From the Organization for Economic Cooperation and Development. OECD.

TECHNICAL NOTE: NATIONAL ACCOUNTS

The conceptual framework underlying the national accounts

At present, most countries produce their national accounts on
a basis consistent with that described in the United
Nations’ publication A System of National Accounts
(SNA) which was released in 1968. However, an updated version of
the SNA was issued in 1993 under the auspices of the
Intersecretariat Working Group on National Accounts, which
consists of officials from the OECD, the International Monetary
Fund, the United Nations Statistical Division, the World Bank and
the Commission of the European Communities. Its official title is
System of National Accounts, 1993 and it is commonly
referred to as SNA93. Several countries have already released
their accounts in accordance with SNA93 and most OECD countries
are expected to produce their accounts on the basis of the
recommendations contained in this new framework by 1999.

There have been some significant changes made in SNA93 with
regard to the assets boundary, the emphasis placed on
institutional sectors, the format of the main summary tables, the
approaches to be adopted in calculating constant price estimates
and the importance of balance sheets as part of the overall
system. However, the broad concepts underlying the 1968 and the
1993 versions of the SNA are very similar and so the following
descriptions are based on SNA93.

The rationale behind national accounts is summarised in the
following three paragraphs, which have been reproduced from the
Introduction to SNA93:

Major concepts in the system of national accounts

Economic production

The main focus of the System is economic production, which
relates to both goods and services. Generally, but not
exclusively, it is goods and services purchased in “the
market” which are included in the national accounts. To
complete the picture, some parts of non-marketed goods and
services (i.e., those for which no explicit charges are made, or
which are sold at prices significantly lower than their cost of
production) are also included in the scope of the national
accounts (see the section below headed ”Imputations for
non-market production”).

Illegal activities

It should be noted that the fact that a certain type of
activity is illegal does not mean that it is outside the
production boundary, despite the difficulties involved in
collecting data on such activities. There are two broad
categories of illegal activities:

those whose production, sale or distribution is
        prohibited by law; and
those which are usually legal but which become illegal
        when carried out by unauthorised producers (e.g.,
        unlicensed medical practitioners).

The main reason for including illegal activities within the
production boundary is that errors will appear elsewhere in the
accounts if such production is omitted. For example, incomes from
illegal production can be spent on purchasing legal goods and
services and, conversely, illegal goods and services can be
purchased using incomes derived from legal sources. Both sides of
such transactions need to be recorded within the national
accounts framework otherwise discrepancies will occur in some
parts of the accounts. It follows that estimates have to be
included for the illegal activities to ensure that the various
parts of the accounts balance.

In practice, it will prove impossible for countries to make
reliable estimates of the income from illegal production.

Therefore, imbalances will be introduced into the accounts in
practice because some parts of such income will be included while
others will be excluded. For example, any illegal income which is
reported through the tax system (e.g., from prostitution) would
be included on the income side of the accounts in those countries
which use tax as a data source, while other (unreported) illegal
income would be excluded.

Unduplicated value of production

In measuring total economic production, the aim is to
calculate the “unduplicated” value of production. In
other words, it is important to avoid double counting in those
cases where the outputs of some economic units are used as inputs
by other economic units. For example, a large part of the output
of the steel industry is used as inputs by other industries such
as construction and motor vehicle manufacturing. In the national
accounts, the unduplicated value of production (referred to as
“value added”) is calculated by deducting the value of intermediate
consumption from the value of gross output. The major
aggregate in the national accounts is gross domestic product (or
GDP) which is defined in paragraph 2.172 of SNA93 as “.....
the sum of gross value added of all resident producer units
(institutional sectors or, alternatively, industries) plus that
part (possibly the total) of taxes, less subsidies, on products
which is not included in the valuation of output. Gross value
added is the difference between output and intermediate
consumption.”

"Net" versus "gross" measures

It is important to recognise that, as noted above, GDP is a
measure of unduplicated production despite the use of the word
“gross” in “gross domestic product”. The
reference to “gross” indicates that consumption of
fixed capital is included in the total. In other words, the
extent to which the economy’s productive assets are used up
over time in producing output has not been deducted from total
production when calculating GDP. The estimate of overall
production which also takes into account the fact that capital is
being used up in the course of production is “net domestic
product” (NDP). While NDP provides a purer measure of
production than GDP, placing a value on consumption of fixed
capital is a difficult process. In practice, consumption of fixed
capital is not calculated on a consistent basis between countries
and so GDP is generally preferred to NDP as the summary measure
of a country’s economic activity.

Alternative methods of deriving GDP

As well as the approach defined above, there are two other
alternative methods of deriving GDP. The first of these is by
summing the final uses of goods and services (i.e., final
consumption, gross fixed capital formation and exports but
excluding intermediate consumption) measured in purchasers’
prices, less the value of imports of goods and services. The
second is by summing primary incomes distributed by resident
producer units (i.e., compensation of employees, mixed incomes,
gross operating surplus and net taxes on production and imports).

Imputations for non-market production

The costs of producing non-marketed goods and services are met
largely, if not wholly, by the person or organisation producing
them so that the goods and services produced can be sold at
prices significantly less than those which would be obtained for
similar products on the market. Examples are most goods and
services produced by non-profit organisations (such as sporting
clubs and churches) and general government bodies (such as the
ministry responsible for administering social security
programmes). It should be noted that the latter organisations do
not include public corporate enterprises such as government-owned
airlines or telecommunications bodies.

An important proviso relating to non-marketed production is
that it is feasible to include an estimate in the national
accounts only for those goods and services for which it is
possible to impute a value on a systematic basis. Typically,
there are two methods of doing so. The first is by using the
prices of similar or closely related market transactions. The
second is by assuming that the output is equal to the sum of
inputs (i.e., the total cost of the compensation paid to the
employees of such organisations and the cost of the goods and
services used by those organisations in the course of production,
including an allowance for the consumption of fixed capital).

Constant price estimates

When data are obtained from, for example, a survey of
businesses, the underlying values are expressed in terms of the
prices of the period for which the data were collected. In
practice, however, prices change over time and so changes in the
values of national accounting aggregates include the effects of
both price and quantity changes. A significant amount of economic
analysis involves examining underlying volume changes and so it
is important to identify means of systematically removing the
effects which price changes have on the values recorded in each
period. Deriving “constant price estimates” is the
means by which this is done.

Constant price estimates are obtained by expressing values in
terms of a base period. In theory, the price and quantity
components of a value are identified and the price in the base
period is substituted for that in the current period. Two main
methods are adopted in practice. The first, referred to as
“quantity revaluation”, is based on a methodology
consistent with the above theory (i.e., by multiplying the
current period quantity by the base period price). The second,
commonly referred to as “price deflation”, involves
dividing price indexes into the observed values to obtain the
volume estimate. The price indexes used are built up from the
prices of the major items contributing to each value.

For some components of the accounts, it is not possible to
identify an underlying quantity and price. For example, gross
operating surplus is derived as the amount left over after wages
paid to employees and intermediate consumption have been deducted
from a business’ total income. As a result, there is no
quantity associated with such a measure, so it is not possible to
revalue the quantities of these components.

In practice, estimates obtained by deflating components such
as gross operating surplus using a price index provide a measure
of the purchasing power of that component over the basket
of goods and/or services included in the deflator. They are
referred to as “real” estimates and they are different
from constant price estimates because they do not reflect any
underlying volume. Another measure which is often of interest
when expressed in real terms is “household saving”. It
is also similar to gross operating surplus in that it cannot be
split into underlying price and quantity components. At present,
no “real” estimates of this type are included in the
OECD’s national accounts publications. However, the possible
inclusion of such estimates will be investigated when the
majority of OECD member countries move to the conceptual basis
described in SNA93 in early 1999.

Purchasing power parities (PPPs)

International comparisons of GDP depend on two conditions
being met. The first is that the basis of calculating GDP is
consistent for the countries under comparison. The second is that
the unit in which GDP is expressed, the numeraire, is comparable.

The simplest way of comparing the GDP of two different countries
is to convert each amount (or, even better, the amount per
capita) to a common currency using official exchange rates.

However, this is widely recognised as being inadequate because
official exchange rates do not adequately reflect the comparative
purchasing power of local currencies in their own markets. The
concept of the purchasing power parity (PPP) was developed to
provide an alternative conversion factor for GDP, so that
internationally comparable price and volume comparisons of GDP
could be established.

PPPs are the rates of currency conversion which equalise the
purchasing power of different currencies. This means that a given
sum of money, when converted into different currencies at the PPP
rates, will buy the same basket of goods and services in all
countries. In other words, PPPs are the rates of currency
conversion which eliminate the differences in price levels
between countries. Thus, when expenditures on GDP for different
countries are converted into a common currency by means of PPPs,
they are expressed, in effect, at the same set of international
prices so that comparisons between countries reflect only
differences in the volume of goods and services purchased.
Expenditures converted to a common currency using exchange
rates, on the other hand, reflect not only differences in the
volumes purchased in the different countries, but also the
differences in price levels between the countries. Exchange rates
do not reflect the relative purchasing powers of different
currencies and are not, therefore, the appropriate currency
conversion rates with which to make international comparisons of
volume. Expenditures on GDP converted at exchange rates remain
essentially nominal measures; the same expenditures
converted using PPPs are real measures.

PPPs also appear in international trade theory in the context
of equilibrium exchange rates - that is the underlying rates of
exchange to which actual exchange rates are assumed to converge
in the long term. The PPPs produced by the OECD are not relevant
for this purpose as they do not refer solely to
domestically-produced tradable goods and services valued at
export prices. Instead, they have been calculated specifically
for statistical purposes in order to enable international
price and volume comparisons to be made for expenditure on GDP
and its components. As such, they refer to the entire range of final
goods and services which make up GDP as a whole including
many items, such as construction and government services, which
are not traded. Moreover, they are valued at domestic
market prices and are calculated using expenditure weights that
primarily reflect domestic demand.

Both real and nominal
expenditures are presented in OECD publications. The real values
are expressed at international prices in US dollars; the
nominal values are expressed at national prices in both US
dollars and national currency. (The choice of the US dollar as
the common unit of currency is purely a matter of convention
which has no effect on the relative positions of countries.)

Inclusions in, and exclusions from, the production boundary

During the process of revising the SNA, there was a
wide-ranging debate on what should be included within the
production boundary of SNA93. The following paragraphs, taken
from chapter 1 of SNA93, describe what is included in the
production boundary and why certain components, such as unpaid
household work, were excluded.

Satellite accounts

One of the major advantages of the SNA is that it provides a
comprehensive framework for collecting, compiling and presenting
economic statistics. As a result, pressures have built to extend
the national accounts to include other interesting aspects such
as the impact of economic activity on the environment. A major
difficulty in doing so is that, because such factors are not part
of “the market” in a country, it is impossible to
directly put a monetary value on them. The solution proposed in
SNA93 is to produce such estimates in the form of satellite
accounts, which are accounting statements consistent with,
but compiled and presented separately from, the main (or core)
accounts. The major advantage of such an approach is that the
core accounts are quarantined from the conceptual and measurement
problems which can have a large impact on the estimates for
environmental factors.

Another area for which satellite accounts could be produced is
unpaid household work. As noted in the previous section, unpaid
household services are excluded from the national accounts
production boundary. While there are significant conceptual and
practical problems involved in producing estimates of the value
of unpaid household work, there is also a great deal of interest
in such estimates. A practical way of meeting this demand is to
produce the estimates in the form of a satellite account, which
enables direct comparisons to be made with different components
of the core accounts, as well as with the overall value of GDP.

Uses of the national accounts

The national accounts provide a detailed set of statistics on
various parts of the economy as well as a summary indicator (GDP)
of economic activity. While the principal use of the accounts is
by governments as a basis for formulating economic policy, they
are also used by a wide range of others. Groups such as private
businesses (particularly financial institutions), journalists,
business lobby groups, academics, trade unions, and the general
public all use the accounts. They form the starting point for
assessments of the current economic situation and for forecasts
which may indicate a need for changes in the direction of fiscal
or monetary policy. Making international comparisons is also an
important use of national accounts and one of the aims of SNA93
was to further improve the international comparability of
national accounts statistics.

Conclusion

The national accounts are designed to provide a coherent and
consistent set of data for various types of economic analysis.

The major summary aggregate, GDP, provides an estimate of the
overall unduplicated production in a country’s economy. It
is an important statistic because, in the longer term, the level
of a country’s production broadly determines the amount of
goods and services which can be consumed by the country’s
inhabitants. However, while being an important indicator of a
country’s economic well-being, GDP is not necessarily a
reliable indicator of a country’s overall community welfare
or quality of life. A number of other, non-economic, indicators
such as public health, education levels, crime rates and
environmental quality would have to be taken into account in any
such measure.
---------------------------------------------------------------