The devil in retail: Consumer financing may attract closer scrutiny by RBI

The consumer financing business is showing signs of heightened stress

The Reserve Bank of India’s (RBI’s) Financial Stability Report (FSR: December 2021) says the retail credit growth model is confronting headwinds: delinquencies have risen, and the new-to-credit segment is showing a dip in originations. It refers to the Bank for International Settlements’ (BIS) observation that in emerging markets, bad loans “typically peak six to eight quarters after the onset of a severe recession.”

Fintech firms — which claim to be smart with technology — are on a podium on which they would not like to be seen. The FSR shows their delinquency levels shot up by 274 basis points (bps) to 4.56 per cent in the year to September 21. True, state-run banks did report the highest figure for the end of this period on this aspect at 5.03 per cent, but it was an improvement of 45 bps.

The bandwagon effect

Rishabh Goel, co-founder and chief executive officer (CEO) of Credgenics, says, “The legacy culture in traditional lending led to a boom in digital lending. The easier paperwork and lack of additional data often becomes the questioning point from a collection angle and we wouldn’t deny it.”

There’s not much granular data on sourcing and delinquency in retail, save for a report by PwC in May 2021. At end-September 2020, over Rs 9 trillion was stuck in accounts classified in various degrees of delinquent and bad loans. This included standard delinquent accounts, sub-standard, doubtful and loss; and in multiple DPD (days-past-due) buckets ranging from 30 to over 720 days (excluding written-off and restructured loans).

In the March-September 2020 phase, the number of delinquent accounts dipped initially due to the moratorium offered by the RBI for loan repayments; it was followed by a resurgence. In the Q3FY21-Q2FY22 phase, a negative trend was seen in the DPD-90 and DPD-120 buckets — the number of delinquent accounts declined by 28 per cent and 5 per cent, respectively; and the amount outstanding in DPD-90 accounts fell by 39 per cent, and in DPD-120 accounts, went up only by a per cent.

This contrasts with a  growth of 12 per cent and 11 per cent recorded in the number of accounts and value of restructured loans respectively; and a growth of 20 per cent in the number (of loans) and 13 per cent in the value of what was written off recorded in the first two quarters of FY21.

Banks went in for retail lending when corporate loans came under stress, and tapped into the relatively new-to-credit segment which has little credit history. It was also aided by the fact that RBI in September 2019 reduced the risks-weights for consumer loans by 25 bps to 100 per cent, at a time when the economy was slowing. “The reduction in risk weights would encourage banks to increase their exposure to this loan segment at a time when credit risks are already increasing from a slowing economy,” said Moody’s!

Krishnan Sitaraman, senior director and deputy chief ratings officer at CRISIL Ratings, says that “retail and small businesses (which together form 40 per cent of bank credit), are expected to see higher accretion of bad loans.” Stressed assets in these segments are seen rising to 4-5 per cent and 17-18 per cent, respectively, by this fiscal-end. “The numbers would have trended even higher but for write-offs, primarily in the unsecured segment,” he adds.

There’s another detail in the RBI’s Report on Trend and Progress of Banking in India 2018-19 — a sharp fall in both incremental and outstanding credit of state-run banks, as nearly a dozen of them were sent under the prompt corrective action framework. And private banks’ share in incremental credit shot up to almost 100 per cent! “This will also be reflected in the share of private banks in incremental retail credit,” adds Sitharaman.

The big picture

The RBI had flagged the issue of retail stress when banks had started to get on to the bandwagon. In FY05, the central bank hiked the risk weights on such loans by 25 basis points to 125 per cent — the opposite of what it did in September 2019, when stress was showing up in the economy.

Shyamala Gopinath (then deputy governor), too, had warned (Retail Banking Directions: Opportunities & Challenges, May 28, 2005) about the dangers of outsourcing, a model which is much in vogue now: “It has the potential to transfer risk, management and compliance to third parties who may not be regulated.”

She referred to the BIS report, Outsourcing in Financial Services, which noted: “…a regulated entity seeking to outsource activities should have in place a comprehensive policy on outsourcing, including a comprehensive outsourcing risk management programme to address the outsourced activities and the relationship with the service provider. Application of these principles in the Indian context is under consideration”.

Seventeen years on, this is exactly what the RBI is trying to address (Working Group on Digital Lending including Lending through Online Platforms and Mobile Apps).

And the first-loss default guarantee (FLDG) extended by lenders — of all hues — is up for a review. The FLDG is the amount offered by digital platforms (or other unregulated entities) as a guarantee to lenders on whose behalf they source business. This is to protect lenders when loans go bad. The FLDG acts as a demonstration of under-writing skills, whereas from the lenders’ perspective, it ensures the platforms’ (unregulated) skin in the business.

“These arrangements between lenders and digital platforms are gaining in traction and RBI is seeking to standardise it. This is akin to what the RBI did when it came to securitisation back in 2005, when it said you have to hold an asset for at least six months on your books before you sell it down, and also that you can’t book the gains upfront but will have to amortise them,” explains Anurag Sinha, co-founder and CEO of OneScore.

For all practical purposes, credit risk is borne by the lending service provider (LSP) without having to maintain any regulatory capital. “The loan portfolio backed by FLDG is akin to [the] off-balance sheet portfolio of the LSP, wherein nominal loans sit in the books of the lender without having to partake in any lending process,” the working group observed.

“There are several shadow banks that have worked on the 100 per cent FLDG model in the last few years with fintechs, which was a bottleneck. With a standard percentage, it would give benefit to fintechs for only locking the amount which can be potential bad-loans for running the business,” says Artem Andreev, country head for RupeeRedee.

Look at it another way. The corporate loan book is relatively clean — due to provisioning, write-offs and bad-loan transfers to asset reconstruction companies. Retail, however, from a regulatory standpoint, is seen as homogenous, and as “secured” and “unsecured” even as the behaviour of each product-line is different.

This is unlike banks’ corporate exposures, where a financial guarantee attracts 100 per cent risk-weightage, while for performance guarantees it is 20 per cent. This lack of differentiated treatment has led to banks going all out to grow their retail books, to compensate for lacklustre corporate credit. The bourses have given retail banks better valuations, exacerbating the plot. If the RBI were now to tighten the screws on retail, it will affect consumption demand. Figure this one out!