A Struggling Ecosystem

Fintech is described as “software, mobile applications and other technologies created to improve and automate traditional forms of finance for businesses and consumers alike.”

The two facets of financial services – fees and lending – allow investors to make money by trading money. Here, I will broach both opportunities and analyse what impedes monetisation in the developing world, including Pakistan.

First, let’s touch upon the banking fees charged to consumers for domestic payments. A good example is a debit card. When you deposit cash at a bank, they give you a card that typically works on the same lines as global processors like Visa and Mastercard or Pakistan’s PayPak, operated by 1LINK. When using a bank card to pay for groceries at a store, you are paying electronically, and the grocer pays two percent of the amount as fees to cover the cost of the machine and the electronic transaction. Since margins are razor thin in Pakistan (grocers often sell below the suggested retail price, further dropping their margins) many merchants ask customers to pay an incremental two percent to cover their charges when they use their cards.

In China, there are very few card machines left despite there being over four bank accounts per adult. Two technology companies, Alibaba and Tencent, spent billions of dollars to give every individual adult and merchant in the country a personalised QR code. These companies charge 0.25% in processing fees compared to the two percent a bank would charge when a credit card is used. The low fee means that merchants rarely ask customers to pay the extra fee.

Visa’s market capitalisation stands north of $400 billion and Mastercard’s north of $300 billion, and taking a slice of their daily transactions is every investor’s dream (very few banks have a market capitalisation close to $300 billion anywhere in the world). To make money requires that the transaction be made electronically. If I give Rs 100 to person A, who pays person B in cash, who pays person C, who then pays person D, the value remains Rs 100 by the time person D receives it. Had these transactions been done electronically, even with a fee of only 0.25% (like in China), everyone, including the government, would automatically make 0.25% times four transactions.

Easypaisa and JazzCash started off charging up to seven percent in transfer fees in the early days. However, competition has driven these fees down to two percent and both are losing money because the cost to shopkeepers of handling this cash is higher than two percent. Most of the cash in circulation bypasses any fees. Businesses in Pakistan avoid coming into the tax net by settling through private IOUs, known as khataas and settling in cash later or through cash cheques paid in a single transaction. The fact is that in Pakistan most people do not put their money into bank accounts because of the excessive documentation required to show the source of the cash. This makes Pakistan’s currency to deposits ratio of 40%, one of the highest in the world.

As fees can only be earned if people transfer money electronically, more than half the money supply in Pakistan is inaccessible in terms of earning fees. Even 90% of the FMCG supply chain and e-commerce is transacted in cash and although there are attempts to digitise this, someone still has to figure out who will manage the risk of ferrying the cash around.

Perhaps an easier path to monetising via the fee channel is to tap into the digital value people already hold in their phones – the mobile credit top-up. In Kenya, when people top up Ksh 100 into their mobile credit, they are not taxed at the outset, and they can use the top-up to make payments. For example, if they want to pay someone Ksh 100, they can transfer the credit to the payee for a small fee. However, they will be taxed at the time when they want to buy a call or data package for KSh 30 – let’s say at the rate of 20% so six Kenyan shillings.

This is what has made Kenya the biggest success story in digital payments in the developing world, because even people in villages pay each other by transferring money denominated in mobile credit, without having to pay heavy taxes. This did not work in Pakistan, because the minute you top up your mobile credit – say Rs 100 – you are taxed between Rs 15 and 30, so any purchase made after that has to be paid from what is left after taxes, say Rs 70. You will never transfer Rs 70 to a person if you owe them Rs 100, because you have already paid a tax penalty of Rs 30. This simple mistake in taxation has cost Pakistan its biggest digitisation opportunity.

Lack of digitisation means that the source of people’s income is never digital. If the source of the income is cash, then the ecosystem will continue to struggle to convert this cash to digital. In the US, digital payments were adopted two to three decades ago with the proliferation of credit cards. Every citizen, including teenage college students, is expected to have a bank account. However, the minute they have a bank account, they are offered a credit card and a credit score that users can access at any time. The advantage the US has is that almost all global trade is not backed by gold, but by US dollars. China for example has $3,000 billion in its reserves, whereas the US has only $200 billion, mostly in euros and yens in reserves. This is because the US does not need to hold a lot of foreign currency reserves; it can simply print money (create it digitally), increase the money supply and pay anyone anywhere in the world.

Banks in Pakistan too can create money out of thin air, but only in rupees, where the demand is limited to Pakistan. Banks have a reserve ratio which means they do not need more than Rs 100 in deposits to lend up to Rs 500. The creation of money increases the money supply and the excess money supply helps fund budget deficits for the Government of Pakistan indirectly.

Half of Pakistan’s total debt is made up of loans from local banks – about $130 billion worth of rupees. If banks in Pakistan lend so much money to the government with no risk of default (the government can always print more money to pay back the banks) why would they bother to lend to people like you and me? This is why there are only 1.7 million consumer credit cards in Pakistan and according to banking officials, only half are unique (many have multiple credit cards).

There is a lot of pressure on banks by the State Bank of Pakistan (SBP) to change course and lend to small businesses. However, even if they want to lend, banks are reluctant to lend to SMEs because they operate mostly on cash and there is little record of their historical financial performance available.Even microfinance banks struggle to lend to small businesses and most lending is collateralised by gold (any lending without collateral comes with a high default risk). There is much excitement right now about the privately licenced non-banking finance companies, but the jury is still out on how reliable the paybacks are from SMEs, given that most work on razor-thin margins amidst stiff competition and low purchasing power. Successful SMEs are mostly those that take the money they are supposed to give the government in the form of taxes and pass it on to the consumer in the form of cheaper prices. This means that what appear to be 15% to 20% margins, are only realised by avoiding paying taxes. With inflation crossing the 20% mark (some say it’s nearly 40% right now), why would banks lend to SMEs?Then there is the problem of recourse. No one in Pakistan knows their credit score. While the SBP has allowed two companies, Tasdeeq and DataCheck, to act as credit bureaus, there is no private company (like Equifax in the US) that allows citizens to check their credit scores. NADRA’s Computerised National Identity Card (CNIC), to which every mobile number and almost every financial transaction is attached, is vastly underutilised. If every CNIC had a credit score associated with it that any individual could access (you can restrict who gains access to it by mandating a one-time password for authentication on the mobile number registered in the name of the user), people would at least have access to their credit score.

It’s not that people do not borrow money in Pakistan. In villages, people borrow from their neighbourhood stores. Then there are private lenders (illegal in three of the four provinces), typically lending at eight to 12% per month or others selling goods on instalments to the unbanked. Although banks could conceivably lend to consumers or initiate Buy Now Pay Later schemes (they have been a source of financing for decades for the unbanked), these lending operations are costly endeavours. Why would banks care to lend small amounts?

Let’s say I want to borrow Rs 10,000 to pay a deposit on a motorbike. If the bank charges two percent interest per month, that comes to Rs 200. Then, after six months, my bike is stolen. So not only am I unable to pay the interest, I cannot even pay my principal back. The bike was not insured, so I default on my loan. The bank lent me Rs 10,000, and got Rs 1,000 back in interest but has no way of recovering the Rs 9,000. So even though the bank was charging an interest rate of 24%, the cost of servicing and recovering this loan is too high relative to the Rs 10,000 that was lent. So banks do not bother to lend to this long tail of customers.

That leaves a window of opportunity for companies that are not banks to lend to the long-tail consumer and the retailer. But there are risks. If the cost of servicing is high for consumers, what about SMEs? What amount is good enough to lend to cover the cost of service?

Let’s say it does not make sense to lend less than Rs 500,000 to a small business. In this case, the SME needs to be turning this Rs 500,000 into Rs 510,000 every month to cover the cost of the loan. This is possible if the inventory turns multiple times in a month, but remember the inventory that turns the most has the lowest margins.

This is the arena where most fintechs are playing. Technology can be used to identify details regarding sales, payables and receivables from a merchant. However, if the merchant defaults on the loan, little can be done, other than sending some thugs to collect from him because he does not care about his credit score and companies that are not financial institutions cannot even report a default to a credit bureau to change a score.

To conclude, banks will not bend over backwards to lend to SMEs. Private lenders may, but they will charge between 20 and 50% in annual interest to cover the risk of default. Fintech lending is sexy but it is not for the soft-hearted. Boots on the ground are required for any lending or money transfer that involves cash and they all have a cost component. There will be more Know Your Customer than tech at the outset. This may change, but consumers are not going to be good citizens unless they know there are repercussions to default or fraud.

The solutions are simple but difficult to implement and require government intervention. I still believe the fastest way to digitise cash is Kenya’s path of not taxing mobile top-ups until they are used, so people can start using this money as a parallel currency to cash. The banks and tax authorities need to stop asking for the source of the cash if they want Pakistan to be digitised. Property, gold, car and currency dealers should be compelled to digitise because unfortunately, even government officials have their savings in assets that provide no growth to the economy. Every citizen has a CNIC and should be given a credit score immediately. The government should borrow from its citizens, in the form of National Savings Certificates that do not require clarification regarding the source of cash.Frankly, unless the government changes its tax regime, I do not believe anyone, except the large commercial banks, will gain from technological advances in financial services.

Muneeb Maayr is Founder, Bykea.

*CLARIFICATION:

I was a little disappointed that the title of my article was changed without my consent or acknowledgement. My reason for writing the article was not to spell doom for Fintech in Pakistan but to identify the pitfalls and infer solutions for a path forward. I believe the current title may be perceived as click bait that casts doubts on my reputation as a forward solution oriented thought leader as opposed to a brooder. My intent was only to lay out my learnings for others to learn from what I knew in order to find a way forward, not cast pessimism on Fintech in Pakistan. I would appreciate that this be respected.
- Muneeb Maayr