MEASURING THE MIDDLE CLASS OF PAKISTAN

Abstract

The aim of this study was to estimate the proportion of middle class in Pakistan and to explore the factors that determines the likelihood of a household of being in a particular class category. To achieve these objectives, I constructed a middle-class index using PSLM 2019-2020 data considering all possible indicators that might affect the middle-class including income, education, occupation, housing, lifestyle, and a new indicator named ICT. To explore the factors that affect household's likelihood of being in a particular class category I used ordered logistic regression model including 6 ordered categories from lower-lower class to upper-middle class. The results showed that in Pakistan the size of expanded middle class is about 37% with ICT index and 35% without ICT, the difference is due to the difference of upper-lower class. The size of vulnerable middle class is 15.3% and secure middle class is 20.4%. Most of the middle-class lives in Punjab province. The proposition that middle class is an urban phenomenon holds true with 68% of secure middle-class lives in urban areas and 32% of secure middle-class lives in rural areas. The results of the ordered logistic model revealed that all the factors included in the study have a significant role on households’ likelihood of being in a particular class. Among all, education level and skill level of household head are very important. As both have a positive significant relationship with ordered categories of class. At the end it is suggested to explore how macroeconomic factors affect the size of middle class and to study the inverse causality of how middle-class affects the economy and society. The present study recommends that public policies should be diverted to strengthen the middle class and enhance education quality and skill development