NSCB - Statistically Speaking... Poverty Statistics 2000-2003: Has Economic Growth Been Pro-Poor?

POVERTY STATISTICS 2000-2003:
HAS ECONOMIC GROWTH BEEN PRO-POOR?
by Dr. Romulo A. Virola ¹
Secretary General, NSCB

For February, I was going to write about statistics on heartaches/heart diseases, but so many questions have been raised about the poverty statistics recently released by the National Statistical Coordination Board (NSCB). So lovers, the loveful and the loveless among my friends will have to wait as I try to clarify a few things.

The major legitimate questions dwell on two things: (1) the seeming inconsistency between the trends in the poverty statistics of the NSCB and the results of the Family Income and Expenditures Survey (FIES) of the National Statistics Office (NSO) and (2) the inconsistency between the national income accounts of the NSCB and the income data from the FIES of the NSO.

On the first question, the perceived inconsistency arises from analyzing and comparing the change in real average family income from the FIES and the change in poverty incidence. The perception is that a decrease in real average family income should go with an increase in poverty incidence! Not always!

Three points to consider. First, the most important and the most basic - the average family income of the FIES is computed basically as an arithmetic mean, which means that the income of every family affects this average – the poor as well as the rich. Thus, the average family income from the FIES can decline when the family income of one John Buendia declines from P500 million to P300 million no matter if the income of many poor families increased above the poverty line. How does this happen? To simplify the exposition, let us assume we have only four families with one member each, two of whom were poor in 2000 as illustrated in TABLE 1. Assuming a uniform 20% increase in prices for all families, the hypothetical Republic of Unrequited Lovers experienced a 50% decline in real per capita income from 2000 to 2003 but poverty incidence went down from 50% to 25%! This is in fact, the essence of what the FIES data show. Get it? No? Well, if you are still groggy over your pre-valentine celebration, the NSCB Technical Staff which serves as the Secretariat of the Technical Committee on Poverty Statistics (TCPS) will be more than happy to answer your questions. The TCPS is also preparing statistical tables for those who want to take a deeper look at the FIES and poverty statistics.

What this means is that, when analyzing poverty incidence, it is not sufficient to look at average income; one should look at the income distribution, in particular, the lower income deciles. Put another way, poverty incidence is not about measures of central tendency; it is about the tails, specifically, the left tail of the income distribution. When the ultra rich have become less rich, it will have no bearing on the poverty incidence, but it will reduce average income! And even if the income of some poor families increased only a little bit, but enough to go above the poverty line, all other things remaining the same, poverty incidence will decline. And our perception, our anecdotal evidences have nothing to do with it! The problem starts and persists when many of us use our intuition and perception to assess our progress and treat them as having scientific basis the way we are enthralled by Krystala and Mulawin.Should this malpractice continue, our future as a nation will indeed, be bleak!

The second point to consider is the concept of real income and poverty estimation. In the official methodology followed by the NSCB for poverty assessment, income is compared with the poverty line. When one’s income is below the poverty line, he/she is considered poor. Income is derived from the FIES, while the poverty line is computed as the cost of a basket of commodities, both food and nonfood, but limited to commodities and services considered basic. Hence, the poverty line does not include the cost of expenditure items, considered nonbasic for poverty analysis, such as alcoholic beverages, tobacco and recreation, which are included in the much bigger CPI basket of commodities. In short, when relating real income to poverty incidence, one must look at the purchasing power of the income vis-à-vis the commodities included in the poverty line basket. Equivalently, it is wrong or at least not sufficient to compare the trend in poverty incidence with the trend in real income computed as nominal income deflated by the CPI for all items. Computations by the NSCB staff show that between 2000 and 2003, the CPI increased faster than the poverty threshold at the national and at the regional levels, except in Regions 8 and 9. An appropriate deflator must therefore be used and voila! The apparent inconsistency between the FIES statistics from the NSO and the poverty statistics of the NSCB disappears. Again, the NSCB Technical Staff has prepared tables on this, which will be uploaded on the NSCB website soon. TABLE 2 is a condensed version.

The third point, although not as major as the first two, is that when determining if a family is poor, the NSCB looks at the per capita income of the family rather than the total income of the family. Two families with the same level of family income may have different poverty status - one nonpoor, the other poor, if one family has a small size while the other is very large. Thus, in the analysis of trends of poverty incidence versus family income, we should look at the per capita income distribution rather than the total family income distribution.

The other issue that has been raised is on the family income captured by the NSO in the FIES versus the income reflected in the Philippine System of National Accounts (PSNA) compiled by the NSCB.

Conceptually, the two are different. First, when we talk of national income accounts in the PSNA, the national income includes income earned by the four institutional sectors²: general government, government corporations, private corporations and households & unincorporated enterprises. Thus, the total family income from the FIES is not equal to total national income from the PSNA. Even the comparison of total family income from the FIES with the total income of the Household & Unincorporated Enterprises of the PSNA, is not completely valid. The FIES covers only individual households. It excludes the so-called institutional households like those in prisons and long-term stay in hospitals. In addition, the PSNA Household accounts cover non-profit institutions serving households like the many charitable institutions in the country such as Hospicio de San Jose, Tahanang Walang Hagdan and other nongovernmental organizations. Moreover, the FIES income, unlike the PSNA income for Households, excludes depreciation and indirect taxes and subsidies. Therefore, not only should the levels of FIES income and national income from the PSNA be different, their trends do not have to be the same! In fact, between 2001 and 2003, the PSNA shows a 10.1% annual increase in income of the Household & Unincorporated Enterprises sector while the FIES shows an annualized increase in total family income of only 4.3 % from 2000 to 2003.

A related issue is the “inconsistency” when comparing trends in regional poverty incidence and in the Gross Regional Domestic Product (GRDP). Again, the “analysis” suffers from a lack of appreciation for the significance of the income distribution on poverty. Nonetheless, NSCB computations show general “consistency” in the sense that in 11 of 17 regions, increases in per capita GRDP were accompanied by a decline in poverty incidence while in 4 regions, decreases in per capita GRDP were accompanied by an increase in poverty incidence.

Having said all that, let me assure the public, that we, in the NSCB remain committed to the Fundamental Principles of Official Statistics adopted by the United Nations. We continue to be independent and professional in our work. Anyone and everyone is invited to observe how we work in the NSCB. In addition, the members of the Technical Committee on Poverty Statistics chaired by Dr. Celia Reyes and composed of representatives from various sectors including representatives from the academic community and the Philippine Statistical Association have been working very hard over many meetings to deliberate on the validity of the poverty estimates. For anyone to imply that these professionals will allow themselves to be used to pursue anybody’s agenda smacks of unadulterated intellectual arrogance!

And now back to the question: Has economic growth been pro-poor? The PSNA compiled by the NSCB shows that from 2001³ to 2003, real GDP and GNP grew at annualized rates of 4.5 % and 4.9%, respectively while per capita real GDP and per capita real GNP grew at annualized rates of 2.1% and 2.5%, respectively. On the other hand, as seen in TABLE 2, the FIES of the NSO shows that the income shares of the lower income brackets have grown but the income share of the richest income bracket has decreased between 2000 and 2003. The share in real terms of the poorest 30% has in fact grown from 7.95% in 2000 to 8.36% in 2003. On this basis, can we not say that, our perception and intuition aside, indeed, growth between 2000 and 2003 has been pro-poor? I say yes, we can, unless you lied to the NSO interviewer about your income!

Happy Valentine’s to all!

TABLE 1
INCOME AND POVERTY TRENDS IN THE REPUBLIC OF UNREQUITED LOVERS,
2000-2003

	Per Capita Annual Income In Pesos		% Change in Nominal Income, 2000-2003	% Change in Prices, 2000-2003	% Change in Real Income, 2000-2003	Poverty Status ¹ / Incidence
	2000	2003	% Change in Nominal Income, 2000-2003	% Change in Prices, 2000-2003	% Change in Real Income, 2000-2003	2000	2003
John Buendia	500,000,000	300,000,000	-40%	20%	-50%	Nonpoor	Nonpoor
Juan de la Cruz	8, 000	10,000	+25%	20%	+4.2%	Poor	Poor
Pedro Santos	30, 000	45,000	+50%	20%	+25%	Nonpoor	Nonpoor
Juana Reyes	10,000	13,000	+30%	20%	+8.3%	Poor	Nonpoor
Total	500,048,000	300,068,000	-40%	20%	-50%	50%	25%

¹ Poverty threshold was P11,451 in 2000 and P12,267 in 2003.

TABLE 2
PERCENTAGE POINT CHANGE IN INCOME SHARE, 2000-2003
PHILIPPINES

INCOME DECILE	2000 SHARE		2003 SHARE		PERCENTAGE POINT CHANGE², 2000-2003
INCOME DECILE	NOMINAL	REAL¹	NOMINAL	REAL¹	NOMINAL	REAL¹
FIRST	1.69	1.69	1.78	1.79	+0.10	+0.10
SECOND	2.73	2.73	2.85	2.86	+0.12	+0.13
THIRD	3.53	3.53	3.70	3.71	+0.17	+0.18
FOURTH	4.42	4.42	4.63	4.64	+0.21	+0.22
FIFTH	5.53	5.53	5.74	5.74	+0.21	+0.21
SIXTH	6.93	6.93	7.13	7.13	+0.20	+0.21
SEVENTH	8.84	8.84	9.04	9.05	+0.20	+0.21
EIGHTH	11.66	11.66	11.82	11.83	+0.16	+0.16
NINTH	16.33	16.33	16.55	16.55	+0.21	+0.21
TENTH	38.33	38.33	36.76	36.69	-1.57	-1.64

¹ Using 2000 based-CPI as deflator
² Rounding off errors

Reactions and views are welcome thru email to the author at ra.virola@nscb.gov.ph.

_______________
¹ Secretary General of the National Statistical Coordination Board (NSCB) and Chairman of the Statistical Research and Training Center (SRTC). He holds a Ph. D. in Statistics from the University of Michigan in Ann Arbor, USA and has taught mathematics and statistics at the University of the Philippines. He is also a past president of the Philippine Statistical Association.

² In the soon to be implemented 1993 System of National Accounts, the five institutional sectors are Government, Financial Corporations, Non-Financial Corporations, Households and Non-Profit Institutions Serving Households.

³2001 is used because in the current PSNA series, 2000 figures are not linked with the rest of the series

Posted 14 February 2005.


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