Key findings
- Based on preliminary 2024 CBMS results, the national MPI at the individual level was 0.045: 12.8 percent of the Philippine population belonged to households classified as multidimensionally poor, with an average intensity of deprivation of 34.7 percent among them (PSA, reference 2026-270, released August 2026).
- At the household level the 2024 national MPI was 0.042: about 12.4 percent of Filipino households were multidimensionally poor, with an average intensity of 33.9 percent.
- The Philippine MPI, as defined by the PSA, uses 17 weighted indicators across 4 dimensions: education (2 indicators), health and nutrition (3), housing, water, and sanitation (9), and employment (3). Some secondary coverage reported 14 indicators; the PSA release and the OPHI directory both state 17.
- In the 2024 individual-level MPI, education contributed the largest share of overall deprivation (40.7 percent), followed by housing, water, and sanitation (31.3 percent), health and nutrition (18.5 percent), and employment (9.5 percent).
- The most widespread single deprivation in the whole population was housing materials (38.7 percent of individuals). Among the multidimensionally poor, educational attainment and housing materials were the most prevalent deprivations, each at a censored headcount of 10.7 percent.
- The MPI is a different measure from the official income-based poverty rate. Based on the 2025 Family Income and Expenditure Survey, income poverty incidence in 2025 was 6.4 percent of families (about 1.90 million families) and 9.7 percent of individuals (about 11.08 million Filipinos). The PSA presents the two measures as complementary; they differ in reference year, data source, and concept.
Important numbers
| Item | Value | Unit and period | Source |
|---|---|---|---|
| MPI, individual level | 0.045 | Adjusted headcount ratio, 2024 (preliminary CBMS) | Source 1, Figure 3 |
| Headcount ratio (H), individual level | 12.8 | Percent of population, 2024 | Source 1, Figure 3 |
| Intensity (A), individual level | 34.7 | Percent of weighted indicators, 2024 | Source 1, Figure 3 |
| MPI, household level | 0.042 | Adjusted headcount ratio, 2024 (preliminary CBMS) | Source 1, Figure 2 |
| Headcount ratio (H), household level | 12.4 | Percent of households, 2024 | Source 1, Figure 2 |
| Intensity (A), household level | 33.9 | Percent of weighted indicators, 2024 | Source 1, Figure 2 |
| Poverty cutoff | 0.25 | Weighted deprivation score; poor if the score exceeds 0.25 | Source 1, page 2 |
| Number of indicators / dimensions | 17 / 4 | Count, 2025-approved methodology | Source 1, Annex A; Source 5 |
| Income poverty incidence, families | 6.4 | Percent of families, 2025 (FIES) | Source 4 |
| Income poverty incidence, individuals | 9.7 | Percent of individuals, 2025 (FIES) | Source 4 |
| Poor families, count | 1.90 million | Families, 2025 | Source 4 |
| Poor individuals, count | 11.08 million | Individuals, 2025 | Source 4 |
How the MPI works
The PSA Board approved the National MPI methodology, which adopts the Alkire-Foster method, on 15 May 2025 through Board Resolution No. 08, Series of 2025. The index asks not how much a household earns but whether it is deprived in a set of basic conditions.
Four dimensions, 17 indicators, nested uniform weights
Each of the four dimensions carries a weight of 0.25. That weight is divided equally among the indicators inside the dimension. Education has 2 indicators (0.1250 each), health and nutrition has 3 (0.0830 each, as printed by the PSA), housing, water, and sanitation has 9 (0.0278 each), and employment has 3 (0.0830 each).
The cutoff
The PSA classifies a household as multidimensionally poor if the sum of its weighted deprivation scores exceeds the cutoff of 0.25, which the PSA describes as being deprived across more than one of the four equally weighted dimensions. Every member of a household carries the same deprivation score, because the PSA assumes people in the same household experience the same deprivations.
The formula
MPI = H x A, where H is the headcount ratio (the share of poor households or individuals) and A is the intensity (the average deprivation score among the poor). The closer the MPI is to 1, the higher the level of multidimensional poverty. For 2024 at the individual level, multiplying the published rounded figures, 0.128 x 0.347, gives about 0.044. The PSA publishes 0.045. The small difference is consistent with the PSA computing the index from unrounded H and A values before rounding each figure for publication; this is Evidence Atlas's reading, not a PSA statement.
Two ways to count a deprivation
The uncensored headcount ratio for an indicator is the share of the whole population deprived in that indicator, whether or not they are multidimensionally poor. The censored headcount ratio counts only people who are deprived in that indicator and live in a multidimensionally poor household. The gap between the two tells you how much of a deprivation sits outside the officially poor group.
Plain-language definitions of all 17 indicators (paraphrased from PSA Annex A)
These are paraphrases. The authoritative wording is in Annex A of the PSA press release (Source 1). The PSA notes that each measure is subject to improvement when new data become available.
| Dimension | Indicator | Household is deprived if | Weight |
|---|---|---|---|
| Education | School attendance | Any child aged 5 to 17 is not currently attending school. | 0.1250 |
| Education | Educational attainment | No household member aged 19 or older has completed secondary education. | 0.1250 |
| Health and nutrition | Food insecurity | Any member aged 18 or older experienced moderate or severe food insecurity on the Food Insecurity Experience Scale. | 0.0830 |
| Health and nutrition | Illness | Any member got ill, sick, or injured in the past six months, needed medical attention, and did not get medical treatment. | 0.0830 |
| Health and nutrition | Proximity to health facilities | The nearest health centre is more than 2 km away and the nearest hospital more than 10 km away. | 0.0830 |
| Housing, water, sanitation | Sanitation / toilet facility | The household does not use a flush or closed-pit toilet, or shares its toilet with other households. | 0.0278 |
| Housing, water, sanitation | Water source | Water is not piped into the dwelling, yard, or plot, and is not from a protected well. | 0.0278 |
| Housing, water, sanitation | Ownership of assets | The household lacks at least one communication asset and one durable asset, or lacks any mobility asset (car, jeepney, van, motorcycle, motorised boat, e-jeep, e-trike). | 0.0278 |
| Housing, water, sanitation | Housing tenure | The household lives rent-free on a house and lot, or in its own house on a rent-free lot, without the owner's consent. | 0.0278 |
| Housing, water, sanitation | Housing materials | The roof or wall is made of salvaged or light materials, or the floor is made of natural or rudimentary materials. | 0.0278 |
| Housing, water, sanitation | Electricity | There is no electricity in the housing unit. | 0.0278 |
| Housing, water, sanitation | Internet access | No household member has internet access, postpaid or prepaid. | 0.0278 |
| Housing, water, sanitation | Overcrowding | Floor area per person is less than 4 square metres. | 0.0278 |
| Housing, water, sanitation | Access to public transportation | No public transport vehicle is available within 500 metres of the dwelling. | 0.0278 |
| Employment | Employment deprivation (unemployment) | More than 50 percent of household members in the labour force aged 18 to 65 are unemployed. | 0.0830 |
| Employment | Labour underutilisation (underemployment) | More than 50 percent of employed members aged 18 to 65 are looking for additional hours of work. | 0.0830 |
| Employment | Working children not in school | Any member aged 5 to 17 is working and not currently attending school. | 0.0830 |
Cutoff tester: would this household count as multidimensionally poor?
Tick the deprivations a hypothetical household has. The tester adds the PSA weights and compares the total with the 0.25 cutoff. It uses the exact nested uniform fractions (1/8, 1/12, 1/36) that the printed Annex A weights round to, so a household deprived in every indicator of exactly one dimension scores exactly 0.25 and is not classified as poor, because the PSA rule requires the score to exceed 0.25. This is an illustration of the published rule, not an official assessment tool.
Worked example in text: a household with no adult who finished secondary school (0.125), light housing materials (0.0278), and no internet access (0.0278) scores about 0.181, which does not exceed 0.25, so it is not multidimensionally poor. Add food insecurity (0.0833) and the score becomes about 0.264, which exceeds the cutoff.
Which dimensions drive the 2024 MPI
The PSA reports how much each dimension contributes to the individual-level MPI. Education contributed the largest share of overall deprivation (40.7 percent), followed by housing, water, and sanitation (31.3 percent), health and nutrition (18.5 percent), and employment (9.5 percent). These are based on preliminary 2024 CBMS results.
Why education dominates. Education has only two indicators, so each carries a heavy weight (0.1250). Educational attainment deprivation, meaning no adult aged 19 or older in the household finished secondary school, affected 20.2 percent of the population overall and 10.7 percent when counting only the multidimensionally poor. By Evidence Atlas's own arithmetic on the PSA weights, one education indicator (0.1250) carries the same weight as four and a half housing, water, and sanitation indicators (0.0278 each). The PSA does not make this comparison itself.
Indicator explorer: who is deprived in what
The table lists all 17 indicators with two headcounts for 2024, in percent of individuals. The uncensored headcount is the share of the whole population deprived in that indicator. The censored headcount is the share deprived in that indicator and living in a multidimensionally poor household. Use the controls to sort or to focus on one dimension; the full table is always present in the page text.
| Indicator | Dimension | Uncensored % | Censored % | Bars (light: uncensored; dark: censored) |
|---|---|---|---|---|
| Housing materials | Housing | 38.7 | 10.7 | |
| Internet access | Housing | 26.4 | 6.8 | |
| Sanitation / toilet facility | Housing | 24.1 | 6.1 | |
| Ownership of assets | Housing | 21.8 | 7.8 | |
| Educational attainment | Education | 20.2 | 10.7 | |
| Food insecurity | Health | 17.6 | 7.2 | |
| Access to public transportation | Housing | 17.4 | 4.7 | |
| Water source | Housing | 15.9 | 5.6 | |
| Overcrowding | Housing | 14.8 | 4.1 | |
| Labour underutilisation | Employment | 9.9 | 3.2 | |
| Electricity | Housing | 5.7 | 3.7 | |
| School attendance | Education | 5.7 | 3.8 | |
| Employment deprivation | Employment | 3.6 | 1.1 | |
| Illness without treatment | Health | 3.2 | 1.2 | |
| Housing tenure | Housing | 2.6 | 0.7 | |
| Proximity to health facilities | Health | 2.2 | 1.5 | |
| Working children not in school | Employment | 0.8 | 0.7 |
Bars are scaled so that 40 percent fills the full width. All values are also in the table cells.
What the gap between uncensored and censored headcounts tells you
Housing materials deprivation affects 38.7 percent of the population but only 10.7 percent when restricted to the multidimensionally poor. So roughly 28 percent of Filipinos live in housing with light or salvaged materials without being counted as multidimensionally poor, because they are not deprived enough elsewhere to cross the 0.25 cutoff. The same pattern holds for internet access (26.4 versus 6.8) and sanitation (24.1 versus 6.1). Educational attainment has a smaller gap (20.2 versus 10.7): about half of the people in households where no adult finished secondary school are in multidimensionally poor households, which is consistent with education's large weight in the index.
These gap figures are simple subtractions of the two published headcounts and are computed by Evidence Atlas, not published by the PSA.
The MPI and the income poverty rate, side by side
The PSA describes the MPI as a complement to its income-based poverty statistics, which are released every two years. The two measures answer different questions, use different data, and refer to different years. They should not be netted against each other or read as a single trend.
| Feature | Multidimensional Poverty Index | Income-based poverty incidence |
|---|---|---|
| Reference year | 2024 | 2025 |
| Data source | 2024 Community-Based Monitoring System (CBMS), preliminary results; participation voluntary | 2025 Family Income and Expenditure Survey (FIES), conducted July 2025 and January 2026 |
| What counts as poor | Weighted deprivation score above 0.25 across 17 indicators in 4 dimensions | Income below the official poverty threshold for minimum basic food and non-food needs |
| Individuals | 12.8 percent of the population in multidimensionally poor households | 9.7 percent of Filipinos, about 11.08 million |
| Households or families | 12.4 percent of households | 6.4 percent of families, about 1.90 million (first single-digit reading) |
| Release | PSA reference 2026-270, August 2026 | PSA reference 2026-272, 21 August 2026 |
| Source | Source 1 | Source 4 |
What moved the income poverty rate, and the critique of the threshold
The PSA attributes the decline in income poverty to income growing faster than the threshold: between 2023 and 2025 the annual per capita poverty threshold rose 5.5 percent while mean annual per capita income rose 22.0 percent; income for families in the first and second deciles rose 23.8 percent and 22.7 percent, respectively (Source 4).
As reported by the Inquirer, the 2025 annual per capita poverty threshold is about PhP 35,121, or roughly PhP 96 per day (Source 6, Source 8). Ibon Foundation, an advocacy group, contends the official threshold understates poverty and estimates a family living wage of PhP 1,277 per day for a family of five, against an average minimum wage it cites of PhP 512 (Source 8). This is an attributed critique of the income measure, not an established finding, and PSA and NEDA do not concur with it. It does not bear on the MPI figures, which do not use an income threshold at all.
Data coverage behind the 2024 MPI
The 2024 MPI uses preliminary CBMS results; CBMS participation is voluntary under Section 10 of RA 11315. The cities of Las Piñas, Pasay, Iloilo, and General Santos were excluded from the national estimate pending data validation; the PSA says city-level MPI results for these four will be provided once available. Enumeration-area completion also varied by area. The percentages below measure completed enumeration areas, not population coverage.
| City | Enumeration areas completed (%) |
|---|---|
| City of Baguio | 50.1 |
| City of Manila | 59.5 |
| City of Muntinlupa | 62.9 |
| City of Makati | 70.8 |
| City of Taguig | 73.7 |
| City of Caloocan | 79.7 |
| Las Piñas, Pasay, Iloilo, General Santos | Excluded from national estimate |
| Most provinces (Annex C) | 98 to 100 |
No regional, provincial, or barangay MPI results were published in this release. Do not read the coverage table as an MPI ranking.
Sources and evidence
- Philippine Statistics Authority, "12.8% of Filipinos are Multidimensionally Poor in 2024", Press Release Reference No. 2026-270, with Annex A (indicator definitions and weights), Annex B and C (CBMS coverage). Web page dated 27 August 2026; retrieved 2 September 2026. PDF · web page. Licence: CC BY 4.0. Supports CL-01, CL-03 to CL-11.
- Philippine Statistics Authority, Poverty Statistics main page (HTML version of release 2026-270 with links to the PDF and XLSX tables). Page dated 27 August 2026; retrieved 2 September 2026. link.
- Philippine Statistics Authority, "2024 National MPI_tables.xlsx". Linked from Source 2; not opened for this version because the PSA file server returned a bot-protection challenge to automated retrieval on 2 September 2026. link.
- Philippine Statistics Authority, "Proportion of Poor Families Reduced to 6.4 Percent in 2025", Press Release Reference No. 2026-272, published 21 August 2026; retrieved 2 September 2026. link. Licence: CC BY 4.0. Supports CL-12, CL-13.
- Oxford Poverty and Human Development Initiative (OPHI), "Philippines MPI", National MPI Directory, structure dated 2025; retrieved 2 September 2026. link. Corroborates the 17-indicator structure and the May 2025 Board approval; documents the earlier 2018 methodology on 2016/2017 data. Supports CL-01, CL-02, CL-15.
- Philippine Daily Inquirer, "New gauge of poverty: 1 in 8 PH homes lack not just income", 28 August 2026; retrieved 2 September 2026. link. Secondary reporting; source for the reported peso threshold (CL-14).
- The Manila Times, "New poverty index finds challenges beyond income", 28 August 2026; retrieved 2 September 2026. link. Secondary reporting; states 14 indicators, which differs from the primary source.
- INQUIRER.net, "Ibon: Gov't poverty standard 'outdated,' 'unrealistic'", 26 August 2026; retrieved 2 September 2026. link. Secondary reporting of an advocacy critique (CL-16).
- PSA Board Resolution No. 08, Series of 2025, approving the National MPI methodology, 15 May 2025. link. The file is a scanned PDF; its existence and date are corroborated by Sources 1 and 5, but its text was not read for this version.
Claim IDs (CL-…) refer to the Evidence Atlas verified claim register CR-20260902-01 for this Drop.
Methodology
All MPI figures are taken directly from the PSA press release (Source 1) and its annexes. The 34 indicator headcounts in the explorer were extracted from the data series embedded in the PSA's Figure 4 bar chart within the PDF and cross-checked against the twelve values the PSA also states in prose; all matched. They were independently re-extracted a second time during claim verification. The XLSX tables (Source 3) were not opened for this version, so the figures have not been checked against a separate labelled data table.
Indicator definitions are paraphrased from Annex A. Weights are shown as printed by the PSA (0.1250, 0.0830, 0.0278). The cutoff tester uses the exact fractions 1/8, 1/12, and 1/36 that these printed values round to, so that each dimension sums to exactly 0.25.
The only computations by Evidence Atlas are: (a) the check that H x A is consistent with the published MPI at both levels (0.124 x 0.339 is about 0.042, matching the published household figure; 0.128 x 0.347 is about 0.044 against a published individual figure of 0.045, a difference consistent with the PSA computing from unrounded H and A); (b) the uncensored minus censored gaps discussed in the explorer; and (c) the worked example in the cutoff tester. Income poverty figures come from Source 4. The comparison table sets the two measures side by side and labels year, data source, and concept; it does not rank them.
Source hierarchy: primary PSA releases first; the OPHI directory as independent corroboration; news reporting only for attributed statements and for documenting where coverage diverges from the primary source.
Limitations
- The 2024 MPI results are based on preliminary 2024 CBMS results, per the PSA's own note, and may be revised.
- The indicator headcounts were read from the Figure 4 chart's embedded data series, not from a separate data table; the PSA XLSX (Source 3) has not yet been opened for cross-checking.
- The PSA PDF's header reads "Date of Release: 18 August 2026", while the PSA web page lists the release as 27 August 2026 and news coverage appeared on 28 August. This Drop cites the reference number 2026-270 and treats 27 August as the public release date; the discrepancy is unresolved at the source.
- Some secondary coverage reported 14 indicators; the primary source and OPHI both state 17.
- Indicator definitions are paraphrases of Annex A, not verbatim PSA text; consult Annex A for the authoritative wording.
- Coverage percentages in the data coverage section measure completed enumeration areas, not population coverage. Four highly urbanised cities are excluded from the national estimate, and CBMS participation is voluntary.
- No regional, provincial, or city MPI figures were published in this release; none appear here.
- The MPI (2024, CBMS, deprivation-based) and income poverty (2025, FIES, income-threshold-based) differ in reference year, data source, and concept. They are complementary and cannot be netted against each other or read as one trend.
- The PSA first published a national MPI methodology in 2018 based on 2016/2017 data. That earlier index used a different methodology and is not comparable with the 2024 results, so no trend is shown.
- The peso poverty threshold (about PhP 35,121 per capita per year) is cited as reported by the Inquirer; it was not found on the PSA release page used for this Drop.
- The Ibon Foundation living-wage figures are presented as an attributed advocacy critique, not as an established finding.
- The text of PSA Board Resolution No. 08 s. 2025 was not read; its date and subject are corroborated by two independent sources.
Freshness
Freshness class F3 (release-driven). Last verified 2 September 2026. Monitored sources: the PSA Poverty Statistics page (Source 2) for a final, non-preliminary 2024 MPI, for city-level results for the four excluded cities, and for any regional MPI tables; the Manila Times article (Source 7) for a correction of the indicator count. Next expected release: not announced by the PSA for the MPI; income-based poverty statistics follow a two-year cycle, so the next full-year release is expected in 2028 for reference year 2027. "Updated" for this Drop means a change to a published number, definition, or conclusion, not a routine source check.
Version
v1.2, 2 September 2026. Changelog: v1 initial build (2 September 2026). v1.1 corrected the H x A rounding explanation in the "How the MPI works" section, labelled the education-versus-housing weight comparison as the publisher's own arithmetic, marked channel and playlist as proposed, and added the Source 2 page date. v1.2 corrected the same rounding sentence in the Methodology section, which v1.1 had missed. No published figure changed in either revision. Drop ID D-20260902-7a2c. Evidence package EP-20260902-01; verified claim register CR-20260902-01.