Over 38pc urban Pakistanis live in unaffordable housing

ISLAMABAD: The World Bank (WB) has estimated that more than one in three urban residents in Pakistan face unaffordable housing conditions, estimating that 38 per cent of the urban population lives in unaffordable housing.

A provincial breakdown shows large differences across the two methods for urban Khyber Pakhtunkhwa (KP), where the rate of housing unaffordability almost doubles. In KP, housing poverty is estimated at 42.1pc whereas the ratio of unaffordability is 23.6pc; in Punjab, housing poverty is 25.8pc while the ratio of unaffordability is 35.8pc; in Sindh, housing poverty is 34pc with an unaffordability ratio of 45.3pc; and in Balochistan, housing poverty is 59.8pc with an unaffordability ratio of 37.1pc.

The results of housing poverty emerged as the WB proposed a modified approach to measure housing affordability by estimating a ‘housing poverty’ rate, and to implement the ‘Residual Expenditure Methodology (REM) for urban Pakistan. For which, it used data from the nationally representative household survey conducted by Pakistan Bureau of Statistics.

The WB report, “Behind on Rent or Left Behind: Measuring Housing Poverty in Urban Pakistan”, made available on Friday, says the REM approach has been proposed, drawing on existing poverty measurement methodology, to measure housing poverty in urban Pakistan. Based on this, households that are unable to afford a minimum threshold of non-housing expenditures are classified as “housing-poor”.

Housing poverty estimated at 59.8pc in Balochistan

The latest data the WB used was from the 2018-29 household integrated economic survey (HIES) which provides a baseline of spending levels and patterns. The bank proposed a modified residual income method (RIM) approach: the Residual Expenditure Methodology (REM) that allows it to measure housing affordability by estimating a “housing poverty” line.

The REM approach relies on two building blocks to estimate housing poverty: (a) households’ non-housing consumption aggregate, as a proxy for residual income, and (b) a minimum socially acceptable standard for these non-housing (residual) expenditures, or a “non-housing poverty line”. In place of budget standards, the REM draws on poverty measurement methodologies to determine a minimum acceptable standard of well-being and its associated spending value. Currently, most developing countries measure monetary poverty using consumption expenditure and absolute poverty lines.

This is in line with global best practice on measuring welfare in developing countries. Moreover, household survey data shows that income and consumption are highly correlated, and this further bolsters the case for using expenditure data.

The WB report says the non-housing poverty line for urban Pakistan is estimated at Rs3,716 per adult equivalent per month in 2018-19. Alternatively, this means that a family of five with three children requires a minimum of Rs16,350 per month leftover after paying for housing to be considered to be living in affordable housing.

Noting that the official measures of monetary poverty are not enough on their own to identify those living in precarious housing conditions, the report says the urban housing poverty rate is three times as high as the official urban poverty rate, the report says.

According to the WB report, the proposed housing poverty measure provides a significant addition to the repertoire of tools available to measure housing affordability, particularly in developing countries. Existing work on housing affordability in Pakistan has been limited in scope, leveraging alternate and often noisy datasets such as officially reported monthly rent and property evaluations from the excise department.

One of the key advantages of using this approach is that the analysis can be targeted to specific sub-populations based on the policy need. These can include specific regions, income levels, and types of employment. While we pilot the housing poverty method for urban Pakistan, this can be simulated at the provincial level, and at even further levels of geographical disaggregation depending on data availability and representativeness, the report says.