradation, unforeseen obsolescence, exceptional losses and other unanticipated events). The same method could be applied to non-monetary flows of existing assets.

3.75. If none of the methods mentioned above can be applied, flows and stocks are to be recorded at the discounted present value of expected future returns. Although this method is theoretically entirely justified, it is not generally recommended since it involves many assumptions and as a consequence the outcomes are highly speculative.

3.76. Flows and stocks concerning foreign currency are converted to their value in national currency at the rate prevailing at the moment they are entered in the accounts, i.e., the moment the transaction or other flow takes place or the moment to which the balance sheet applies. The midpoint between the buying and selling rate should be used so that any service charge is excluded. In conformity with the general rule, provision of assets, services, labour or capital in exchange for foreign cash is recorded at the actual exchange value agreed upon by the two parties to the transaction. The exchange value should then be converted to national currency at the midpoint rate prevailing at the time the transaction takes place. That moment may be different from the times the payments are made; as a result, the value in national currency of the transactions in question may differ from the value in national currency of the related payments when they take place.

3.77. Business accounts, tax returns and other administrative records are main sources of data for drawing up the national accounts. One should be aware, however, that none of these necessarily satisfies the valuation requirements of the System and that accordingly adjustments may have to be made. In particular, in the interest of prudence, business accounting often adopts valuations that are not appropriate for the national accounts. Similarly, valuations for tax purposes often serve objectives that differ from those of macroeconomic analysis. For example, the depreciation methods favoured in business accounting and those prescribed by tax authorities almost invariably deviate from the concept of consumption of fixed capital employed in the System.

3.78. Extending the general rule in paragraph 3.71 above, where a single payment refers to more than one transaction category (such as they are defined in the System), the individual flows need to be recorded separately. For example, the System recommends dividing interest transactions with financial enterprises between two transaction categories whenever possible: one standing for pure interest and the other representing the implicit payment for financial intermediation services. Earlier in this chapter, the partitioning of financial leasing and transactions of wholesalers and retailers were discussed. Partitioning is not limited to transactions; an example is real holding gains, which are separated for analytical reasons from neutral holding gains that are simply proportionate to changes in the general price level.

3.79. In some cases partitioning is connected with deceptive behaviour. Values put on an invoice may deviate systematically or to such a large extent from the prices paid in the market for similar items that it must be presumed that the sums paid cover more than the specified transactions. An example is so-called transfer pricing: affiliated enterprises may set the prices of the transactions among themselves artificially high or low in order to effect an unspecified income payment or capital transfer. Such transactions should be made explicit if their value is considerable and would hinder a proper interpretation of the accounts.

3.80. A less obvious mingling of transactions occurs when the provision of an asset and the related money payment or payments do not take place simultaneously. When the time gap becomes unusually long and the amount of trade credit extended is very large, the conclusion may be that implicitly an interest fee has been charged. In such extreme cases, the actual payment or payments should be adjusted for accrued interest in order to arrive at the correct value of the asset transferred. Such adjustments are not recommended for normal trade credit.

3.81. Usually, the producer and the user of a given product perceive its value differently owing to the existence of taxes and subsidies on products, transport costs to be paid and the occurrence of trade margins. In order to keep as close as possible to the views of the economic transactors themselves, the System records all uses at purchasers’ prices including these elements, but excludes them from the value of output of the product.

3.82. Output of products is recorded at basic prices. The basic price is defined as the amount receivable by the producer from the purchaser for a unit of good or service produced as output minus any tax payable and plus any subsidy receivable on the product as a consequence of its production or sale. It excludes any transport charges invoiced separately by the producer. If it proves impossible to obtain the required information at basic prices, output may be valued at producers’ prices. The producer’s price is defined as the amount receivable by the producer from the purchaser for a unit of a good or service produced as output minus any value added tax (VAT), or similar deductible tax, invoiced to the purchaser. It also excludes any transport charges invoiced separately by the producer.

3.83. Use of products is recorded at purchasers’ prices. The purchaser’s price is defined as the amount payable by the purchaser, excluding any deductible VAT or similar deductible tax, in order to take delivery of a unit of a good or service at the time and place required by the purchaser. The purchaser’s price of a good includes any transport charges paid separately by the purchaser to take delivery at the required time and place.

3.84. The difference in value recorded for a product between when it is produced and the moment it is used for, say, final consumption expenditure can be considerable. Components of this difference may be:

• (a) Taxes less subsidies on products payable by the producer;

• (b) Trade and transport margins, including taxes less subsidies on products payable by wholesale and retail traders;

• (c) Transport, including taxes less subsidies on products, paid separately by the consumer;

• (d) Predictable quality increases producing additional output volume less current losses during storage;

• (e) Holding gains while the product is with the producer and with wholesale and retail traders.

As one can see from the above, the difference between the original basic price and ultimate purchasers’ price of a particular good encompasses both pure price and volume elements. In practice, of course, the estimates do not keep track of individual products but are made at a more global level for groups of products.

3.85. Imports and exports of goods are recorded in the System at border values. Total imports and exports of goods are valued free-on-board (f.o.b., that is, at the exporter’s customs frontier). As it may not be possible to obtain f.o.b. values for detailed product breakdowns, the tables containing details on foreign trade show imports of goods valued at the importer’s customs frontier (c.i.f. value), supplemented with global adjustments to f.o.b. C.i.f. values include the insurance and freight charges incurred between the exporter’s frontier and that of the importer. The value on the commercial invoice may of course differ from both of these.

3.86. As the overall balance of imports and exports must conform to actual circumstances, border valuation of goods has consequences for the recording of freight and insurance in the System. Usually, the values of both imports and exports for these service items have to be adapted to compensate for the special conventions on goods trade with the rest of the world. Further details on this treatment are in chapters XIV and XV.

3.87. When discussing timing in the System, an essential distinction should be made between stock data as recorded in balance sheets, on the one hand, and flow data as recorded in the accounts, on the other. Balance sheets, by definition, refer to specific points in time. In contrast, flows are aggregations, over some chosen accounting period, of individual transactions or other flows, which are themselves scattered over the accounting period.

3.88. Thus, the System does not show individual transactions or other flows, but there are two reasons why precise rules on their individual timing must be given. First, rules have to be formulated to say in which accounting period the discrete flows are to be recorded. Secondly, an exact timing of individual flows within the accounting period is crucial to make the distinction between changes in net worth due to transactions and those due to holding gains or losses. This distinction is particularly important in situations of high inflation.

3.89. One of the problems in pinning down the timing of transactions is that activities of institutional units often stretch over periods in which several important moments can be distinguished. For instance, many commercial sales commence with the signing of a contract between a seller and a buyer, encompass a date of delivery and a date or dates on which payments become due and are only completed as of the date the last payment is received by the seller. Each of these distinct moments in time is to some extent economically relevant.

3.90. Similarly, in analysing public spending one can distinguish the day that a budget is voted upon by the legislature, the day on which the ministry of finance authorizes 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 days payment orders are issued and cheques 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 actually paid or refunds are made.

3.91. Clearly, making entries for all successive stages discernible within the activities of institutional units, although theoretically possible, would severely overburden the System. A choice has to be made, recognizing (a) the needs of macroeconomic analysis, (b) microeconomic views, and (c) commonly available sources. Often, in this respect, a distinction is drawn between recording flows on a cash basis, due-for-payment basis and accrual basis. The System recommends recording on an accrual basis throughout.

3.92. Some systems closely related to the SNA, most notably the IMF Government Finance Statistics, recommend recording on a cash basis, at least in respect of aggregate data. Cash accounting records only cash payments and records them at the times these payments occur. This method is widely used to record government revenues and expenditures and it is also employed for certain business purposes. A practical advantage is the avoidance of problems connected with valuing non-monetary flows. Yet, cash accounting cannot be used generally for economic and national accounting as the times at which payments take place may diverge significantly from the economic activities and transactions to which they relate and it is these underlying activities and transactions that the System seeks to portray. Moreover, cash recording cannot be applied to the many non-monetary flows included in the System.

3.93. Due-for-payment recording shows flows which give rise to cash payments at the latest times they can be paid without incurring additional charges or penalties and, in addition to these, actual cash payments at the moments they occur. The period of time (if any) between the moment a payment becomes due and the moment it is actually made is bridged by recording a receivable or a payable in the financial accounts. Due-for-payment recording furnishes a more comprehensive description of monetary flows than does cash accounting. A disadvantage is, of course, that the registration is still limited to monetary flows.

3.94. Accrual accounting records flows at the time economic value is created, transformed, exchanged, transferred or extinguished. This means that flows which imply a change of ownership are entered when ownership passes, services are recorded when provided, output at the time products are created and intermediate consumption when materials and supplies are being used. The System favours accrual accounting because:

• (a) The timing of accrual accounting is in full agreement with the way economic activities and other flows are defined in the System. This agreement allows one, for instance, to evaluate the profitability of productive activities correctly (i.e., without the disturbing influence of leads and lags in cash flows) and to calculate a sector’s net worth correctly at any point in time;

• (b) Accrual accounting can be applied to non-monetary flows.

3.95. Many transactions, such as everyday purchases of households in shops, are monetary transactions in which some asset is delivered against immediate, or nearly immediate, payment in cash. In those instances there are no differences between the three methods discussed here. Accrual accounting is particularly relevant to the timing of various internal transactions (such as output that is added to the inventories of the producer), exchanges in which the parties deliver at differing times (such as sales with deferred payments) and obligatory transfers (taxes and flows connected with social security).

3.96. Usually, accrual accounting arises naturally to the institutional units involved. Numerous transactions consist of an exchange between two enterprises of, say, goods for financial assets. In such an exchange, accounting entries will be made in the books of each enterprise, showing the same dates for the acquisition of the goods and the surrender of the financial assets, on the one hand, and for the acquisition of the financial assets and the surrender of the goods, on the other. Sometimes, however, the two parties involved in a transaction will not perceive it as occurring at the same moment. Furthermore, some transactors, in particular government units, do not keep records of purchases on an accrual basis. In these cases, the rules of consistency in the System require that efforts should be undertaken to correct basic statistics for major deviations and flaws. The application of the general rule of recording on an accrual basis to the most common circumstances is discussed below.

3.97. With respect to exchanges and transfers of non-financial assets, the accrual principle usually comes down to recording at the moment when the legal ownership of those assets passes. When change of ownership is not obvious, the moment of entering in the books of the transaction partners may be a good indication and, failing that, the moment when physical possession and control is acquired. These subsidiary rules apply in particular to internal transactions or when a change of ownership is taken to occur under a financial lease or hire-purchase arrangement. Imports and exports of goods are recorded when change of ownership occurs. In the absence of sources specifying the date on which ownership changes, there is a strong presumption that the goods will cross the frontiers of the countries concerned either shortly before or soon after the change of ownership takes place and trade statistics based on customs documents reflecting the physical movement of goods across the national or customs frontier may often be used as an approximation.

3.98. Services are recorded in the System when they are provided. Some services are special in the sense that they are characteristically supplied on a continuous basis. Examples are operating leasing, insurance and housing services (including those of owner-occupied dwellings). These services are recorded as provided continuously over the whole period the contract lasts or the dwelling is available.

3.99. Following the general rule, distributive transactions are recorded at the moment the related claims arise. As a result, for example, compensation of employees, interest, rent on land, social contributions and benefits are all registered in the period during which the amounts payable are built up. Equally, entries for taxes are made at the moment on which the underlying transactions or other flows occur which give rise to the liability to pay. This implies that taxes on products and imports are recorded at the times the products in question are produced, imported or sold, depending on the basis for taxation. Current taxes on income are recorded when the income to which they pertain is earned although taxes deducted at source may have to be recorded when they are deducted. With respect to some distributive transactions, the time of accrual depends on the unit’s decision when to distribute income or make a transfer. The level of dividends is not unambiguously attributable to a particular earning period, and dividends are to be recorded as of the moment they are declared payable. Other examples are withdrawals from income of quasi-corporations and various voluntary transfers, which are recorded when effected.

3.100. Transactions in financial assets (including payments of cash) are recorded in the System on a change-of-ownership basis. Some financial claims/liabilities defined in the System, in particular trade credits and advances, are the implicit result of a non-financial transaction and do not involve an explicit piece of evidence. In these cases the financial claim is deemed to arise when its non-financial counterpart occurs. The same holds for financial transactions that the System records between a quasi-corporation and its owner.

3.101. Both parties involved in a financial transaction may record it at varying dates in their own books because they acquire the documents evidencing the transaction at different times. This variation is caused by the process of clearing, the time cheques are in the mail, etc. The amounts involved in such “float” are generally substantial in the case of transferable deposits and other accounts receivable and payable. Again, reasons of consistency require that the transactions are entered on the same date for both parties. If no precise date can be fixed on which the change of ownership occurs, the date on which the transaction is fully completed (thus the date on which the creditor receives his payment) is decisive.

3.102. The principle of recording on an accrual basis implies that output is recorded over the period in which the process of production takes place. Thus, additions to work-in-progress are recorded continuously as work proceeds. When the production process is terminated, the whole of the work-in-progress accumulated up to that point is effectively transformed into a stock of finished product ready for delivery or sale.

3.103. Similarly, the intermediate consumption of a good or service is recorded at the time when the good or service enters the process of production, as distinct from the time it was acquired by the producer.

3.104. Inventories may be materials and supplies held as inputs by producers, output as yet unsold, or products held by wholesale and retail traders. In all cases, additions to inventories are recorded when products are purchased, produced or otherwise acquired. Deductions from inventories are recorded when products are sold, used up as intermediate consumption or otherwise relinquished.

3.105. The timing of consumption of fixed capital is nearly inseparably linked with the question of its valuation. Consumption of fixed capital is a cost category which accrues over the whole period the fixed asset in question is available for productive purposes. The exact proportioning to accounting periods depends on the rate of depreciation which is usually assumed to be either linear or geometric (declining balance) (see chapter VI).

3.106. Transactions which are measured as the balance of two or more other transactions follow the timing of the constituent basic flows. For example, financial intermediation services indirectly measured (FISIM) are entered according to the moments property income is earned on funds put out by financial intermediaries and interest payable by these units accrues. The same rule applies to balancing items.

3.107. In order to yield significant balancing items, the System aims at timing resources in the same period as related uses (and, in particular, at matching output with the various costs of production). However, due to the variety of transactions and other flows covered, each with its own characteristics, some thought is needed in interpreting balancing items. For instance, in analysing the balancing item “saving” of non-financial corporations, one should be aware that the time when the operating surplus arises does not necessarily tally with the timing of the other factors, such as when dividends are payable.

3.108. Other volume changes in assets are usually discrete events which accrue at precise moments or within fairly short periods of time. Changes in prices often have a more continuous character, particularly in respect of assets for which active markets exist. In practice, nominal holding gains or losses will be computed between two points in time:

• (a) The moment at which:

  • (i) The accounting period begins; or

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

  • (iii) An asset is produced; and

• (b) The moment at which:

  • (iv) The accounting period ends; or

  • (v) The ownership of an assets is relinquished (through sale or a transaction in kind); or

  • (vi) An asset is consumed in the production process.

3.109. One may wonder why nominal holding gains and losses are not calculated over a period beginning at the moment on which two units agree to a mutual exchange of assets instead of the period which starts with the moment on which the assets are acquired. After all, does not the signing of the contract fix prices, implying that the risk for any later price changes is being transferred? The System, however, regards commitments resulting from a contract as contingent until one of the parties has performed its obligation (by passing the ownership of some asset to the other party, providing a service or providing labour or capital). Also, a unit can incur holding gains and losses exclusively on the assets or liabilities it effectively holds. The combination of these two rules implies that during the period between the signing of the 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.

3.110. Changes in structure and classification should be entered at the very moment when, according to the rules adopted in the System, a unit or an asset comes into a different category than it was classified previously. An integrated stock-flow system like the SNA requires that all reclassifications be recorded. Consequently, such a system does not allow reclassifications between two consecutive accounting periods.

3.111. In order to obtain statistical series which are more comparable over time, 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. However understandable this procedure might be, it does not conform to the System’s recommendations, which aim at correct estimates on levels. Keeping records of reclassifications makes it possible in principle to reconstruct time series based on the situation in any accounting period.

3.112. Fundamentally, balance sheets can be drawn up at any point in time. The System defines balance sheets for all sectors at the moment when one accounting period ends and a new accounting period begins. The closing balance sheet of one period is identical to the opening balance sheet of the next one, so there remain no price changes, reclassifications or other economic flows which are not duly recognized by the System.

3.113. In principle, any lapse of time may be chosen as the accounting period. Periods that are too short have the disadvantage that statistical data are influenced by incidental factors, while long periods do not adequately portray changes going on in the economy. Merely seasonal effects can be avoided by having the accounting period cover a whole cycle of regularly recurrent economic phenomena. Most business and government accounting refers to complete years. In general, calendar years or quarters are best suited for drawing up a full set of national accounts tables.

3.114. The immense number of individual transactions, other flows and assets within the scope of the SNA have to be arranged in a manageable number of analytically useful groups. In the System, such groups are constructed by crossing two or more classifications. As a minimum, a combination of the “who” and “what” questions asked in chapter II is made, i.e., a classification of institutional sectors or industries is crossed with the classification of transactions, other accumulation entries or assets. Additionally, resources must be distinguished from uses and assets from liabilities. In order to accommodate more detailed analysis, the classes thus generated may be further subdivided: examples are specifications of “in what” (kind of product or asset), of “why” (function) and of “with whom” (transaction partners).

3.115. Since the classifications in the System contain a number of levels made explicit in the codes, corresponding levels of aggregation may be distinguished.

3.116. Although conceptually the value for each aggregate is the sum of the values for all elementary items in the relevant category, in practice other estimation methods are frequently used. First, information on elementary transactions, other flows and assets may be incomplete or even non-existent. Secondly, the data obtained from different primary sources are usually not fully consistent due to deviating definitions and varying coverage, and adjustments at aggregate level are necessary to reconcile them. The practical techniques for estimating and balancing national accounts data are to be extensively discussed in handbooks which are planned for publication by organizations of the Inter-Secretariat Working Group on National Accounts.

3.117. Individual units or sectors may have the same kind of transaction both as a use and as a resource (e.g., they both pay and receive interest) and the same kind of financial instrument both as an asset and as a liability. Combinations in which all elementary items are shown for their full values are called gross recordings. Combinations whereby the values of some elementary items are offset against items on the other side of the account or which have an opposite sign are called net recordings.

3.118. The System recommends gross recording apart from the degree of netting which is inherent in the classifications themselves. In fact, netting is already a feature of many of the System’s recommendations. It mostly serves to highlight an economically important property which is not apparent from gross data.

3.119. Netting is implicit in various transaction categories, the most outstanding example being “changes in inventories”, which underlines the analytically significant aspect of overall capital formation rather than tracking daily additions and withdrawals. Similarly, with few exceptions, the financial account and other changes in assets accounts record increases in assets and in liabilities on a net basis, bringing out the final consequences of these types of flows at the end of the accounting period. Of course, all balancing items involve netting as well. To avoid confusion, the System uses the words “gross” and “net” in a very restrictive sense. Apart from a few headings (“net premiums”, “net equity of households on life insurance reserves and pension funds”, “net worth and net lending/net borrowing”) the System’s classifications employ the word “net” exclusively to indicate the value of variables after deduction of consumption of fixed capital.

3.120. The System defines one situation in which exactly the opposite of netting should be applied. A grossing up is recommended in respect of processing to order vis-à-vis the rest of the world. In agreement with external trade statistics, this type of processing is not recorded “net” as a service item. Instead, the movement of goods before and after processing is registered as imports and exports for their full value as if they were purchased and subsequently sold back.

3.121. Consolidation is a special kind of cancelling out of flows and stocks which should be distinguished from other kinds of netting. It involves the elimination of those transactions or debtor/creditor relationships which occur between two transactors belonging to the same institutional sector or sub-sector. Consolidation should not be seen as a sheer loss of information; it entails an elementary specification by the transaction partner. Consolidation may be most relevant for monetary institutions and general government. For certain kinds of analysis, information on the transactions of these (sub)sectors with other sectors and the corresponding “external” financial position is more significant than overall gross figures. As a rule, however, the entries in the System are not consolidated.

3.122. The rule of non-consolidation takes a special form regarding the transaction categories “output” and “intermediate consumption”. These transactions are to be recorded throughout at the level of establishments. This implies specifically that the accounts for institutional sectors and for industries should not be consolidated in respect of output delivered between establishments belonging to the same institutional unit.