Searching for the Binding Constraint to Digital Financial Inclusion in Pakistan: A Decision Tree Approach

Over the last decade, Pakistan has seen improvements in the coverage of its networks of bank branches, ATMs, and mobile money agents. However, the country is lagging behind comparator countries when it comes to the financial inclusion of its population; according to the latest estimates, barely 20 percent are currently included. By using the Claessens and Rojas-Suarez (2020) decision tree methodology, this paper assesses the potential demandand supply-side constraints limiting the usage of digital payment services to identify which constraints are binding. Our main finding is that Pakistan’s institutional weaknesses constitute the most important binding constraint. These weaknesses have incentivized the creation of a sizable informal economy that has resulted in a preference for cash over the use of formal financial channels for the majority of the population. Reflecting institutional deficiencies, the imposition of a withholding tax on financial transactions undertaken by individuals who did not file tax returns has further encouraged a move toward cash and away from digital finance. On the demand side, Pakistan lags behind comparable countries on various indicators such as technical literacy, awareness about the functionality of products, and social inclusion of women. Nevertheless, we find that these constraints are severe only for specific subpopulations. Consistent with our conclusion that institutional weaknesses, the binding constraint, have created deep incentives to remain informal, a considerable proportion of mobile phone–owning financially excluded individuals continue to choose to be financially excluded in spite of having high levels of technical literacy, functional awareness, and trust.