Creating a 2000 IES-LFS Database in Stata

PROVIDE Project Technical Paper 2005:1

SARB 2000 does not make provision for inter-household transfers because, in theory, net
(national) domestic transfers in any economy should be zero. Hence, it was decided to form a
single transfer income category. In the case of SARB 2000 this is made up of net transfers
from abroad and from government. In the case of IES 2000 it is made up of inctrans plus
incgov minus hhtrans (household transfer expenditure). These two income sources are not
strictly comparable, although it does give a general idea of the extent of total transfer income.
Other income (incother) and income from the sale of home production (inchphc) in the IES
2000 was netted out by increasing the other income categories pro rata.

The relative contribution of each income source is shown in Table 5. The percentages
reported in columns two and four show the relative contribution of the three income sources
to total current income. Income tax and savings are expressed as percentages of current
income.

Table 5: Comparing IES 2000 and SARB 2000 income and expenditure patterns

• IES 2000 figures are weighted and multiplied by the number of households (11 million) to obtain
an estimate of the national totals. Data was also adjusted so that total income and expenditure
matches (see section 4).

• (*) SARB 2000 income from labour quoted directly from source; IES 2000 labour income includes
income from the sale of home produce.

• (**) SARB 2000: income from property plus transfers from incorporated business enterprises; IES
2000: incgos plus inccorp.

• (***) SARB 2000: net transfers from general government plus net transfers from the rest of the
world; IES 2000: inctrans plus incgov minus hhtrans.

• Note: Analytic weights assumed (variable weight) for IES 2000 data

From Table 5 it is clear that the income patterns differ quite substantially between the two
data sources. In both data sources labour is the most important source of income, but the
relative contribution of labour is much higher in IES 2000. Notably, the definition of income
from labour is fairly broad in IES 2000, as it includes non-monetary forms of remuneration
such as food, housing and clothes. As far as the other income sources are concerned
differences are most likely as a result of definitional differences. However, Simkins (2003)
also notes a large drop in income from net profits in the IES 2000 data compared to IES 1995,

21

© PROVIDE Project

<<   <   23   24   25   26   27   28   29   >   >>