4.1 A People-Centred Lens The preceding sections examined poverty through a geographic and structural lens: where deprivation is concentrated, what contextual and structural factors drive it, and how shocks amplify it. Section 4 turns to a people-centred perspective, asking who bears the greatest burden of multidimensional poverty across income, employment, education, migration status and household type. Figure 7 compares household groups across three dimensions simultaneously. The horizontal axis shows how many are poor (incidence), the vertical axis shows how deeply poor they are (intensity), and bubble size reflects the absolute number of MPI-poor households in each group. Groups in the upper-right quadrant face the worst outcomes: both widespread and deep poverty. Groups with large bubbles carry the greatest share of the national poverty burden, even when their MPI score is not the highest. Figure 7 Multidimensional poverty across household types Figure 7. Multidimensional poverty across household types Position shows MPI headcount and intensity · colour shows MPI score · bubble size shows MPI-poor population Larger bubble = larger MPI-poor population Dashed lines = national levels (H 32.6%, A 46.1%) MPI head Education Figure 8.Household Non-linear relationship between income and multidimensional poverty higher Migration status No formal Migrant 49% Predicted probability of being MPI-poor by monthly per-capita income (MMK, log scale) Female-headed Predicted probability Income axis shown on a log scale Primary Structural trap 46% MPI intensity (A) Predicted probability of being MPI-poor Initial incomeMale-headed gains yield limited poverty reduction Transformative middle High/higher 42% 20% 40% Diminishing returns Largest marginal reductions in poverty risk 55% 20% Remaining deprivation is increasingly structural Non-migrant Middle 40% 55% 20% 40% 55% 65% Employment type 49% Unemployed Income quintile Shock exposure Q1 Conflict Casual worker Economic Q2 46% Employee Health Natural Own-account Employer 42% 19% Q4 Q5 30,000 20% 100,000 40% 55% 5% Q3 300,000 20% 40% 55% Monthly per-capita income (MMK, log scale) 1,000,000 20% 40% 55% Model: penalized logistic regression with natural cubic splines (df = 3), survey-weighted MPI headcount (H) Source: PPS 2025 (UNDP) Source: PPS 2025 (UNDP) In practice, many of these characteristics overlap, as migrant households are often income-poor, reliant on casual labour and exposed to shocks, reinforcing the compounding nature of multidimensional poverty. The following sections examine these intersections, moving from structural traps to acute amplifiers to gendered dimensions of deprivation. L AYE R S O F DE P R I VAT I O N : M u l t i di m en si on a l Pov er t y i n M y anmar 29