The average hides the country
Pakistan's national averages are useful accounting figures, but official HIES quintiles show how sharply household resources and living standards diverge across the country.
Five fifths, not one average.
PBS groups the population into five equally sized consumption quintiles after adjusting household expenditure for family size and composition. They are welfare groups, not simple income bands.
First fifth
Rs 40.0k
monthly household consumption
Rs 5,291
per person, per month
Rs 41,851 household income
Second fifth
Rs 53.3k
monthly household consumption
Rs 7,792
per person, per month
Rs 55,138 household income
Middle fifth
Rs 62.9k
monthly household consumption
Rs 10,235
per person, per month
Rs 65,741 household income
Fourth fifth
Rs 77.3k
monthly household consumption
Rs 13,920
per person, per month
Rs 80,387 household income
Highest fifth
Rs 134.7k
monthly household consumption
Rs 28,962
per person, per month
Rs 139,317 household income
The welfare gap
Household totals understate the difference.
The highest fifth consumes about 3.4 times as much per household as the first fifth. Per person, the gap is about 5.5 times.
The first fifth averages 7.56 people per household; the highest averages 4.65. A household total can look substantial while each person still has very little to live on.
The national consumption mean of Rs 79,150 sits above the average for each of the bottom four fifths. It is an aggregate statistic, not a representative household budget.
Official trend
The latest official comparison shows wider inequality and higher poverty.
The national Gini coefficient rose from 28.4 in 2018-19 to 32.7 in 2024-25. The official national poverty headcount rose from 21.9% to 28.9% in the same comparison.
These are distinct measures: the Gini describes the spread of the distribution; poverty counts people below a threshold. Together, they support a wider distributional problem. They do not identify a single cause.
Distribution shape
In our household-weighted reconstruction, most households sit left of the mean.
The long right tail pulls the average upward. This reconstruction makes the shape visible; it does not replace an official PBS distribution table.
The quintile figures and inequality trends come from published PBS and Finance Division tables. The distribution chart on the visual report is a PK Ventures reconstruction from the public HIES 2024-25 microdata, not an official PBS table. It groups household consumption into Rs 20,000 monthly bands, applies the official household weights, and reproduces a weighted mean of Rs 78,557, 0.75% below the official Rs 79,150 mean. Use it to read the shape of the distribution, not as an official continuous estimate or a median.
HIES quintiles are based on consumption welfare adjusted for household size and composition. The income figures shown are average incomes observed within those consumption quintiles.
Operating capacity
Products, jobs, and services cannot be designed around a fictional average customer.
The practical test is whether households across the distribution can earn reliably, absorb shocks, maintain useful assets, afford essential services, and build a path to ownership.
Do not stop at agreement
Challenge the argument, add what we missed, or show us where it should become work.
Sources
- [1]Pakistan Bureau of Statistics, HIES 2024-25 Economic Report — Official household income, household consumption, per-capita consumption, household-size, and consumption-quintile tables; see PDF pages 32, 34-35, and 44-46.
- [2]Pakistan Bureau of Statistics, HIES 2024-25 Microdata (Stata) — Public microdata used for the PK Ventures household-weighted reconstruction shown in the distribution chart; it is not an official PBS distribution table.
- [3]Finance Division, Pakistan Economic Survey 2025-26: Social Protection — Official poverty and inequality estimates, including the national Gini coefficient and urban-rural trends; see PDF pages 2-4.
Related insights
Continue the thread.
Each note should move into the next useful question, not end as a standalone argument.