Two questions. First, on the contingency provision creation of Rs. 21 crores during Q4 mentioned towards some specific accounts - which segment does this pertain to (NBFC, real estate, Commercial Banking) and what is the aggregate exposure at Bank's level? Second, on the rationale behind hiking interest rate on savings account and TD well ahead of industry - are you experiencing challenges in raising incremental deposits, or anticipating better growth ahead and hiking rates to maintain CD ratio?
On the contingency provision, these are normal business banking working capital cases. We made a risk assessment of what amount is covered through security and there was a recommendation from risk, hence we provided it. There is nothing specific to it. These are standard accounts based on risk assessment and we just provided an additional provision. On the liability rate hike, Sanjay responded: liabilities is not a one-data-point business. It is a combination of variables - how much you want to raise, cost of money, how you want to play CASA, retail versus wholesale liabilities, the overall CD ratio. Our whole last year performance saw CASA ratio stable at around 28%-29%, stable money around 80%, cost of money dropped by around 32 bps against our expectation of 15-20 bps, and ALM is a perfect match. We need to play with the environment - there is ALCO every month and RBI monetary policy every two months. Our 9-year journey has taught us that liabilities is a day-to-day business and we need to play every day.
First, on ROA - we are almost at 1.8% now on exit level. What would be the focus going forward - drive it up further or sustain at 1.8%? What are the levers available? Second, on margins - there was a 6 bps benefit from lower slippages and 7 bps from lower day count in Q4. Was there also an IT refund element and if so, how much? What is the overall margin outlook?
On ROA, Q4 is seasonally strong so 1.8% reflects that. Our goal is to maintain or achieve 1.8% on a full-year basis next year. Levers are: continued improvements in opex-to-assets ratio, and normalization in credit cost from MFI and credit card. These two in particular should drive full-year credit cost lower than this year. On the IT refund in margins, there was some IT refund but it was not material to the overall movement - it helped by a tiny bit but nothing specific to call out. On the overall margin outlook, we've seen strong improvement in cost of funds continuing for the last two quarters. But with the rate increase we've taken, cost of funds may have bottomed. Some seasonal factors from this quarter won't be there for the next quarter or two, so there will be some impact on margin.
First, on RBI ECL norms coming through - given the Fincare book experience and since we are growing that book again, what would be the impact on steady-state credit cost and how does the 90 bps guidance look with new ECL guidelines? Second, on geographical liability expansion - as we transition to Universal Bank over next two years, what will be the strategy for liabilities geographically, especially given higher contributions from UP and Karnataka on the liability side but asset-liability mismatch in those geographies?
On ECL, it's too early to comment and we as an SFB may not be covered under the ECL program. Vivek added: it's too early to comment. 90% of our book is retail secured plus secured commercial banking with very different loss given default characteristics - loss given default in retail asset has always been low. In MFI, 100% of incremental book is covered under CGFMU and around 92% of overall book is covered. Any guideline including the draft had a provision that government coverage would continue to be benefited. On liability geography expansion, the liability franchise is our most important strategic focus reviewed daily. Without any right to win in markets, we grew liability franchise by 23% this year. We have all facets: retail bank (~60% of overall deposit franchise), government business, corporate bank, commercial-led deposits, FIG, and treasury-led deposits (CDs). All are building up very nicely. As we become universal, the value of our franchise will go up, visibility will increase, and we'll double down on building it faster, more granular, and more impactful. We became a bank to build a liabilities franchise - our core leadership acceptance is around our liabilities franchise.
First, how should we look at AUM growth from southern geography and what proportion of incremental growth should come from south in three to five years? Which segments will drive this growth? Second, in two years since the merger, the performance of non-MFI segments often gets clouded by MFI weakness - how have early-stage delinquency trends been in non-MFI segments in southern geographies now that we have two years of growth data?
It's too operational now - we have become one Bank and it's not about Southern or Northern market. Certain branches perform, certain don't. Our overall growth estimation is linked to nominal GDP growth rate at 2x or 2.25x. In that, some products will work, some won't, some geographies will work, some won't. We have built 10-12 products in asset classes, around 2,500 touchpoints, want to grow distribution and team and expand into the market. We also want to cross-sell more to existing customer base. Our approach is to focus on overall growth matrix, overall NIM, other income, and credit cost. After 10 years, we will become more predictable in every sense. On the ROA target of 1.8% - margins might stay flat to slightly down due to mix shift, credit cost we are asking to build around 90 bps. We want to work on cost and credit cost - internally we project 0.90% but performance should be better than that. Once external environment becomes more predictable, India looks internally very bright and we are building this bank for India.
First, on technology - we have spent a good time discussing technological investments including Gen AI and Agentic AI. How do you see this translating into business volumes and what kind of cost ratios will you target over next two-three years as the Bank transitions to Universal Bank? As an SFB, cost ratios are in a narrow range (late 50s to 60s), but as a Universal Bank the range is wider. Where will AU want to position itself on cost ratios? Second, on asset quality - the Bank delivered 1.8% ROA which was guided for FY27 right in Q4 itself. How is asset quality shaping up and should we benchmark credit cost estimates around this quarter's number?
On technology and AI, we are investing a lot and AI is really helping us. We've launched our first AI-led LOS in Gold loans and given two-three loans in a 5-10 minute frictionless journey. AI will allow us to connect with people internally and externally seamlessly in their own language. Being a retail physical-oriented franchise, AI will help in two ways: productivity will go up and scale management will be better from a risk perspective. On cost targets, we've gone from 4.3% to 4.1% this year. Next year we should be lower than 4%. The first benchmark should be around 3.5% in three to five years. On asset quality, Q4 is seasonally strong and I wouldn't advise building credit cost around this quarter. We should build it around 90 bps, which allows franchise some risk-taking capability. The 1.8% Q4 ROA should not be seen as our permanent ROA given external challenges. AU is building a solid long-term franchise across people, distribution, channels, geographies, products and tech. After 10 years, we will become more predictable in every sense.
First, on asset quality - we have come out from the asset quality cycle. How are we looking to strengthen residual asset quality metrics like PCR, contingency provisions and ECL going forward? Second, on home loans philosophy - the book has been mostly flat. Will we focus on asset duration over yields in the next few years?
On asset quality, Q4 numbers tell the exact story. We've always been very strong on retail secured assets; commercial assets have been range-bound. The stress was coming from credit cards and MFI, both of which are now settling down. PCR is not a defined number - it goes by provisioning policy and there is no change in provisioning policy. On ECL, guidelines come into effect from 1st April 2027 and some impact on Stage 1 and Stage 2 will occur but accelerated provisioning at Stage 3 might get released. Sanjay added: Risk Committee meets every quarter to assess any additional provisioning requirements. If you take the credit guarantee book out, our PCR is around 70%, and our focus always remains to secure any probable loss through PCR. On home loans, the market has become too competitive with every new NBFC and HFC building their book in the affordable housing space. The risk-reward is no longer there so we are not going irrationally. We have a lot on the table to grow - we want to grow around 2x to 2.5x nominal GDP and we are playing a risk-reward game every year, growing wherever risk-reward is favorable.
First, how should we think about margins going into next year - the day count benefit won't repeat but benefit of lower reversals should stay since slippages are moderating year-on-year? Second, on fees - in Q4 there is usually a sharper seasonal uptick in loan assets and general banking fees, is there something missing this time? Third, on provision coverage - in the last call you talked about how you provide for CGFMU assuming recoveries come through on covered portfolio, but PCR is going up again this quarter - has anything changed?
On margins, cost of funds may have bottomed out with the rate increases we've taken. The 6 bps for lower slippages had two elements: lower gross slippages QoQ and higher NPA reversals. To the extent Q4 is seasonally strong and Q1 is weaker, you won't see this benefit in the next quarter. Your asset yield will reflect asset mix - unsecured portfolio will probably grow at a pace slower than the rest of the book, so on a net-net basis there may be some asset mix related pressure on yield. On fees, there is nothing specific to call out. On PCR, it is a function of accounting policy and where NPA is coming from, which asset class and what is the provisioning policy for that particular asset class. It's an outcome rather than an input.
First, on deposits - do you have an internal target for cost of funds through the cycle, relative to banks with the lowest cost of funds after you get the universal banking license? And what is the amount of retail deposits on the balance sheet as on March 26? Second, on growth outlook given macro uncertainty - when do you expect to take action on curtailing risk or pulling back credit, or do you not intend to do that?
On the long-term cost of funds target, internally we are pushing that cost of money should be around the repo rate prevalent at that time. Currently repo rate is 5.25% and my cost is around 6.75%. Mid-sized banks are around 6.25% and the old three-four banks are around 5.25%. Once we become very mature in our universal banking avatar, our cost of funds should be around the repo rate prevailing at that time - that's a long-term dream and target. On retail deposit amount, Gaurav added: branch banking constitutes about 60% of overall deposit and is primarily targeting retail deposits. Total stable deposits (CASA, retail TD, and non-callable wholesale deposits) is at 79%, stable versus last year. On risk management, we want to remain very risk-averse and wherever we find the indicators, which as of now are not there, we don't onboard customers in segments that might get affected. It will happen automatically because we have built our credit underwriting model around this.