BLOG

Banking 100 Million Pakistanis

Earlier this year, a team from Bankable Frontier Associates – David Porteous, Brian Le Sar, Johan Bezuidenhoudt, Ignacio Mas, Khurram Sikander, David Reed – and I started a project to understand how interoperability in retail payments affects the outcome of financial inclusion in Pakistan.

We picked a sizeable target: 100 million Pakistanis financially included electronically by 2020. This is a big stretch from the reality in 2012, but we wanted to push for some ambitious thinking with the lens of interoperability. The number is not that arbitrary because it took Pakistan roughly the same amount of time ending in 2011 to get to 100 million mobile subscribers.
 
Today we share with you the final report. You can download it here.
 
The project involved working closely with the State Bank of Pakistan and Pakistan Telecom Authority and a diverse cross section of stakeholders from banking, telco, third party entities and government agencies.  The report draws on case studies and data from South Africa, Ghana, Colombia, UK, Kenya, Brazil and on the diverse expertise of the team.
 
The report provides specific ideas for the Pakistani market which we intend to use in a workshop with the industry there. But the bulk of the report also provides insights on why interoperability might be important, how we should think about it from a policy and market development perspective and how it should be measured, especially as it relates to financial inclusion. Below are some brief highlights of the report:
 
Interoperability of retail payment instruments is not an objective in its own right; rather it is a means of achieving other desirable objectives. Like a lot of other countries, Pakistan has two such objectives: greater financial inclusion and a reduction in the use of cash for government payments. Both of those objectives combine in one vision we can call an ‘inclusive cash-lite society,’ where cash is still used but not pervasive.
 
We discuss in the report how policy objectives for and provider benefits from interoperability need to come together to realize that vision of inclusive cash-lite.  If interoperability is not an end in itself, then objectives used by policy-makers for interoperability in retail payments – from productive efficiency (no duplicative infrastructure) to dynamic efficiency (competition and product diversity) -- need to be prioritized and sequenced differently to achieve that objective.
 
Whether in Pakistan or elsewhere, it will be a challenge for business models to achieve a target of 100 million people with electronic accounts in 8 years. For Pakistan, the investment figure is staggering: USD 5.2 billion would be needed for the country to set-up one million new units of banking infrastructure, up from the 74,000 total branches, agents, POS devices and ATMs today.  USD 5.2 billion is four times the entire profit of the banking sector in Pakistan in 2011 and close to the total CAPEX of the all Pakistani MNOs reported to GSMA’s Wireless Intelligence in the five years to 2011. Even once that investment is made, it is unclear if the industry will meet revenue targets.
 
While using branchless channels just might half that total investment, it need not resolve all business issues. For instance, the industry would still need to resolve a paradox:  you need a massive build out of infrastructure where people can cash-in and cash-out of their accounts but the need for that infrastructure declines, and so do revenues for agents and owners of that infrastructure, as more and more people transact electronically.  The paradox suggests that there might be a time when interoperability is economically attractive to providers.
 
To enable adoption on a massive scale of the kind that would lead to 100 million people with electronic financial access within 8 years, and to improve the business case, markets would need to not just interconnect, but “effectively interconnect.” In the report, we introduce and define different levels of interconnection and how to use them in a new way to measure interoperability. For a country like Pakistan, effective interconnection means achieving both -- cost reductions passed on to customers and strong positive network effects on the demand side. Various scenarios which have either cost reduction or network effects are discussed in the report. If neither occurs, the trajectory of adoption will be modest.
 
Interoperability could change the trajectory of growth in Pakistan and even greater interoperability could unlock a pathway to cash-lite financial inclusion but it would have to be in defined-use cases and it would have to be handled with great care. We provide specific advice to the Pakistani regulators and explain how they could manage interoperability. For instance, agent-level interoperability could help rationalize that large investment that is needed, but if forced prematurely by the regulator it would depress incentives. On the other hand, by focusing in the short-term on full interconnectivity in specific payment uses case at the platform level, like Interbank Funds Transfers between bank accounts and mobile wallets, the regulator could help the industry take a step towards those network effects.
 
Download the report here.
 
 
 

Comments

07 September 2012 Submitted by Anonymous (not verified)

The author is overlooking an essential point that needs to be taking into account in Pakistan in a similar manner as co-authors Messrs. Porteous and Bezuidenhout should have done years ago in their country South Africa. Namely, government and banking sector need to be coherent in their strategy, its implementation and be held accountable on results (or the absence or weakness) according to agreed fixed objectives after a fixed period of time.

In South Africa, the 4 banks that dominate the banking sector promised government banking inclusion in exchange for not having legal obligations a decade ago. Still over half of the population is as then unbanked. As you know technology and partnership with world class telecom firms was not the issue in South Africa. The banks, 3 of the 4 largely being foreign owned, hold on to their oligapoly, including owning the National Payment System and their fat profit margins. They exclude financial institutions such as Capitec and Postbank from the National Payment System. Government did not take responsibility for the state of its economy, over its banking sector policy; the separation of responsibilities between Reserve Bank, Finance Ministry and Industry ministry is is hopeless. It still allows the big banks to do what they want and, on the other side, they have not come any closer to integrating semi- and informal finance (including Postbank, pawnshops, regulated moneylenders, the Stokvels and the Burial societies, the last two probably holding more deposits than the private deposits of banks) into the formal bank sector. And then I don’t speak of the cooperative MFIs; small, perpetually ailing and totally dependent on charity, political bodies and donors, while members, the key to success of coops, have nothing to say.

In Pakistan an increasingly more important problem between banks and government is how to implement financial inclusion of poor people. Also with modern technology, profits will be insecure and small as there is no legislation on labor contracts or minimum income (as salary or social security). Furthermore, Supreme Court of Justice already recognised in several cases that RIBA, the Islamic prohibition of Interest Rate, is (with Zakah, to take 2.5% of current accounts each end of Ramadan, which also faces challenges) the determinant obligation for banks but, because of also the secular legal system and the external relations of the banks and the economy, the interest rates may still co-exist. However, in the politicised area of “banking the poor” the Interest Rate prohibition approaches and the closure of the politicised credit focused cooperative banks (central government not being able to change the constitution that ensured provincial responsibility over coops) has made government believe that it is a good idea to quickly transform credit programs and NGOs into commercial banks.

Discussing “interoperability” is not the issue in financial inclusion. State Bank has produced good Branchless Banking legislation as a result of excellent collaboration with telecom firms, where legal responsibility is with the licensed banking partner – reason for which Telenor, a worldclass (Norwegian owned) telecom company, got a banking license by buying Tameer bank.

As I said, the core issue in Pakistan is to ensure the poor that government will collaborate with banks and civil society (including especially religious organisations, parliament and political groups) to develop a coherent strategy and policy with subsequently, exact performance indicators and a regulatory framework that can hold all stakeholders responsible for having emancipated the poor as citizens that are allowed to have a decent life, including managing their money as they wish with the help of professional, sustainable and accountable banks.

07 September 2012 Submitted by Kabir (not verified)

Peter, I share your view that no one thing is a cure-all for financial inclusion in South Africa, Pakistan or elsewhere. I am not sure if there is “the issue in financial inclusion” and I am sure a random poll of various stakeholders in both countries would throw-up different priorities. Ultimately, in this report, we take an approach that is similar to your suggestion in your last para that government action needs to be collaborative and results driven. In fact, from pages 27 to 38 of the report, we suggest how the SBP and PTA need to define success and build on the “good” work they have already done to date in branchless banking, as you put it. David, Johan, Brian and the rest of us are keen on a proposed new measure of interoperability in part because we think it is a step towards that exactness in measurement that you wish government interventions had. The new measure is described on pages 15-19 in the report and we will also summarize it in a forthcoming blog post.

Add new comment