Bank of the Philippine Islands (BPI) Earnings Call Transcript & Summary

Okay. Good afternoon, ladies and gentlemen, and welcome to BPI's Fourth Quarter 2021 Earnings Call. I am Chinky Lukban, Head of Corporate Strategy. Just some housekeeping reminders before we proceed, we would like to request all participants to use the log-in convention of your company name and then your first and last names. Keep your lines on mute so as to minimize background noise. [Operator Instructions] I am pleased to introduce our speakers this afternoon. With us are BPI President and Chief Executive Officer, TG Limcaoco; and Executive Vice President and Chief Finance Officer, Tere Marcial. We will start with a few words from Mr. Limcaoco, and Ms. Marcial will go through the deck sent out earlier. This will be followed by a Q&A session and then wrap up with some final thoughts from our CEO. TG, please go ahead.

Thank you very much, Chinky, and a warm welcome to everyone on this call. Our -- very pleased with our year-end results. Our income is PHP 23.88 billion, as Tere will mention, which is 11.5% better than last year's, really driven by lower provisions, but really also what a great result given the fact that our trading income was practically very little compared to the previous years. Tere will have more details on that. After her, I want to talk a little more about our digital strategies and the results we're getting to give you numbers from that. And we have purposely left a lot of time in this call to answer many of the questions that we've been getting. So let me, at this point, turn it over to Tere.

Thank you, TG. Good afternoon to everyone. Let me present our financial and operating performance for the fourth quarter as well as the full year 2021. To start off and to give context, here is a recap of the macroeconomic highlights. 2021 was a recovery year, notwithstanding the resurgence of the pandemic with new COVID-19 variants. Global GDP is estimated to grow at 5.9% in 2021 from a 3.1% contraction in 2020. Major central banks have kept policy rates unchanged. The Fed, however, turned more hawkish as U.S. inflation was at 40-year high and unemployment rate has declined to nearly pre-pandemic level. Global inflation rates have increased rapidly, reaching its highest in 2 decades. In most cases, higher inflation reflects pandemic-related supply disruptions and higher commodity prices compared to their low base from a year ago. Philippines' GDP grew at 5.6% in 2021, following a contraction of 9.6% in 2020, underpinned by the reopening of the economy and the fiscal and monetary stimulus put in place to support the recovery. Labor market has improved substantially since the onset of the pandemic. And while consumer prices have increased sharply, the BSP has committed to keep monetary policy rate accommodative until the economy has firmly recovered. Domestic interest rates increased and the yield curve has steepened on inflation and supply concerns. We have seen meaningful progress in the country's vaccination program. The country's COVID-19 cases appear to have peaked in January 2022. And with a slowdown in cases and continued vaccination program, we expect sustained reopening of the economy. The economic recovery in 2021 was reflected in our business volumes, our clients' activities and our profitability. Notwithstanding the weak first quarter, the second, third and fourth quarters showed sequential growth in loans and deposit balances and the improvement in asset quality. For the full year 2021, the bank generated a net income of PHP 23.88 billion, up 11.5%, driven by lower provisions and record fee income. NIM stood at 3.3% down by 19 basis points as asset yields remained low, offset by lower cost of funds. Cost to income ratio was 52%. Asset quality remained resilient during the pandemic. The bank's NPL level turned out better than initially expected and better than industry average. Improving asset quality and ECL outlook allowed for lower credit cost. Our balance sheet expansion reflects recovery in demand across deposit and loan portfolios. Credit cards, mortgage and micro finance led in loan growth, while we saw strong deposit growth across Corporate, SME and retail segments. The decline in borrowings due to bond maturities reduced our funding cost. We continue to make progress in our digital agenda. We saw increasing digital propensity among our clients with higher transaction count and higher transaction value in digital versus branch channels. Strong growth in digital fees was sustained across all platforms. Higher earnings for the full year delivered an ROE of 8.4%, an improvement from 7.7% in the previous year. The bank posted a net income of PHP 6.4 billion in the fourth quarter, 51.2% higher than the same period last year and 13.2% higher quarter-on-quarter. Improving mobility conditions fueled the sequential growth in our revenues, up by 9.7% quarter-on-quarter to reach PHP 25.8 billion, a record high. Fourth quarter 2021 fee income was up 15.5% year-on-year with each quarter showing sequential gains. Full year fee income of PHP 23.8 billion, up 23.2% provided revenue stability during a period of weak trading income and NIM contraction. Moving on to our loan book. Demand was generally tepid for most of the year before gaining traction in the third quarter, which was followed by a strong pickup in the fourth quarter. Loan balances stood at PHP 1.53 trillion, up 5.2% year-on-year, led by growth in corporate, credit cards, SME and mortgage. Notably, mortgage, credit cards and micro finance loans have rebounded to pre-pandemic levels. Let's take a look at net interest margin. Quarter-on-quarter, NIM increased 2 basis points to 3.3%. NIM has been trending higher in the last 3 months of the year, with December NIM at 3.36%, the highest since January. We expect this trend to continue in a rising interest rate environment. Deposits grew 13.9% year-on-year, with each quarter showing sequential growth and all client segments contributing to the growth. CASA ratio modestly declined in 2021 after a sharp increase recorded in 2020. Borrowings declined in the third quarter due to bond maturities, improving our funding costs. This was replaced by PHP 27 billion in bonds listed recently with a coupon rate of 2.8%, 125 basis points lower. Additional bond maturities this year should further improve funding costs. We delivered another record fee income in 2021 with all businesses showing strong growth. Touching on a few highlights, beginning with our 3 biggest businesses. For credit cards, the growth in fees was largely from increase in billings and from aggressive collection efforts. Wealth management reported a new high in AUM, which stood at PHP 1.1 trillion, up 6% year-on-year, driven by -- largely by strong net flows. The increase in branch service charges was largely from recovery in volume and reinstatement of fees that were suspended in 2020. The following businesses posted the highest growth in fees, ATM and digital channels. Growth in fees was largely from the implementation of ATM acquirer-based fee charging beginning 2021 as well as higher fees from open banking as we onboarded new partners in our platforms. Stock brokerage and investment banking, on higher trading volume for BPI securities and increase in the number of deals arranged by BPI Capital last year. Pipeline for 2022 continues to be robust, a potential indication of pickup in corporate loan demand. Retail loans. The increase in fees from higher volume resulting in higher auto and housing loan processing fees and service charges. The chart on the right shows the increasing contribution of fee income to total revenue, which stood at 24.5% in 2021 from 22.3% prior to the pandemic. We expect this number to grow and provide stability in our revenue base over time. Moving on to operating expenses. Operating expenses stood at PHP 50.7 billion, up 5.4%, mainly on tax spend and variable expenses with corresponding revenues generated. Cost to income stood at 52%, retracing 2019 [indiscernible] process more transactions with less branch footprint and a much leaner organization with headcount declining by 10% from pre-pandemic level. In 2021, we co-located 55 branches, bringing our physical branch count to 814. More branches are scheduled for co-location and closure at this year as clients increasingly migrate their transactions to our digital platforms. In 2021, we booked much lower provisions, amounting to PHP 13 billion, as our forecasts on key macroeconomic variables continue to improve and expected credit loss declined accordingly. Credit costs for the year was 93 basis points, significantly lower than 196 basis points in 2020. Asset quality has been resilient and demonstrated steady improvement since the second quarter of 2021. Our NPL level was lower at PHP 38 billion. NPL ratio was at 2.49%, down 19 basis points from the previous year. And NPL cover increased to 136% from 115% the prior year. As we have bought significant provisions in 2020 and 2021, any potential upside in asset quality will bode well for future earnings. Our capital position remained robust with a CET1 ratio of 15.9% and CAR of 16.8%, slightly lower compared to 2020 levels, primarily on higher credit risk-weighted assets. Last year, the bank paid regular cash dividends of PHP 1.80 per share, equivalent to a dividend payout ratio of nearly 38%, one of the highest among listed banks. In closing, we ended 2021 and ushered in 2022 from a position of strength, a fortress balance sheet, high-quality loan book, a diversified and growing fee income base and cost-efficient growth engine. We look ahead with optimism as we aim to lead in this period of economic recovery. Let me thank you, and let me now turn you over to our CEO, TG Limcaoco, to give you an update on our digital initiatives.

Thanks very much, Tere. So really, despite the challenges presented by COVID-19, we remain committed to our digital agenda. Last year, we created a digital governance framework where the business units as platform owners worked closely with our group called digital channels, who are responsible in building out those platforms and creating the digital ecosystem. This collaboration and reorganization has yielded results. We clearly expanded our digital customer base, and we did not only serve our customers better, but also improved our efficiency and our profitability. We are on track to roll out our fifth platform, [ BizCo ] in March. [ BizCo ] is tailored to the needs of our SME clients, providing them with solutions for payments, for payroll, invoicing, billing and collection. Work is ongoing on our other platforms for the broad market, which is an E-wallet with loyalty, and on our platform for the high net worth individuals. Both are lined up for release second half of this year. Also for release in the second half is the BPI trade mobile app, where aside from stock trading, our customers will be able to transfer cash in real time, subscribe to IPOs and investment funds, access reports and realtime market data. We are very excited about these 3 new platforms. The progress we are making on our customer platforms is really just the tip of the iceberg. Below the surface, there is also a transformation that is happening as we reengineer the bank to be operationally efficient, safe and secure from cyber threats and being able to leverage the data as banking becomes more and more data driven. Now let me talk about our customers' digital propensity. The growth in enrolled accounts and active users continues to be strong. Today, of our 8.46 million client base, 4.9 million or 58% are enrolled in our platforms, while 3.24 million are active users, up 20% from the previous year. As at year-end, our active users penetration rate stood at about 38%, up from 31% in the previous year and from 17% in 2016. Clearly, these numbers are good, but they can be improved in the years to come. Talking about our clients' digital propensity, this was no doubt accelerated by the pandemic. As of December 2021, 91% of total transactions in the branches are digital, up from 89% in the previous year and 85% prior to the pandemic. In terms of the transaction amount, 29% go to our digital channels from 25% prior to the pandemic. The digital propensity of our retail and corporate clients also improved year-on-year. As of December 2021, 92% of our retail client transactions are digital, up from 89% in the previous years, while 81% of our corporate client transaction [ counts ] are digital, up from 78%. We saw similar upward trends in the digital transaction amounts in both retail and corporate clients. If we want to talk about gross transaction value, the total value of our digital transactions amounted to PHP 5.9 trillion in 2021. That's up 47% from the previous year. We recorded improvements in the daily average transaction count and value from July 2019 to December 2021. As of the end of last year, we processed a daily average of 946,000 digital transactions, higher 27% than the previous year and 2.9x the average transaction count in July 2019. The daily average transaction value also exhibited the same trend. As of December 2021, it reached PHP 17.3 billion, up 30% from December 2020 and 2.8x that level in July 2019. The digital fees generated by our platforms, including ATM and open banking, amounted to PHP 7.3 billion in 2021, 30% higher year-on-year. Each platform showed strong fee growth, which we attribute to our expanded digital client base, our clients' strong digital adoption, the enhancements we have put on our platforms, including the ability to onboard new products and new partners. Last year, we onboarded close to 35 new partners in our open banking platform. In the succeeding charts, I will be discussing our digital metrics. These are the metrics we use to monitor our progress and the value we create for our shareholders. Let me start with our definition of digital customer, and we will continue to refine this definition and explain our definitions as we change them going forward, but we will try to maintain some consistency. Today, we define our digital customers based on their behavior. First, if they open an account in any of our digital platforms. For now this pertains to deposit accounts opened using the New to Bank or the New to Product features in our retail platform. New to Bank has received over 322,000 applications since its launch in June last year. Secondly, they are also considered digital customers if more than 50% of their client-initiated financial transactions are done via the digital platforms. For example, fund transfers, bill payments or purchase of investment funds. Then we requalify digital customers on a rolling 12-month basis. If a digital customer doesn't meet the criteria, the client is reclassified to a non-digital customer in the next period. So all income and expenses associated with the product that a digital customer has, whether it is a digital product or a non-digital product will be part of that digital segment. The number of digital customers we have, increased almost fourfold in the past 3 years. As of last year, last month last year, 1.87 million or 22% of the bank's retail client space are what we call digital customers, up from 6.5% in 2019. The chart on the right shows that digital customers are more engaged, doing 3.2x more online transactions compared to non-digital customers. Not surprisingly, digital customers generate 2.4x more revenue than a nondigital customer. We saw this across deposits, across credit cards, across loans and other products, including investments and insurance. Moving on to expenses. In the past 3 years, the cost of servicing a digital customer has been declining, in contrast to the increasing cost of serving a non-digital customer. In 2021, serving a digital customer costs 10% more than the non-digital customer. The higher cost is largely from higher engagement. And while it does cost more to serve a digital customer, they generate significantly higher revenues. Therefore, digital customers are more efficient as they generate more income per unit cost. From 2020 to 2021, we saw a decline in the cost-to-income for digital customers and an increase in the cost-to-income for non-digital customers and the widening differential between the two. In 2021, the cost-to-income ratio for a digital customer was 26% versus the 58% cost-to-income of a non-digital customer. We expect the differential to continue to widen over time, with digital revenue growing faster than cost as we add more products, more services and expand their ecosystem to improve -- and to improve the UI/UX of our platforms. In summary, serving digital customers costs more, but generate significantly higher revenue and therefore is more efficient than a non-digital customer. Moving forward, we will use customer count, revenue cost and cost-to-income to measure our progress in digitalization. Finally, we get a lot of questions about our relationship with GCash, so let me summarize what we've done. On December 20 -- on December 6 last year, we launched 2 mutual funds on the GCash platform. The first one is a feeder fund that invests in a multi-asset global fund managed by BlackRock. The second is a local fund that tracks a PSEI index managed by BPI Investment Management Inc. Since its launch, we have acquired about 113,000 clients -- sorry, since it's launched [ 2nd ] January '19, we have acquired about 113,000 new clients or about 3,800 clients daily. The aggregate investment amount is small, but we are happy with the acquisition of new BPI clients, which would not have been possible without the partnership of GCash. We also launched BPI/MS auto Insurance on GInsure last December 13. The volume is understandably small as the product is only relevant for vehicle owners who are looking to register new or existing cars. We are, however, pleased that the minimum viable product is already out with GCash, and we will continue to look for ways to improve our product offerings. GCash remains our top partner in open banking, with transaction value boasting a 116% increase year-on-year. From January to December 2021, average daily transaction -- the average daily transactions increased nearly 60%, while average daily transaction value increased nearly 70%. We also have a partnership with GCash on client acquisitions for our credit card and personal loan products. This is an ongoing arrangement that started last August. We will be launching a deposit product under GSave this first quarter, and the products under GLoan will be in the second half of this year. Finally, as we reinvent the bank and banking, we continue to make strategic investments in technology. We have committed significant resources to advance our technology agenda in 2022, investments which we believe are needed to enable us to deliver banking excellence and [indiscernible] and the best digital offers and create value for our shareholders over time. Allow us to open the floor now for questions. Chinky?

Yes. Thank you, TG and Tere. We will now begin our Q&A session. [Operator Instructions] Our first question comes from Robert Kong of Citi.

It's Robert from Citi. Can you hear me?

Yes.

Okay. My questions are mostly on 2022 guidance. Clearly, the asset quality has come in much better, and I think the original peak was supposed to be 4%. It will be very helpful if you could guide on where you think the NPLs are going for this year, and of course, the credit costs are very important. Second question is, could you explain how your NIM has improved in the fourth quarter? I'm assuming you've taken advantage of improving yield curve. And what are you guiding for NIMs this year, including whatever assumptions you have on the yield curve utilization as well as maybe any BSP rate hikes? And final question is if you have any guidance on ROE for this year, that would be very helpful?

Robert, let me answer your questions. So first, on the NPL 2022 guidance, you will note that we ended 2021 with 2.49% NPL ratio. And we are expecting some further NPL formation, though benign, such that we're looking at around 3% NPL ratio for 2022. We think this is a very conservative assumption, especially given the recent leading indicators for better economic activity, better mobility and the whole reopening story, plus the 7.3% GDP growth assumption that we have put in our 2022 guidance. So 3%, but I think it might be quite conservative and could surprise on the upside. Related to this expected NPL forecast, we're looking at credit costs to go down significantly from 90-something basis points in 2021 to maybe 60 to maybe 65 basis points in 2022. So that significant improvement is because of the fact that we have really aggressively provisioned in 2020 and 2021, such that we are very comfortable with the ECL buffers that we have accumulated such that it really allows for even moderate to adverse scenario in terms of our ECL and NPL expectations. NIM for Q4, how did we show improvement in October, November and December? Large part, really, Robert, is attributed to the reduction in our cost of funds. We were able to benefit from the reduction in cost of borrowing, particularly our debt securities given some maturities that we have seen in -- large maturities that we have seen in September. There are some segments in the loan portfolio that benefited from higher repricing, particularly on the SME and maybe some parts of corporate, the middle market corporate. But for the most part, it was still -- we saw lower our P&Ls for the large corp. But the big benefit is really from the lower cost of funds, thus allowing us to increase our NIMs for October, November and December. What is our expectation for 2022 NIM? A lot of it really has to do with the cyclical nature of the banking book where we expect higher loan repricing. We are putting in a 75 basis point assumption in terms of policy rate hike delivered in 3 tranches. That's our base case assumption. And with a 75 basis point policy rate hike, we do expect NIM expansion of between 15 to 20 basis points for the full year 2022. And finally, ROE expectations. I think we have seen the low at 7.7% in 2020. 2021, 8.4%. We expect a further increase to maybe 9% to 10% if we deliver on our revenue projections. So looking really to move in closer to pre-pandemic ROE and ushering in higher ROE targets for the medium term.

Can I just clarify 2 things. One is on the NIM guidance. You said you're assuming 75 basis points higher rates this year, which will give you a NIM expansion of 15 to 20 basis points. But I assume that some of that NIM expansion -- sorry, those rate hikes are going to be back-ended to the second half of the year. So could you just give us just a sensitivity on a full year reprice rate hike, what would that mean for your NIM on a gross basis? So that's one clarification, if I may.

I'll just give you a very rough estimates, Robert. I would say around 10 basis points for every 25 basis point policy rate hike. This could be better or worse depending, really, on timing of the hikes. As you said, it's most likely going to be back-ended. Especially, we think the BSP may not hike policy rates prior to elections. So a large part of the NIM improvement is not just coming from loan yield repricing, but also on the significant reduction in our cost of deposits because we don't -- we are not yet reflecting the full benefit of lower CASA cost, lower CASA interest rates, a lot of which we did in the second half of last year. So we expect the full benefit for the most of 2022. And then finally, we do expect further bond maturities in 2022 at much higher cost of borrowing, which we don't intend to roll over completely. And if we do roll over, as we have seen in the recent bond issuance, where we saw a 125 basis point improvement in the Q1 rate.

So one of the very small final clarification. This is on the relationship with GCash. It's quite clear you've got an enormous amount of product that's going through their platform. But is there any degree of, if not exclusivity, but at the least preferential use that would sort of drive your products in front of others on the platform?

To be fair, Robert, there is no exclusivity because we are independent. And GCash really is only 40% owned by Globe, right? But what we do have is a relationship where I would like to think is a -- let me be very careful about what I say, right? It's a relationship where we know we are very close, and therefore, we get some advantage. So for example, when you go to the GCash platform and look at the investments available, we are the first on the list even though we are not alphabetically first. And that's what we would use. That's what we would take advantage of.

Thank you, Robert, for your question. The next question will come from Selvie Jusman of Morgan Stanley.

Can you hear me?

Yes, very well. Go ahead.

I have 2 questions. So firstly, on potential cost saving going forward in the medium term as well as longer term. As TG mentioned earlier, that the cost of income ratio of the digital side is actually much lower and you could see potential to further scale up. What does it mean for the overall group and also taking into account your near-term investment need? So I'm just trying to think about the cost-to-income ratio in the longer term, how much potential we could see? So that is my first question. And then my second question, could I get a little bit more detail on the credit lease RWA in the fourth quarter as well as what is the expectation of loan growth in the coming years, and therefore, the capital consumptions? Yes, so those are my 2 questions.

Yes. I'll leave it to Tere. Tere, maybe you address also the cost-to-income ratio target that we're looking at.

Sure, TG. For 2022, we ended, as I mentioned, at 52%, which almost retraced 2019 level. 2022 is an investment year where we will significantly increase particularly tech expenses that will increase our cost-to-income ratio to high 50s. And then the long-term target, so immediately after -- by 2023, we are expecting this ratio to immediately go back to today's level, around 52%, 53%, and the long-term target of around 45% by the end of our medium-term plan, which is 2022 to 2026. A lot of this really will -- because of contribution of higher digital revenues, higher fee income and of course, the growth in our revenue coming from core intermediation business. So it's both coming from the revenue -- largely from the revenues from both digital as well as traditional customers. In terms of credit risk RWA growth in Q4, it's really in large part the top corp and large corp exposures. As I mentioned, it was -- demand was tepid for most of the year. There was some pickup in Q3 and a further acceleration in Q4, which is, again, a good sign of recovery in the quarters ahead. So the increase in credit RWA was mainly on -- from corporate credit. What is our loan growth expectation for 2022? We really are looking at 8% to 10% growth. I think from all across our loan portfolios, we're expecting a recovery. But the big contributors, at least in terms of accelerated growth will be coming from our retail business, particularly credit cards and personal loans, especially as we have seen improvement already recorded in Q4, and that we expect to be sustained for the rest of 2022. We also expect corporate growth -- corporate loan growth to also pick up in 2022. But I believe this is more back ended, so we're looking at higher growth in the second half for large corporate accounts.

Sorry, can I clarify? [indiscernible] For the cost-to-income ratio, did you mention earlier, the cost-to-income ratio will go up further to 50 -- sorry, that is 53%, you mentioned, in the middle of 2023, is it, before it starts to fall to about a [indiscernible] level in -- by the end of the year?

Yes. Well, of course, we are looking at projections over a 5-year period. And starting off with this year, which is an investment year, we expect cost to increase. And then we'll go back immediately in 2023 to around 52%, 53% before it goes to 45% target by the end of 2026.

Thank you, Selvie. Our next question is from Rachelleen Rodriguez on the chat box. She's asking, may I ask the [ driver ] for the downward NPL trend to 2.5% in 4Q versus 3Q 2021? What is your NPL outlook for 2022?

For the fourth quarter, we really saw improvement in NPL levels from across our loan portfolios. The biggest improvement actually came from credit cards, which is a combination of the result of aggressive collection efforts and at the same time, some regular write-offs, which is really our standard for credit card loans. And then we saw also significant improvement in NPLs for housing, mortgage as well as auto, and we also saw an improvement in SME. So those were some of the drivers for our 4Q NPL improvement. And did you ask the guidance for 2022?

Yes.

We had mentioned earlier that we're looking at 3% NPL ratio, but we believe this might be a quite conservative. And with the whole reopening story, we might be in for a lot of upside on NPL. Upside, meaning better portfolio, lower NPLs.

Thank you, Tere. Our next question is coming from DA Tan of JPMorgan.

TG, Tere, can you hear me?

Yes, we can hear you.

All right. Just 2 questions from me. One is on the costs in 2023. I understand you're programming a sharp increase this year, but it will go back to 2021 levels next year. Question is, what kind of cost growth are you thinking next year? Or are you seeing that digital would bring in significantly higher fee base? So what are the drivers of that sharp decline in cost-to-income? And I'll go with my second question as well. Second question is on digital, I mean, number of customers. If we look at it, I think the increase in digital customers have been impressive the past few years. But I do see that the overall customer count is down. I just want to understand the reason why it's down and how you're thinking about customer acquisition going forward. So this 8 million, should we expect it to increase very sharply in the next few years?

DA, let me address the question on cost. So really, these are long-term assumptions, but the significant decline in 2023 will really be because of the fact that we are front-loading a lot of investments this year. So there'll be a spike this year and an immediate correction in 2023. What are these? A lot will have to do with our tech investments as we enhance existing platforms and invest in new digital platforms and digital functionalities, as explained by TG earlier. And that will be front-loaded this year and will be normalized in 2023, which is why we expect cost-to-income to go back almost immediately to 2022 levels, the next year. So that's our plan. And in terms of -- the question is on customer count?

The revenue growth.

Revenue growth, Tere.

Sorry, revenue growth from digital?

Yes.

You want to address that, TG?

Sure, sure. DA, yes, we're also looking at significant revenue growth in 2023 versus 2022. Primarily, you've got the non-interest income. That one, we project will grow maybe just under 20%. And then obviously, the interest -- the net interest income should also be better as we -- as you will have the full effect of the higher rates given what we expect the rate hikes towards the end of this year. So those 2 will contribute a significant lift in the revenue side, thereby bringing your cost-to-income down back to the current levels of about 51%, 52% -- sorry, 52% or 53%. Your other question was on the customer count...

Customer count, we have actually shown almost flattish level in terms of customer count because the focus over the last, call it, decade, the past decade really has been on deepening. So we wanted to -- we have focused on increased engagement, not only in our digital channels, but also of digital through financial advisory, better offerings, increase in revenue per customer, increase in product penetration and product per customer. So that has been the focus, and that was part of our strategy in this past decade. But what we have also determined, particularly in our last strategic planning exercise, was that there is a lot of potential and this is -- there is a lot of growth to be had if we are to or aggressively grow our customer base. And we have mentioned in some discussions with investors that we really have very aggressive customer acquisition targets over the medium term. So it's been flattish in the past decade because of focus on deepening, but we are changing that strategy going forward as we increase customer count aggressively.

Sorry, can I just follow up on the cost side? Just wondering on 2023, you did see there's a significant deceleration. My question is, if we get this cost base this year, should we expect a decline in costs next year or just a deceleration in the pace of growth?

Yes. We're really looking at a deceleration in pace of growth for 2023. It's not cast in stone. Of course, we have a program. We have an investment program, and we have various projects at different stages of implementation. And as we have seen in 2020 and 2021, there were some project milestones that were delayed that really pushed expense this year. So part of the increase this year is really related to some of the delays related to COVID that is pushed forward. And depending on our speed of implementation this year, there is also a potential -- possibility for a slide. I guess the way to look at it, DA, is look at our cost increase between the end of 2020 to the end of 2023. So to give you really a more -- I guess a more long-term view of how our cost evolution or cost trajectory is really shaping because these are long-term investments, and project implementation take, on average, about maybe 2 years and then the revenues associated with that when we look at the financials. It's really over -- the revenue and cost is really over a space of 5 years. So the way we look at investment is really long term and give some, I guess, leeway for a slide and look at it point-to-point over maybe a 2-year period. That might give you a better picture.

Our next question comes from Karthik Chellappa of Buena Vista.

Yes. Am I audible?

Yes.

Okay. Great. I have 3 questions. The first question is if we were to look at the time deposit growth in 4Q, the sequential increase of almost close to PHP 90 billion, it's one of the sharpest that we have seen sequentially. What is the incremental cost of these time deposits?

Okay. I don't know the actual incremental cost. But what we were doing there, Karthik, was really a decision that we had made at ALCO to see how well we could really tap that base. So we went out towards the end of middle November, towards December, just to build it up to see how much we could put on the books given the strategy this year is really to -- we have a bond maturity that -- what happened in September, we have 2 bond maturities this year. And we only finance maybe less than half of that with the bond that we just launched. So going forward, we wanted to make sure -- I wanted to make sure with my team that we had the ability to tap the market -- the branches to tap the deposit market. So we went out -- talk about that amount, PHP 90 billion, probably like, I think, 1% on the average for maybe a month. So that was the incremental cost. And we're letting that go now. It's really to make sure that we're in front of our clients at all times and you don't disappear. I have a belief that you just can't be absent from the market for a long time and you've got to be able to show your clients a product, and then let it go and then make sure that you just keep in touch with them, so you're talking all the time.

Because [indiscernible]...

Sorry, Karthik, let me also put in context that in September of last year, we had a PHP 34 billion maturity. That bond carried of around 4.05%. And because we had planned for our bond issuance only for January, which was actually listed a couple of days ago where we generated PHP 27 billion, we really wanted to make sure that the funds stayed with the bank and took advantage of the liquidity of these customers and place it in the interim in time deposit. So that contributed to the significant increase in TD in Q4, but not even close to the PHP 34 billion maturity. And if you look at the cost, as TG said, 1% for TD versus the 4.05% bond that it replaced.

Yes. The reason why I brought it up is because your year-end funding cost is close to about 80 basis points. So a 1% means on an incremental basis, it is slightly higher than your weighted average cost, right? Not by a lot, but it still seems higher, right? Will that have any impact at all on your cost of funds going forward? Or will these just wind down by 1Q?

They'll all practically wound down by January.

Okay, okay. That's good. The second question I had was if you look at your credit cards, the growth in the loan balance of credit cards is about 11.7% for the year. But the growth in the fee income from cards is 14.9%, which means it's far higher than the loan balance. So what are all the non-interest portion of the card-related fees [indiscernible].

Well, the -- I guess the bread and butter of card -- you're asking card fee, Karthik, right?

Yes, yes. Card fee growth is higher than the card balance growth, so obviously there is some non-interest portion, whether it's bank card fees or any other kind of income that has grown strongly. Just want more color on what that stream of income actually is?

Okay. Well, I....

Okay. [indiscernible] on the line.

The biggest contributors to our card fees; number one, the interchange that we get through the merchants; number two, the annual fees charged to customers; number three, the late payment fees; and number four, other fees in relation to foreign exchanges. But the biggest contributor really is the combination of the annual fees charged to our client.

[indiscernible] At least the late payment fee may see some kind of a decline in 2022, right? Because if customer cash flows are actually improving and they're a little more prompt in payments, at least, that stream of income should slow down, right? Is that a reasonable assumption to make?

Yes, it's a reasonable assumption, but that will be made up for by increasing interchange from the merchants as we see increased mobility and the ability of customers to spend more in face-to-face transactions at point-of-sale as well as their resumption, although gradual, of travel across domestically and maybe by the end of the year internationally.

Excellent. This is really helpful. My last question is one of the charts that TG put up on the digital one, which is the retail cost-to-income ratio where the non-digital cost-to-income ratio for retail went up from 43% in 2020 to 58% in 2021. How much of that is on account of all-in revenue? And how much of that was just on account of the increase in costs?

I think when you look at the revenue, it's mostly from the fall in revenue because the revenue for 2021 for a non-digital customer is lower than the per capita revenue of a non-digital customer the [ previous year ].

Okay. Because the decline is almost 30% to 35%. If I assume entirely in some account of decline in revenue, that's almost 35%, which looks like a steep decline. Is that more or less right, ballpark?

I don't know the real details. But I think it's because, right, we have a lot of fixed costs on the non-digital customer, right? So a small fall -- a revenue fall in the non-digital customer translates to a bigger increase in cost-to-income because you've got a fixed cost.

Okay. Thanks, Karthik. Our next question is in the chat box from Samin Reza of Maple-Brown Abbott. His question is, what's the non-interest income growth in 2022? I understand that 20% is expected in 2023. And also, which segments are you expecting to drive noninterest income?

We're really -- the 20% estimate is really for 2022, and we're seeing significant growth from across all businesses. The big contributors being our credit card business, our asset management business and also continuing increase in branch services and branch service fees. Those are the big contributors. We also expect significant increases in fees coming in transaction banking, combination of cash management as well as trade and supply chain. We do expect an increase in digital fees from open banking partner fees and the remittance. It's really all across. And on average, we're looking at around 20% for 2022.

And for '23, it's also about 20%.

Yes, we're looking at a 5-year CAGR increase and same growth target.

Any other questions? I don't have any more questions in the chat box and none in queue. Okay, we have another 1 from Robert of Citi.

Sorry, I just have a couple of follow-ups for me as there's no more questions elsewhere. Just coming back to the cost side, is there not some upside to costs from the integration of BPI and BPI Family Bank? Surely, that would be a positive cost this year because of the branch reduction and the headcount reduction you've already mentioned. And could you also specify what the further branch reduction or co-location will be for the next couple of years? I just want to understand where the steady state is for the branch count. I think you said 814 or so at the end of last year. So what were the steady state will be? And then just on the cost increase from investments in 2022, I would have thought a lot of this will be capitalized and amortized, but it sounds like it's just hitting the P&L directly. So if you could clarify on that, that would be helpful.

Yes, Robert. So for cost reduction from Family Bank, so when we looked at this integration, a big benefit is really coming from revenue synergies. But sure, there are costs synergies arising from integration of various operations units. And related to that also, on your question on branches, we do expect closing of probably around 70 branches in 2022. And a large part of that is because of the integration of the Family Bank. So there will be corresponding reduction in headcount, therefore, reduction in manpower costs. And the long-term plan for our branch transformation is really to bring down our branch footprint to around 6 -- from, what, 814 in 2021 to around 600 roughly by the end of 2026. And really, that whole digital transformation that we are seeing and the shift in transactions to digital and increasing digital propensity will really allow us to aggressively execute on that plan. So that will have a corresponding reduction in branch premises expense and also, of course, in terms of manpower. But I think importantly, when we look at the Family Bank integration, it's really the big contribution coming from revenue synergies as we offer better customer experience under a one BPI brand and also be able to cross-sell better under one BPI brand. All of these revenue synergies will more than offset whatever increase in cost of funds that we might see because of the higher reserve requirement ratio for the main bank compared to the thrift bank. And even that is also offset by the fact that the time deposit costs for the main bank is actually lower compared to the thrift bank. In terms of cost, what we have done over the past several years is really shift our whole technology cost model to a more variable and less, what we call, for instance, expenses that we amortize over a period of between 2 years and 5 years. So a lot of that has changed quite significantly. Just the projects that we approved in 2021 alone, approximately 95% is considered OpEx and only 5% is capitalized. And this is really the current wave the whole industry has evolved more on a pay-per-use model. So a lot of it really goes into P&L almost immediately.

Thanks, Robert. Before I get to Samin, there is a question in the chat box from Cristina Ulang. Given the ESG lending team and more mining-friendly regulatory climate in the country and higher commodity prices, what's the lending policy toward mining? Can you give some color also on your lending appetite for tourism and aviation?

Go ahead, Tere. And then maybe John-C is around so he can talk about it.

Well, long term, we're really looking at aggressively pushing for what we call sustainable finance. And we have big targets in terms of increasing our sustainable finance portfolio by the end of 2026. Sure, we have an existing policy on coal development. We are -- we don't have a specific policy excluding mining from our lending book. And we are looking at it also more holistically in terms of the economic contribution of the sector, this whole mining sector. And also not all, I guess, mining companies are created equally. We have seen in the recent regulations released by the government when it opened up mining such that there are really stringent requirements to ensure responsible mining. So that's speaking in general. So the quick answer is, in terms of lending to mining sector, it doesn't necessarily go in opposite, I guess, philosophy to our ESG thrust in the medium term. Appetite for lending, tourism and aviation, I think maybe John-C, our Head of Corporate Banking can respond to that.

Thanks, Tere. Thank you for that question. Yes, so for mining, I have nothing much to add except maybe that there will be certain subsectors in mining that could be very interesting. And then that's probably largely dependent also on what the policy of the new government, Philippine government going forward. Having said that, as Tere said, we were flexible there, and we will maintain the ESG standards that we have set out to do. From a management standpoint, we are beginning to look at KRAs with our teams that are ESG-focused. So that's on the first item. On tourism and aviation, to be very clear, we have retained our approach of taking a look at businesses that have really dealt with a setback due to the pandemic. And what I mean by that is that pre-pandemic, we know who have been able to do well, who have been able to operate profitably, and therefore, are just challenged by the situation and are great operators. So we've continued to actually lend in the aviation sector and the tourism sector. It is classified. Those are still classified in our internal rating as high-risk sectors. But when mobility goes up, we are watching those 2 areas closely because they will require funding to accommodate increased activities as we get mobilized and [indiscernible] increases. Thank you.

Chinky?

Why don't we go to Samin? Chinky have dropped.

In terms of longer term, do you expect the loan composition to change from primarily corporate? So that's what, roughly 75% of the book? What's your outlook on the loan composition moving forward for this year and maybe the next few years as well, just roughly?

We're looking to add 1 point per year. So right now, just, Samin, in terms of our loan mix, we have about 75% corporate and 25% consumer and SME. As we position for higher growth in consumer and SME, we do expect naturally this 25% to go up gradually over the medium term. But it's very difficult to just add 1 point increase even if we grow at the 20% or 25% CAGR for consumer. Our long-term target is to bring this up from consumer and SME from 25% to 30%. And that already assumes aggressive growth in consumer and SME and still showing a growth that is above industry for our corporate book.

Has Chinky come back?

We've really lost her.

I think that's about it. I just want to -- if there are no more questions, I just want to thank everyone on the call. I hope you found it useful. And at BPI, my colleagues and I remain very committed to try to pursue our digital strategy while not forgetting, obviously, the core intermediation business. Our belief is strong that we want to acquire more customers going forward. DA had a question about customers. It's really about our new thesis is that whereas before, it was about deepening customers and looking at customers' profitability based on their deposits. We will continue to deepen, but customers based on transaction volumes and fees are just as profitable as customers who have deposits. So we are looking to acquire more customers. And therefore, our digital strategy is based on that. I think that's everything I wanted to say. And I just want to thank everyone for being on this call. Thank you, guys. Happy New Year.

Thank you.

Thank you.