Journal Issue

The Concepts of Territory and Resident in the Social Accounts

International Monetary Fund. Research Dept.
Published Date:
January 1967
  • ShareShare
Show Summary Details

THE PURPOSE OF THIS PAPER is to clarify some of the basic ideas underlying national accounts as well as their relationship to the balance of payments. It is not intended to provoke changes in internationally accepted procedures for national accounting, but merely to suggest ways in which these procedures could be more adequately described and rationalized. While certain changes in actual accounting procedures might result from a more rational description of their conceptual framework, such changes would in any case be marginal and statistically insignificant.

The main contention of the paper is that the concept of territory should be discarded altogether from the national accounts as a basis for defining their basic categories, and that these should be defined, like those of the balance of payments, solely in terms of the concept of resident (i.e., in terms of a certain number of economic units rather than a piece of land).

National and Domestic Product

It is customary to define production in the national accounts from two points of view. The domestic product, it is said, is the aggregate value added in production in the domestic territory. Some of this value accrues to foreign (nonresident) suppliers of factors of production. On the other hand, residents derive income from factors of production operating abroad. The aggregate product attributable to factors of production owned by residents is termed national product. It is equivalent to domestic product plus factor income received from the rest of the world less factor income paid to the rest of the world. Domestic product is thus associated with territory, and national product with resident.

This explanation has been repeated so often that it has been accepted as almost axiomatic by most national accounts statisticians and by nontechnicians using the national accounts. Yet it is incorrect as a description of domestic product as this term is used. Domestic product is not, in practice, a territorial concept. This was brought out in discussions in 1956 between national accounts specialists of the United Nations and the Organization for European Economic Cooperation and balance of payments specialists of the International Monetary Fund. In the report on these discussions, addressed to the Statistical Commission of the United Nations,1 some suggestions were given for clarifying the concept of domestic product by redefining it. These suggestions were in part adopted in a new edition of the System of National Accounts,2 while the earlier language was also in part retained. Accordingly, one can now find the category of domestic product defined in two ways in SNA. However, the explanations and modifications given in this document, which are based on the discussions referred to above, help to illuminate the inconsistent manner in which the territorial definition is applied.3

The definitions of national and domestic product are given in SNA as follows:

  • 46. If the production accounts of all domestic producers are consolidated, the resulting total will measure the production taking place in what is called the domestic territory of the country. Some of the value added in this production accrues to foreign suppliers of factor services rather than to normal residents of the country. These residents derive, in addition to income which accrues to them from domestic production factor, income from abroad. If the total of domestic product is adjusted for these external flows of factor income by subtracting the outward flow and by adding the inward flow there results a total termed the national product. This total refers not to production originating with the territory but to the total of production all over the world attributable to factors of production supplied by normal residents of the territory.
  • 47. As an alternative formulation of the definition of domestic product given in the preceding paragraph, domestic product can be defined as the product attributable to factor services rendered to resident producers. Factor services rendered to resident producers of the given country when they are located abroad, including resident institutions functioning abroad, are included in domestic product. Domestic and national product can thus be defined in analogous terms: domestic product as the total of production attributable to factor services rendered to resident producers of the given country; national product as the total of production attributable to factor services supplied by residents of the given country.4

These paragraphs give two essentially different definitions of domestic product, whereas the two definitions of national product are, in fact, identical. The alternative definition of domestic product in paragraph 47 is expressed in such abstract terms that its content may not be self-evident. The product of a given production unit is usually conceived as the balance between an output of goods and (nonfactor) services and an input arising from purchases of goods and (nonfactor) services from other production units, at home or abroad, with an allowance for changes in inventories. This balance is described as the value added in production. It is matched by payments to owners of factors of production, i.e., for services rendered to the production unit concerned.5 Domestic product is the aggregate of value added in all resident production units, and it can thus be ascribed to factor services “rendered to resident producers.” From this standpoint, it may also be conceived as product arising in the domestic economy, but only if the domestic economy is defined as the totality of residents rather than in terms of the national territory. The definition of national product as product attributable to factor services supplied by residents is the traditional one.

Certain difficulties inherent in the territorial definition of domestic product are brought out in paragraph 48 of SNA, from which the four opening sentences are quoted below:

  • 48. The domestic territory of a country is here defined to exclude overseas territories and possessions and to include, in addition to the territory lying within its political frontiers, ships and aircraft operated by domestic carriers even while in the territorial waters of another country or in the air above it. The operation of ships on time-charter should be treated as an enterprise activity of the country in which the ships’ operators are resident, the value added in production by that enterprise being net of the charter fee paid to the ships’ owners. This charter fee would constitute a payment for a non-factor service produced in the country of the owner. The ships involved would remain part of the domestic capital of the countries owning them….6

The difficulty in attributing product of ships operating under charter hire by country on a territorial basis is an example of a problem of a more general character referring to capital equipment (including also, for instance, construction equipment and motion pictures) leased by residents of one country to those of another. Ostensibly, the services of such equipment are rendered in the country in which the equipment is located and, therefore, should apparently be included in the domestic product of that country. Yet the prevailing practice, which makes perfectly good sense, is to include them in the domestic product of the owner’s country. This is the implication of the statement just quoted: “This charter fee would constitute a payment for a non-factor service produced in the country of the owner.”7 For if a territorial concept were applied, it would be hard to deny that the service was produced in the country of the operator of the ship, since it is considered to be part of that country’s territory.

Another departure from the territorial principle of product allocation refers to certain personal services produced abroad but included in domestic product (or vice versa produced on domestic territory but excluded from domestic product). For instance, a small part of domestic product as conventionally conceived refers to work performed abroad by employees of resident enterprises (e.g., business travelers). Another example is that of the services of diplomatic and military personnel stationed abroad. The inclusion of their services in the domestic product of the country of which they are residents is rationalized by the somewhat irrelevant legal construction of extraterritoriality. Indeed, such extraterritoriality apparently is attributed not only to the national representatives of a given country but also to the locally recruited staff of embassies and military establishments.

Paragraph 51 of SNA includes the following statement:

The official diplomatic and consular representatives of a given country, including members of official missions and members of armed forces stationed abroad, are to be considered extraterritorial by the country in which they are located and therefore as residents of the given country. They are considered to contribute to the domestic product of the countries in which they are residents. Wages and salaries paid to locally recruited staff of foreign diplomatic and military establishments are regarded as factor income paid by these establishments and the corresponding factor services would be included in the national but not the domestic product of the country of location….8

All these departures from the territorial definition of domestic product disappear if the resident definition is substituted. It is apparent that only this definition can bring order out of the conceptual disorder. What is the rationale of this definition? It may perhaps be expressed as follows:

Domestic product as it has developed in practice is a purely pragmatic concept. Orderly accounting procedures require that, to avoid duplication, a distinction be made in production accounts between the input that represents the output of other producers and that which does not and which is, therefore, regarded as the services of the original factors of production. The product attributed to a certain producer is equivalent to his sales (or his own net use) of product less his purchases of products from other producers. For accounting purposes, therefore, it is necessary to draw a clear distinction in production accounts between the use of the original services of factors of production and that of other goods and services (i.e., goods and nonfactor services).9

Domestic product covers the production that a national statistical service can record conveniently following these procedures by addressing itself to residents, i.e., to enterprises operating in a given country (or using it as the main base for its operations, as for shipping companies); the government and all its agencies at all levels; and the households of all persons normally staying in the country. Defined in such pragmatic terms, domestic product lacks, of course, any ideological or nationalistic appeal that may perhaps be associated with the concept of territorial product. For statistical purposes, however, it has the great advantage of lending itself readily to a classification by industry while national product cannot easily be so classified. Domestic product, moreover, coincides by and large with that generating income which the authorities of the country concerned can subject to taxation. In the highly important case of a direct investment company, for instance, the authorities of the country in which the company operates can effectively tax the income generated by the foreign-owned factors of production.

Other Domestic and National Concepts

The territorial idea has had some influence on other concepts in national accounting. In an article published in 1955, one author, G. Stuvel, defined a whole series of national and domestic concepts (product or income, consumption, exports, and imports).10 Domestic consumption is consumption in the domestic territory, and domestic exports and imports refer to goods and (nonfactor) services crossing its borders. In the article, Stuvel does not differentiate between domestic and territorial concepts, taking the view that “domestic production units … can exercise their production activity only on the domestic territory on which they are located.”11 In his later work, however, Stuvel defines domestic product as equivalent to factor income payable by resident production units, acknowledging the (rare) possibility of such units “operating abroad on their own account rather than through the intermediary of a subsidiary, branch, or agent abroad.”12

Territorial concepts are not widely used. Consumption is thus generally conceived as consumption by residents (although this may have to be estimated from consumption purchases recorded in the national territory with certain adjustments for international transactions). Exports and imports are also generally conceived in balance of payments terms as covering transactions in goods and services between residents and foreigners, rather than in terms of physical flows between territories. The concept of capital formation is more ambiguous. In practice, it is based on a confusion of the ideas of territory and resident. The international systems of national accounts—and most national systems—accept the principle that all construction of fixed assets in the national territory should be included in capital formation independent of ownership. For consistency with the rest of the national accounts and with the balance of payments, it is necessary to assume—what is, at times, an unrealistic fiction—that fixed foreign-owned assets belong to a resident economic unit in which, in turn, a foreigner may own the full equity—except for direct investment a rather contrived notion. A special departure from the territorial principle for attributing capital formation refers to ships leased on time charter, where allocation on the basis of the residence of the owner is substituted. Such ships, as appears from paragraph 48 of SNA quoted above, are included in the territory of the country of the operator; from the standpoint of capital formation, however, they are part of the country of the owner in accordance with the residence principle of allocation. This is only common sense: rigid application of the territorial principle here would require a number of fictitious entries for capital formation, exports and imports of ships, and offsetting international financial flows in the national accounts to account for the leasing of ships from country to country.

Considerable ambiguity arises in the treatment of inventories that are located outside the national territory and owned by residents, on the one hand, and inventories that are located in the national territory and owned by foreigners, on the other. In the existing UN system of national accounts, changes in these inventories are treated as part of domestic capital formation on the basis of ownership in accordance with the resident rather than the territorial principle. The relevant rule is given in paragraph 181 of SNA, which reads as follows:

181. Care should be taken to see that the coverage of stock changes agrees with the definition of imports and exports. This involves including the changes in stocks of commodities owned by normal residents but located abroad and excluding the changes in stocks of commodities owned by non-residents and held in the given country.13

In recent discussions of a UN revised system of national accounts, this procedure has come under attack. One argument, which is not without merit but should not be decisive, is that of statistical expediency: we cannot obtain, it is said, accurate information on inventories owned by residents and held abroad, or on inventories owned by foreigners and held in the country; therefore, we cannot adjust trade and investment data from a territorial basis to a resident basis. The other main argument is that a territorial definition of domestic capital formation makes it consistent with that of domestic product. This argument would be valid only if domestic product were a territorial concept. But, as shown above, it is not.

Nowhere has the territorial idea been more consistently applied than in the French national accounts. These accounts, for instance, do not accept the fiction that a government operating abroad is within the national territory because it is extraterritorial. The government is not part of the national economy as far as its activity abroad is concerned. Consumption, capital formation, and international transactions, as well as production, are all defined on a territorial basis. This procedure makes the national accounts of France not fully comparable with those of other countries and specifically makes its international transactions account depart rather radically from the generally accepted conventions. It is assumed that purchases of goods and services by households and governments outside the national territory are undertaken by them in their capacity of “agents extérieurs” and are paid for with funds which they provide in their capacity of “agents intérieurs” without obtaining any quid pro quo from the standpoint of the national economy. These transactions (like transactions of foreign households and governments in the domestic territory) are treated as distribution of income (“repartition”) similar to the payment of interest and dividends, rather than international transactions in goods and services, which are confined to physical flows between territories. It thus results in interpretations of transactions in terms that appear somewhat unrealistic from the standpoint of the “decision units” undertaking them and, hence, also from that of economic analysis. In this respect, the French system of national accounts appears to be less satisfactory than the international systems, which, though conceptually less precise, do embody a more practical approach.

Territorial Bias of Resident Concept

It would be unrealistic to deny that the scope of the economic activity to be recorded in the national accounts is largely determined by the national territory. This is because it is, by and large, the economic activity in the national territory which determines the welfare of the country’s population and which can, as a practical matter, be the subject of economic policy. The definition of resident, as used in the national accounts and the balance of payments, takes account of this politico-economic reality and, accordingly, has a strong territorial bias. Resident means resident of the national territory, and the concept covers all economic units that have a certain permanent association with it. I submit, however, that once resident has been defined in terms of the national territory, the latter concept should be discarded altogether from the national accounts. From that point on, it can only confuse rather than help thinking.

This is not the place to define resident in detail but only to outline its broader application. It has been indicated above that all fixed assets in a country are assumed to be owned by residents, although this may sometimes be carrying the territorial principle a little too far. All business enterprises operating in a country are thus considered to be residents even if they are subsidiaries or branches of foreign enterprises. The national government is always a resident, as are its personnel, wherever they operate. Conversely a foreign government and its personnel are always foreigners even when they operate on the compiling country’s territory. The tendency is to treat individuals as residents of the country where they work unless their stay in that country is only of a very short duration, for example tourists and border workers. The concept of resident is defined flexibly, permitting the national accounts to be adjusted as much to a territorial principle as is considered realistic. But use of the resident concept permits stopping short of unrealistic interpretations to which we may be forced by a rigid application of a territorial concept.

Toward a More Rational Set of Basic Definitions

It is suggested above that all the conventional national accounts aggregates can be defined in terms of the resident concepts. Below is a suggested set of definitions:

Domestic product (gross or net)Value added in production in those economic units regarded as residents1
National product (gross or net)Value added by production factors owned by residents1
ConsumptionConsumption by residents
Capital formationConstruction of fixed capital, and change in inventories, owned by residents2
Exports and importsTransactions in goods and services (including or excluding factor services) between residents and foreigners

Domestic and national product valued at market prices include indirect taxes net of subsidies.

Real assets of foreign-owned direct investment companies in a country are assumed, in accordance with normal accounting practice, to be the property of the subsidiary or branch in which they are employed, i.e., to be owned by residents.

Domestic and national product valued at market prices include indirect taxes net of subsidies.

Real assets of foreign-owned direct investment companies in a country are assumed, in accordance with normal accounting practice, to be the property of the subsidiary or branch in which they are employed, i.e., to be owned by residents.

Whatever the usefulness of domestic product in certain types of analysis (say, in input-output computations), national product appears to be more relevant than domestic product from the standpoint of the balance between aggregate supply and demand in an economy. This is because national product coincides with income that accrues to residents and is generated by production, whereas domestic product does not. However, while national product is a better expression than domestic product of the resources available to a country, it is not the best. These resources also include the transfer income which a country’s resident receives from abroad, but exclude the transfer income they pay to nonresidents. National product (at market prices), plus or minus such income transfers, yields what has been termed national disposable income in a system of national accounts recently proposed by the United Nations. This is the total income available to residents. Deducting consumption (of residents) from national disposable income, we obtain saving of the whole economy defined as comprising all residents. National disposable income and saving are two further national accounts aggregates that can be added to the list set out above.

It is particularly when a system of national accounts is described by sectors and is extended to cover financial transactions that the value of the resident rather than the territorial approach becomes evident. A sector of the domestic economy is defined in terms of a certain type of economic unit (households, the central government, nonfinancial enterprises, the central bank, deposit money banks, insurance companies, and other financial institutions). Thus, these sectors represent a sub-classification of the economy conceived as the totality of residents. The economy defined on a territorial basis does not lend itself to a meaningful subclassification into sectors.

Is Change to Resident Definition Significant?

Why should we be concerned with clarifying the definition of the national accounts by discarding the territorial concept from its underlying principles? Is it not enough that we have a definition that gives a general idea of what these accounts are all about, even if it is not completely accurate? After all, since everyone knows what we are dealing with, why bother about describing it precisely?

Some people may think clarification is merely a matter of taste. Most, however, will undoubtedly find it more satisfactory to apply a conceptual framework that puts every transaction affecting the national accounts, and not just most of them, into place. Much is gained in clarity and simplicity of exposition by applying a fully logical conceptual framework. Although the inconsistencies in the basic concepts of the conventional systems of national accounts may not be too bothersome, it is perhaps to be expected that a system providing for greater terminological clarity such as that outlined above would seem more attractive and more likely to prevail over the long run. It would represent a happy combination of a pragmatic approach and logical precision.

Les notions de territoire et de résident dans la comptabilité nationale


Un Système de comptabilité nationale des Nations Unies (A System of National AccountsSNA) et de nombreux autres systèmes utilisés à l’échelon national, définissent l’économie d’un pays de deux façons différentes, à savoir, en fonction du territoire et en fonction des résidents, c’est-à-dire des cellules économiques associées d’une manière relativement permanente au territoire national. On peut, conformément à ces définitions alternatives, considérer le produit en tant que produit intérieur ou en tant que produit national. Par définition, le produit intérieur est la valeur ajoutée sur le territoire national et le produit national la valeur attribuable aux facteurs de production appartenant à des résidents. Le produit national est égal au produit intérieur plus les revenus d’investissements et les autres revenus des facteurs reçus de l’étranger, et moins les revenus d’investissements et autres revenus des facteurs versés à l’étranger.

La présente étude se propose de montrer que la définition territoriale de la production ne concorde pas avec les méthodes utilisées pour calculer le produit intérieur. En effet, le produit intérieur ne recouvre pas la valeur ajoutée sur le territoire intérieur, mais la valeur ajoutée aux cellules économiques résidentes. La notion de territoire n’est utile que pour définir le terme de résident, et il conviendrait de la rejeter en tant que notion fondamentale de la comptabilité nationale. Le présent article propose une série de définitions des notions fondamentales de la comptabilité nationale, telles que le produit intérieur, le produit national, le revenu national disponible aux prix du marché, la consommation, l’épargne et la formation de capital, toutes étant fondées sur la notion de résident plutôt que de territoire. L’utilisation de cette série de définitions permettrait de supprimer les contradictions de la méthodologie de la comptabilité nationale, inhérentes à la notion territoriale, sans pour autant modifier la signification des catégories pragmatiques en cause, telles qu’elles se sont dégagées au cours des années.

Los conceptos de territorio y de residente en las cuentas sociales


En Un Sistema de Cuentas Nacionales de las Naciones Unidas (A System of National AccountsSNA) y en muchos otros sistemas nacionales de dichas cuentas, la economía de un país se define de dos maneras: en función del territorio y en función de los residentes del país, es decir, de aquellas unidades económicas que tienen cierta asociación permanente con el territorio nacional. Conforme a estas definiciones alternativas al producto se le considera de dos maneras: como producto interno y como producto nacional. El producto interno se define como el valor añadido a la producción en el territorio nacional, y el producto nacional como el valor atribuible a los factores de producción de propiedad de residentes. El producto nacional es igual al producto interno más las sumas recibidas del exterior por concepto de ingresos provenientes de inversiones y de otros factores de producción y menos las sumas pagadas al exterior por ese mismo concepto.

En este estudio se mantiene que la definicóon territorial de producción no concuerda con las prácticas que se siguen para calcular el producto interno. En realidad, el producto interno no abarca el valor que se añade en el territorio nacional, sino el valor añadido por las unidades económicas residentes. El concepto de territorio no sirve sino para definir a los residentes, y debe descartarse como concepto básico de las cuentas nacionales. Este artículo propone una serie de definiciones de conceptos fundamentals de las cuentas nacionales, como son producto interno, producto nacional, ingreso nacional disponible a precios de mercado, consumo, ahorro y formación de capital, que se basan en el concepto de residente y no en el de territorio. El uso de esa serie de definiciones ayudará a eliminar de la metodología de las cuentas nacionales ciertas incongruencias inherentes al concepto territorial, sin que haya que modificar el contenido de sus categorías pragmáticas, que han ido evolucionando con el transcurso del tiempo.


Mr. Høst-Madsen, Deputy Director in the European Department, was formerly in the Fund’s Research and Statistics Department. He is a graduate of the University of Copenhagen. Before joining the Fund staff in 1946, he was with the Danmarks Nationalbank.


United Nations, Statistical Office, Final Report, UN-OEEC-IMF Discussions to Coordinate the Balance of Payments Manual, A Standardized System of National Accounts and a System of National Accounts and Supporting Tables, New York, November 16, 1956 (mimeographed).


United Nations, Statistical Office, A System of National Accounts and Supporting Tables, Studies in Methods, ST/STAT/SER.F/2/Rev.2 (New York, 1964), hereinafter cited as SNA.


In other words, the definition represents an incorrect generalization based on existing practice.


SNA, p. 7.


For simplicity of exposition, this statement disregards the existence of indirect taxes and subsidies. In fact, payments to factors of production may exceed value added because of subsidies, or fall short of it because of indirect taxes. Subsidies and indirect taxes create a difference between product measured at factor cost and at market prices, but whether one or the other valuation basis is adopted, the product, conceived as a real flow, can equally be ascribed to the input of factor services.


SNA, loc. cit.


The charter hire should not be confused with the net income (after depreciation and other cost, if any) on the financial capital invested in the ship, which is part of the profit of the enterprise owning it. Such profit constitutes factor income whether retained in the enterprise or paid out in the form of interest and dividends.


SNA, loc. cit.


This does not mean, however, that the two are necessarily different in substance. The payment by a government or a household for personal services (say, the services of a government official or a domestic servant) refers at the same time to consumption of product (i.e., nonfactor services) and to the services of factor of production. The distinction of the two aspects of the transaction reflects merely their different function within the given accounting framework.


G. Stuvel, “A System of National and Domestic Accounts” Economica, Vol. XXII (1955), pp. 207-217.


Ibid., p. 209.


G. Stuvel, Systems of Social Accounts (Oxford, 1965), p. 20.


SNA, p. 29.

Other Resources Citing This Publication