SiteMap •  Links • Search

Explanatory Notes on the 2003 Poverty Estimates

B. Inappropriateness of using CPI as income deflator in analyzing poverty

• In the FIES results released by the NSO, income in real terms is computed by deflating the nominal income by the consumer price index (CPI). However, for examining the trends in poverty incidence, CPI would not be an appropriate deflator.

• The main reason why the use of CPI is inappropriate in analyzing poverty trends is the differences in market baskets/commodities covered by the CPI and the poverty threshold. Further, the manner of deflating (that is, deflating just the average income of the decile group instead of deflating individual family income by the appropriate CPI and then obtaining the distribution of the real income at the family level) adds to the seeming inconsistency between the FIES and the poverty estimates.

B.1 Differences in market baskets/commodities covered by the CPI and the poverty threshold

• As shown in Table 2, CPI and poverty threshold (PT) increased at different rates from 2000 to 2003. In particular, CPI rose faster than PT both at the national level and at the regional level, except in Regions VIII and IX. This is due to the differences in the market baskets/commodities covered by the CPI and the PT.

Table 2
CPI and Poverty Threshold by Region, 2000 and 2003

• Table 3 shows the differences in the market baskets/commodities covered by the CPI and the PT. The CPI has a wider coverage of items that includes non-basic commodities as opposed to the PT, which covers only the expenditures on basic necessities. Items considered non-basic, such as alcoholic beverages, tobacco, recreation, durable furniture and equipment, taxes, and miscellaneous expenditures, are excluded from the PT. Moreover, for those expenditure groups that are covered by both CPI and PT (e.g., food, transportation and communication, household operations, personal care and effects, etc.), PT has a narrower coverage of items since it is limited only to those commonly purchased by poor families. For this reason, it would not be appropriate to use the CPI as deflator of per capita income in analyzing changes in poverty incidence.

Table 3
Comparison of the Market Baskets/Commodity Coverage of the CPI and Poverty Threshold

B.2 Manner of deflating

• In the released FIES results, the regional real average income for each income group (decile) is obtained by deflating the nominal average income for that decile by the regional CPI. However, a more precise way of obtaining the real income would be to deflate the individual nominal income of each family by the provincial CPI and then obtain the distribution of real income at the family level.

• As shown on Table 4, the use of regional CPI results in an inconsistent pattern between the trends from 2000 to 2003 in the country’s poverty incidence and the real per capita income of the third lowest income group, that is, poverty incidence dropped by 2.7 percentage points despite the decline, although minimal, in the average real per capita income of the income group whose per capita incomes are closest to the poverty threshold. In fact, a reduction was observed in the average real per capita income of all income groups. However, using provincial CPI, it becomes evident that the improvement in poverty incidence was mainly due to the positive growth in the average real per capita income of the third lowest income group, notwithstanding the 4.4 percent drop in the overall real per capita income.

Table 4
Real Per Capita Income (national/regional CPI versus provincial CPI as deflator) by Decile, Philippines, 2000 and 2003

• The complete tables by region using national/regional CPI and provincial CPI are found in Appendix Table B.1 and Table B.2 , respectively. The use of regional CPI results in inconsistent patterns at the national level and in NCR, Regions IV-A and V. On the other hand, the use of provincial CPI results in inconsistent patterns only in NCR and Region IV-A.

 

 

For these notes, inquiries may be made with Ms. Redencion M. Ignacio or
Ms. Glenita V. Amoranto at telefax number (632) 896-7981 or at their
e-mail addresses rm.ignacio@nscb.gov.ph / gv.amoranto@nscb.gov.ph.

 

 

	E-mail the webmaster	Terms of Use					Home • Top of Page
		1997-2008, National Statistical Coordination Board Makati City, Philippines

Poverty Statistics

Main Page

Explanatory Notes on the 2003 Poverty Estimates

Trends in poverty incidence should be analyzed vis-à-vis changes in the real income of lower income groups instead of changes in the overall real average income

Inappropriateness of using CPI as income deflator in analyzing poverty

Per capita income versus family income as the more relevant indicator for analyzing poverty

Analysis of trends in poverty incidence versus average per capita income (at constant prices using implicit price index as deflator) by region

Growth rate of average per capita income (at current prices) versus the growth rate of the poverty threshold provides an alternative way to analyze trends in poverty incidence

GRDP not a welfare measure and should not be compared directly with poverty incidence without examining income patterns among households

Download PDF version [156kb]