Sam 1
Sam 1
Sam 1
Elsenburg
September 2003
Overview
The Provincial Decision-Making Enabling (PROVIDE) Project aims to facilitate policy
design by supplying policymakers with provincial and national level quantitative policy
information. The project entails the development of a series of databases (in the
format of Social Accounting Matrices) for use in Computable General Equilibrium
models.
The National and Provincial Departments of Agriculture are the stakeholders and
funders of the PROVIDE Project. The research team is located at Elsenburg in the
Western Cape.
Private Bag X1
Elsenburg, 7607
South Africa
[email protected]
+27-21-8085191
+27-21-8085210
For the original project proposal and a more detailed description of the project,
please visit www.elsenburg.com/provide
PROVIDE Background Paper 2003: 4 September 2003
Abstract
1 The main author of this paper is Kalie Pauw, Senior Researcher of the PROVIDE Project.
i
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
Table of contents
1. INTRODUCTION ........................................................................................................................................ 1
2. AN OVERVIEW OF SAMS .......................................................................................................................... 1
2.1. Economic accounting .................................................................................................................... 1
2.2. Circular flow in the economy ........................................................................................................ 3
2.3. The accounts of a SAM .................................................................................................................. 5
3. A SAM APPROACH TO MODELLING ........................................................................................................ 14
4. CONCLUDING REMARKS AND FUTURE RESEARCH ................................................................................... 17
5. BIBLIOGRAPHY ....................................................................................................................................... 18
List of figures
List of tables
ii
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
1. Introduction
Social Accounting Matrices (SAMs) have become the database of preference for most
macroeconomic modellers. A SAM is a comprehensive, economy-wide database that contains
information about the flow of resources associated with all transactions that have taken place
between economic agents in a certain economy during a certain period of time. As such it
presents a snapshot picture of the economy at hand.
Initial theoretical developments in social accounting are largely attributable to Sir Richard
Stone who addressed the matter of integrating disaggregated production accounts (in the form
of input-output systems)2 into the national accounts. The aim was to form an economy-wide
database, which not only included information about productive activities in the economy, but
also incorporated other non-productive institutions and markets, such as factor markets,
capital markets, households, government, and the rest of the world.
This paper provides an overview of social accounting and is organised as follows. Section
2 provides an overview of social accounting and the way in which the circular flow of
resources is captured in a SAM. Section 3 describes the SAM approach to modelling. Section
4 makes some concluding remarks.
2. An overview of SAMs
When economic agents are involved in transactions with each other financial resources
exchange hands. The first objective of a SAM is to organise data. For this purpose accounts
are included in the SAM to represent agents that are involved in economic transactions.
Transactions are captured in the relevant accounts of the SAM, showing the values and
direction of the flow of resources. A SAM thus forms a complete database of all transactions
that take place between agents in a given period, presenting a ‘static image’ or ‘snapshot
picture’ of the structure of an economy for that period.
2 Input-output analysis was developed by Wassily Leontief in the 1930’s. Its primary aim is to provide a tool to
analyse the production side of the economy, with a specific focus on the intermediate input requirements
and final outputs of industries.
1
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
SAM accounts are created for all economic agents, including producers, households,
government and the rest of the world. Accounts also exist for commodity, factor and capital
markets. Some SAM accounts can therefore be seen as intermediate clearing accounts. For
example, the transactions recorded in a SAM are not limited to the purchase or selling of
goods and services but can incorporate any type of transaction. This includes transactions that
take place during the production process such as the purchasing of intermediate goods and
hiring of factors. It also includes current account transactions of institutions (households,
enterprises and government), such as inter-institutional transfers, consumption expenditure
and the payment of various taxes. It further includes capital account transactions of
institutions, such as savings and investments. Finally, it can include any transaction that takes
place across international borders, such as direct foreign investment and international trade
transactions.
The consistency of a SAM stems from the fact that it adheres to the fundamental principle
of economics described in the quotation above. In financial accounting double entry
bookkeeping is performed with the aid of various T-accounts with debit and credit entries. For
example, a government transfer of R1000 to households would be accounted for as follows
(Figure 1):
Since resources flow from the government account to the households account, the former
is debited and the latter is credited with R1000. The SAM approach to bookkeeping is much
more concise. A SAM is a square matrix where column of the SAM represents a specific
account in the economy. The exact same accounts also appear in the rows. Incomes or receipts
2
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
are shown in the rows of the SAM, i.e. if households receive R1000 the amount should appear
in one of the cells along the household row account. Expenditures or outlays are shown in the
column, i.e. if government parts with R1000 the amount should appear in the government
column account. In the government transfer example used here the transfer can be entered in
the cell where the government column and household row accounts intersect (Table 1). In
general, a single-cell entry in row i and in column j represents a flow of resources from
account j to account i.
In financial accounting each account is balanced at the end of the financial period. The
debit side equals the credit side and profits or losses are matched off with an entry in a profit
or loss account. Similarly, in a SAM, the income of each account (row total) must equal the
expenditure of each account (column total). This holds for individual accounts, but also for
the economy as a whole. A SAM thus goes further than T-accounts. Where T-accounts are
balanced individually, a SAM ensures that all accounts are simultaneously balanced. For this
reason a SAM can be described as complete and consistent. Since all transactions are captured
in a SAM, a discussion of the circular flow in the economy may help to understand the
concept of a SAM better.
The conventional way to present a circular flow diagram is to show the movement of goods
and services through the economy. Since a SAM is interested in the flow of resources in the
economy, the arrows in Figure 2 point in the opposite direction.
3
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
Inter-
institutional Value added Activity/
transfers
Factors
Production
accounts Intermediate
accounts
input demand
National product Demand for
domestic goods
Notes: 1) Government, households and incorporated business enterprises are all regarded as institutions.
Although some inter-institutional tax transfers and tax transfers from factors to government are
captured in Figure 2, tax payments from activities and commodities have been omitted for simplicity
reasons.
2) Flows indicated with a star (*) can be in the opposite direction as well. These are mainly flows
associated with the rest of the world accounts. For example, if imports exceed exports there would be
a flow from the commodity account to the rest of the world account. Similarly for net transfers from
(to) abroad, the balance of payments etc.
Figure 2 can be explained by starting with the commodity accounts. Resource flows into
these accounts originate from the rest of the world (net exports), capital accounts (investment
demand), the institutional accounts (household and government consumption), and the
activity accounts (intermediate input demand). The combined income of these accounts pays
for domestic demand from the activity accounts as well as commodity taxes to the
institutional accounts (not shown).
Activities are involved in production; hence the activity accounts are also called
production accounts. Final goods supplied to the commodity accounts are produced by
employing factors that add value to intermediate inputs purchased from the commodity
accounts. These resource flows are represented by payments to the factor accounts and the
commodity accounts respectively. Factor use taxes or production taxes may also be payable to
government in the institutional accounts (not shown).
Apart from receiving remuneration from the activity accounts, factors supplement their
income with transfers from abroad (rest of the world accounts). Since factors are ‘owned’ by
institutions, factor account income is paid to the institutional accounts. Typically households
own capital and/or labour, while enterprises and government may own capital. Since
transactions also take place between institutions, inter-institutional transfers are also included
4
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
in the figure. These include tax payments from households and enterprises to government,
inter-household transfers, transfers from government to households such as social security
grants, and all other inter-institutional transfers. Inter-institutional transfers can easily be
identified in the more detailed institutional breakdown typically used in a SAM (see section
2.3). Net transfers from abroad supplement institutions’ income, provided these are positive.
All income not spent by households, enterprises and government is saved. Government
savings can be negative (budget deficit) or positive (budget surplus). Institutional savings flow
from institutional accounts to the capital accounts. A positive balance of payments would also
contribute to the pool of savings. The entire pool of savings is utilised for investment
purposes. This is shown here as a flow from in the capital accounts to the commodity
accounts, i.e. investment goods are purchased from the commodity accounts. The following
section proceeds to explain how these flows can be represented in a SAM-format that uses the
same accounts as those in Figure 2. It will however be shown that greater detail is now
possible since accounts can easily be disaggregated into various sub-accounts.
SAMs are usually constructed with six types of accounts that relate directly to those in Figure
2 above. Each of these accounts can be disaggregated further should more detailed
information be available or necessary from a modelling perspective. Table 4 shows the
structure of a SAM using these accounts. In this SAM the institutional accounts are
disaggregated into separate accounts for households, incorporated business enterprises and
government. Each account is represented by a row and a column in the SAM, where row
entries represents flows into the account, while column entries represent flows from the
account. An important feature is that each account must be balanced, i.e. flows into the
account must match flows from the account.
The commodity accounts show all transactions relating to intermediate input demand,
institutional consumption demand and investment demand. The row accounts show how
commodities are distributed between intermediate input demand (USE matrix) and final
demand. Final demand is made up of consumption demand by households, governments and
enterprises (if applicable). It also includes investment demand and export demand from the
rest of the world. All domestically consumed commodities are valued at the same price, which
is inclusive of all relevant sales taxes and tariffs. Thus, all prices along the row are the same
irrespective of which agent purchases the commodity. This is the so-called law of one price.
The inclusion of exports in the same row is slightly misleading, as exports are usually valued
at export prices, which are a function of exogenously determined world prices.3
3 See Technical Paper 2003: 3 for a more detailed discussion. Also see the discussion in section 3.
5
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
Domestic production is contained in the MAKE matrix along the row of the activity
accounts and the commodity account column. Total domestic supply is made up of
domestically produced goods as well as imported goods. The SAM therefore separates
imports from exports. Apart from expenditure on intermediate inputs, activities pay for value-
added, which is represented by a flow from the activity accounts to the factor accounts.
Adding production taxes gives total activity expenditure, which equals activity income from
the commodity accounts.
Factors earn income from employment in domestic production activities. They can,
however, also earn income from abroad. The income of the factor capital is usually in the
form of profits and rent, while labour earns wages. The factor column account shows how
factor income is disbursed. Institutions, the owners of factors, receive income from factors net
of factor taxes such as social security payments or taxes on profits. Factor income is typically
in the form of wage-income or distributed profits. Transfers from factors to the rest of the
world can also be included in the SAM.
There are three institutional accounts in the SAM in Table 2. Households’ usually earn the
majority of their income from factors, but can also supplement this with transfers from other
institutions or the rest of the world. Total household income is distributed between
consumption, transfers to other households, direct taxes and savings. Incorporated business
enterprises earn income from non-distributed firm profits and transfers. Surplus income is
distributed between taxes and transfers to other institutions, transfers to the rest of the world,
and enterprise consumption (if applicable) and savings. Government receives income from
various tax sources, from the ownership of factors (if applicable), and from transfers from
other institutions and the rest of the world. Expenditure is made up of transfers and
government consumption demand.
The capital accounts contain all transactions relating to investment and the funding
thereof. Domestic institutions and the rest of the world can contribute to the pool of savings as
shown in the capital row account. Savings are injected in the economy in the form of direct
investment, which is captured as demand for commodities. Finally, the rest of the world
account(s) record all transactions between domestic institutions, factors or commodity
accounts and the rest of the world.
An important feature of a SAM is that the jth row total should equal the jth column total,
i.e. for each account, income or receipts should equal its expenditure or outlay. This feature of
a SAM ensures compliance with the basic economic accounting principle. If this does not
hold, the SAM is either incomplete or inconsistent. This means that certain transactions are
not recorded, or certain SAM entries do not properly match the agents involved in the
transactions. Consider the previous example of a government transfer to households. If
6
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
households report income from transfers of R980, while government budget accounts suggest
that the transfer was R1000, the accounts are not reconciled and the SAM will not balance.
There is also an important difference between balancing a SAM and estimating a SAM.
When SAM-entries are missing, but there is reason to believe that the specific transaction did
take place, it is necessary to estimate the missing value. Once all available data, whether
observed or estimated, has been included in the SAM, it is still unlikely to balance due to data
inconsistencies and estimation errors. Such a SAM then has to be balanced. Various statistical
techniques can be used to remove ‘small’ errors, which will ensure that the SAM balances.
The difficulty with balancing a SAM has to do with the fact that any change to row entries
will automatically affect various column totals. Rows and columns must therefore be adjusted
simultaneously. A discussion of the various balancing/estimation methods falls beyond the
scope of the paper.
7
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
6. Government Sales taxes, tariffs, Indirect taxes, Factor income to Transfers to Transfers to Transfers to Government
export taxes factor use taxes government, factor government, direct government, direct government from income
taxes household taxes enterprise taxes RoW
8. Rest of the Imports Factor income to Transfers to RoW Government Foreign exchange
RoW transfers to RoW outflow
world
TOTAL Supply Activity Factor Household Enterprise Government Investment Foreign exchange
expenditures expenditures expenditure expenditure expenditure inflow
Source: Löfgren et al (2001)
8
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
Next we consider some simple examples, first of a closed economy without a government
sector, and secondly of an open economy with a government sector. These examples show
how the various accounts can be disaggregated further to include more detailed information
about the structure of the economy.
In this example data from a basic two-sector economy with only firms and households is
provided (adapted from Löfgren, 1999). There are two markets, one for factors and one for
commodities. Households earn all their income from labour and capital, which they supply to
the factor market. They spend all their income on consumption, i.e. there is no capital account
to which they can transfer their savings. There is also no government sector and hence
households do not pay taxes nor do they receive income from government transfers. There is
also no foreign sector, i.e. our economy is closed. In this example, as with the one that follows
(section 2.3.2), it is assumed that each industry only produces a single product. This implies a
diagonal MAKE matrix. In larger more disaggregated SAM’s it is often necessary to relax this
limiting assumption by allowing multi-product producers, i.e. the MAKE matrix will then
have off-diagonal entries.
AGR-C 60 40 50 75 225
NAGR-C 40 60 100 50 250
AGR-A 225 225
LAB 62 55 117
CAP 63 95 158
U-HHD 60 90 150
R-HHD 57 68 125
9
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
producers. Factors of production consist of labour (LAB) and capital (CAP). The only
institutions included in the model are households. Households are disaggregated into urban
(U-HHD) and rural (R-HHD) households. No government sector is included. Since this is a
closed economy there is also no account for the rest of the world.
The amounts that appear in this SAM all refer to the flow of resources between agents in
monetary terms. Receipts appear in the rows, while expenditures appear in the columns.
Agricultural activities purchase agricultural commodities (60) and non-agricultural
commodities (40) as intermediate inputs into their production process. The amounts paid for
labour (62) and capital (63) represent value added. The total value of agricultural production
amounts to 225. In the absence of government no production taxes are payable. The
agricultural commodities are sold on the agricultural commodity market for 225. Similarly,
non-agricultural activities produce goods to the value of 250, which is then also sold on the
non-agricultural commodity market for 250. The income of activities is captured as an entry in
the activity row and the commodity column. The income and expenditure of all activities are
thus in equilibrium.
In this SAM we make the assumption that households derive their entire income from the
factor market, i.e. we assume that households own factors of production and supply these
factors on the factor market in exchange for remuneration. This remuneration is exactly equal
to the value-added payments of activities. Urban households earn 60 from labour and 90 from
capital employed in the production process. Rural households earn 57 from labour and 68
from capital. The total income earned by households is spent on consumption. Thus, in the
absence of a capital account (savings) and a government account (taxes), household income
and expenditure balances out.
2.3.2. Example 2: Open economy with a government sector and a capital account
In the second example a government sector is added, as well as a capital account for the
institutions (households and government) and an account for the rest of the world. The
10
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
inclusion of the government sector has tax implications for various agents. Various tax
accounts are included, namely accounts for income tax (YTAX) payable by households, sales
taxes (STAX) payable by the commodity market and import tariffs (TAR) payable by the
commodity market on imported goods. A core government account (GOV) represents the
government budget. The capital account (S-I) or savings-investment account captures savings
and investment by institutions. The inclusion of the rest of the world account (ROW) also
allows for foreign capital and current account transactions to be captured. Consider Table 4
below.
11
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
Table 4: SAM for an open economy with a government sector and a capital account
AGR-C NAGR-C AGR-A NAGR-A LAB CAP U-HHD R-HHD GOV S-I YTAX STAX TAR ROW TOTAL
AGR-C 84 55 30 49 13 28 30 289
R-HHD 82 83 5 16 186
GOV 25 30 39 15 109
S-I 70 40 -1 4 113
YTAX 20 5 25
STAX 10 20 30
TAR 39 39
TOTAL 289 558 279 394 177 208 285 186 109 113 25 30 39 105
Source: Löfgren (1999)
12
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
The column account of the commodities market is different to the previous SAM due to
the introduction of government. The commodity account is now required to pay sales taxes to
the amount of 10 for agricultural commodities and 20 for non-agricultural commodities.
Furthermore, import duties are payable on imported commodities. We make the assumption
that only non-agricultural commodities are imported, hence no import tariff is payable on
agricultural commodities. Non-agricultural commodities to the value of 105 are imported
from the rest of the world. A tariff of 39 is payable on these imports. The row account of the
commodity market is also altered. As before, activities consume intermediate goods
households consume final goods. However, in addition to this, government and the rest of the
world also now consume goods. Government spends 13 and 67 on agricultural and non-
agricultural products respectively, while the rest of the world pays 30 for agricultural
commodities. We assume that only agricultural commodities are exported to the rest of the
world.
The column (expenditure) accounts of activities remain the same as before, i.e. producers
purchase intermediate inputs from the commodity markets and add value through the use of
factors of production, labour and capital. The total value of agricultural production is 279,
while the value of non-agricultural production is 394. As before, producers sell these products
on the commodity markets and receive revenue equal to the amounts spent on intermediate
inputs and factors (value added).
The factor accounts are unchanged. Factors receive income from value added payments by
producers. Households, the owners of these factors, in turn derive part of their income from
factors. In some cases factor taxes (e.g. a capital asset tax) can be included. However, in this
SAM no provision is made for factor taxes. The column accounts of households now include
household savings (70 and 40 for urban and rural households respectively), as well as
payments to the income tax account of government (20 and 5 for urban and rural households
respectively). Household income is supplemented by transfers from government (25 and 5 for
urban and rural households respectively) and transfers from the rest of the world (40 and 16
for urban and rural households respectively).
Government receives revenue from the various tax sources (25, 30 and 39 from income
taxes, sales taxes and import tariffs) as well as transfers from the rest of the world (15).
Government thus receives a total income of 109 from these sources. This can now be spent on
government consumption expenditure and transfers to households. The remainder of
government revenue is saved. In this example government savings are negative, i.e. they run a
budget deficit and need to borrow money in order to balance budget. An alternatively to this
negative entry in the SAM would have been to include a +1 in the capital account column.
13
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
The income of the capital account includes institutional savings and the balance of
payments, while expenditure is made up of investment demand. Investment is captured in a
similar way as consumption of institutions, i.e. it appears in the commodities row account.
Since all the other accounts are in balance, the rest of the world account will be in balance.
The rest of the world account receives income from the sale of exports. As mentioned above,
the commodity market imports non-agricultural commodities to the value of 105. The rest of
the world imports 30 worth of agricultural commodities. Domestic institutions also receive
transfers from the rest of the world. The balance of payments is equal to 4.
The preceding sections showed how a SAM provides a quantitative picture of the economy by
summarising the resource flows between agents that are involved in transactions. Pyatt (1988)
refers to this as the transactions value (TV) approach to social accounting. However, the
SAM-format can also be used to present economic theory, i.e. it can provide both an empirical
and a theoretical description of the economy. This section draws on Pyatt’s “SAM approach to
modelling” (1988).
t jk = t jk ( y; p, f , λ ) [3.1]
Pyatt (1988) uses a simple proof to show how prices in a column account for commodities
or activities are interdependent although they relate to different commodities. The vector of
row-sums of a SAM represents account incomes or total revenue (TR). Similarly, the vector
of column-sums represents total outlay or cost (TC). The basic underlying accounting identity,
which requires that the sum of the jth row should equal the sum of the jth column, implies that
TR equals TC and hence average revenue (AR) equals average cost (AC), which is dependent
14
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
on prices. Therefore, prices are interdependent and column summation of the commodity or
activity accounts in a SAM yields a set of equations
p = p( y; p, f , λ ) [3.2]
Equation 3.2 provides a general description of the price system and is the first set of three
sets of equations of a macroeconomic (general equilibrium) model. The set of equations are
linear homogenous, i.e. if input prices are doubled, output prices will also double, provided
that the scale of production stays constant.4 The interdependency of prices can easily be
explained with the aid of a 3x3 matrix such as the one below, which represents a small SAM
(commodity/activity accounts):
Each matrix entry represents the transaction value and is made up of quantities and prices.
Thus, the entry piqij represents the transaction between accounts i (row, income) and j
(column, expenditure), valued at price pi. Summing along the row k of the matrix is possible,
as prices and quantities are the same (they relate to the same commodity). Thus, the revenue
of account k can be expressed as follows:
TC k = ∑ pi q ik [3.4]
i
But, given that TC k = TRk , it follows that a change in pi (i ≠ k) in a given column account
that affects TCk, will lead to a change in TRk and hence pk has to adjust to maintain the
balance. However, when pk changes it affects the TC and hence TR of the other columns as
well. Thus, all prices are interdependent, as can be seen from the worked example below for
k = 1:
4 This is ensured by using linearly homogenous production functions to model production in a model.
15
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
TR1 = TC1
p1 q11 + p1q12 + p1 q13 = p1q11 + p 2 q 21 + p3 q31
p1 (q12 + q13 ) = p 2 q 21 + p3 q 31
p 2 q 21 + p 3 q31
∴ p1 =
q12 + q13
Pyatt (1988) makes use of a simple exposition to show how a system of equations for an
economic model can be developed with the SAM as basis. The basic principle underlying
Pyatt’s SAM approach to modelling is that each entry in a SAM can contain either exogenous
transaction that are independent on income or the scale of production or endogenous
transactions that are dependent on income. The matrix of endogenous transactions is denoted
by N and the matrix of exogenous transactions by X. A SAM can then be written as the sum of
the endogenous and exogenous matrices, i.e. T = N + X, where T is a square matrix or SAM.
If y is the vector of row sums, i.e. y is the vector of account incomes of T, then
y = n+ x [3.5]
where n and x are the vectors of row sums of N and X respectively. This vector equation
represents the second set of model equations, the first being equation 3.2, the price system of
the model. Equation 3.5 relates to the demand side of the system of equations. It simply
explains how total income is derived from endogenous and exogenous demands.
Now, let [x] denote the number of elements in a vector x. Equation 3.1 thus contains
[y] + [p] + [f] + 1 variables, the total number of variables in the model (assuming that there is
only a single element in the vector λ). The price system is made up of [p] equations (equation
3.2), while the demand system contains [y] – 1 equations (equation 3.5). Note that one of the
equations in the set y is linearly dependent on the others, which explains the reason for
subtracting the one. Combining these two sets of equations gives [p] + [y] – 1 equations.
Thus, there are [f] + 2 degrees of freedom.
In a system where all prices are flexible the system will only solve (or ‘close’) if a set of
[f] + 2 equations known as closure rules are defined. The closure rules form the third and final
set of equations in the model. In general (using the accounts in the SAM in Table 2) closure
rules apply to three types of accounts, namely the government account, the capital account
and the rest of the world account. Since the system is over-identified certain variables are
fixed in order to reduce the degree of freedom. Economic theory and evidence should guide
the modeller in defining the closure rules.
Each closure rule defines how equilibrium should be reached. Firstly, one has to decide
whether government savings are fixed or flexible, i.e. how is the government budget
16
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
At least one closure rule must not be linear homogenous, otherwise a solution will not be
found. This is achieved by fixing one of the prices, the numeraire, in the system. Due to
interdependency of prices all other prices can be expressed relative to the price of the
numeraire.
This paper aimed to introduce the basic concepts of social accounting. A SAM is an
ingenuous development that not only reconciles the production side of the economy with the
rest of the economy, but also presents flows of resources associated with economic
transactions in a concise way. SAMs are consistent and comprehensive because they adhere to
the fundamental law of economic accounting and they capture all economic transactions
between all agents. This ensures that a balance is maintained at a microeconomic level. They
also capture the net flow of resources between agents in the economy, thus adding to our
understanding of the structure of the economy. This ensures that macroeconomic balances are
maintained. These features have contributed to the popularity of SAMs and their ever-
increasing use as a statistical basis for multisector and macroeconomic modelling.
However, SAMs are not only used to show resource flows. Although a SAM in itself does
not constitute an economic model, it can be used to present economic theory as illustrated by
Pyatt (1988). The SAM approach to modelling is a useful way of showing how the model
statement (price and demand system, behavioural equations and macroeconomic
balances/closures) of a macroeconomic or general equilibrium model can be developed. It is
also a useful way of helping to understand the process of calibration. The SAM approach to
modelling clearly shows that the various properties of a SAM, particularly the fact that
account row and column totals are balanced, are very important when a SAM is used as the
database for a general equilibrium model.
There are some issues that have not been touched on in this paper that need further
attention. It is envisaged that some or all of these topics will be added to a revised edition of
this Background Paper. These include the following. Firstly, the price system underlying a
SAM is important. This was only briefly mentioned in section 3 about the SAM approach to
modelling. A revised edition will explain the distinction between basic prices (consumer
prices) and purchaser prices and how these are related or interdependent. Also refer to
Technical Paper 2003: 3.
17
© PROVIDE Project
PROVIDE Background Paper 2003: 4 September 2003
Secondly, some issues surrounding production and trade need to be raised. Specifically,
the incorporation of multi-product activities, how trade and transport margins are accounted
for, and how multiple trade partners can be included in a SAM needs to be discussed. Some of
these issues are discussed in Technical Paper 2003: 3.
Thirdly, the selection and the degree or level of disaggregation of accounts in a SAM was
not mentioned in this paper. Although a macro-SAM will usually follow the basic structure
presented in Table 2, the construction of a more detailed or disaggregated micro-SAM
requires that some thought be given to the commodity/activity breakdown, the number and
types of factors, the choice of representative household groups and whether multiple trade
partners should be included in the rest of the world accounts. Forthcoming Background
Papers on the formation of representative household groups and on South African
international trade discuss some of these issues.
5. Bibliography
King, B.J. (1985). “What is a SAM?” In Social Accounting Matrices: A Basis for Planning, edited by G. Pyatt
and J Round. Washington, D.C.: The World Bank.
Löfgren, H. (1999). “Exercises in General Equilibrium Modeling Using GAMS.” Microcomputers in Policy
Research 4, International Food Policy Research Institute.
Löfgren, H., Harris, R.L. and Robinson, S. (2001). “A Standard Computable General Equilibrium (CGE) Model
in GAMS.” International Food Policy Research Institute: Trade and Macroeconomics Division
Discussion Paper, No.75, May 2001.
McDonald, S. and Punt, C. (2001). A Social Accounting Matrix for the Western Cape. Mimeo.
McDonald, S. and Robinson, S. (1998). Developing a Social Accounting Matrix for South Africa. Paper
prepared for the 1998 Conference of the Development Economics Study Group, University of
Reading, July 1998.
Pyatt, G. (1988). “A SAM Approach to Modeling.” Journal of Policy Modeling 10(3): 327 – 352.
Pyatt, G. (1991). “Fundamentals of Social Accounting.” Economic Systems Research, Vol. 3 No. 3.
Reinert, K.A., and Roland-Holst, D.W. (1997). “Chapter 4: Social accounting matrices”, in Applied methods for
trade policy analysis, Cambridge University Press, Cambridge, UK
Robinson, S. (1989). “Chapter 18: Multisectoral models”, in Handbook of Development Economics, Volume II,
Ed. Hollis Chenery and T.N. Srinivasan, North Holland.
18
© PROVIDE Project
Background Papers in this Series