Okay. Good afternoon, ladies and gentlemen. Welcome to our earnings call to discuss BPI's Results for the Second Quarter and First Half of 2025. I am Haj Narvaez, your moderator for this session. We are conducting this briefing in a hybrid manner with BPI speakers and panelists here in our headquarters at Tower 2 of Ayala Triangle Gardens, Makati City, while the rest of our participants are dialing in remotely. I am pleased to introduce you to our speakers and panelists this afternoon, starting off with TG Limcaoco, our President and CEO, Eric Luchangco, CFO and CSO. They will be joined in the panel for the Q&A by Tere Marcial, Head of BPI Wealth, Jenny Lacerna, Head of Mass Retail Products, Rico Peña, Head of Enterprise Services, Dino Gasmen, Treasurer and Head of Global Markets, Luis Cruz, Head of Institutional Banking, Boy Isarte, Head of the Payments Council.
Ginbe e Go, Head of Consumer Banking, is unable to join us today, but representing her are Dennis Fronda and Dex Quahotor. We are also joined by the rest of the BPI leadership team in this call. This afternoon's agenda will begin with opening remarks from our President, TG Limcaoco, followed by our CFO and CSO, Eric Luchangco, who will walk you through the second quarter and first half performance highlights, digital and sustainability updates. The floor will then be open to questions from the audience. Please note that this call is being recorded and legal disclaimers apply. Let me now turn you over to TG for his opening remarks.
Thanks very much, Haj. Good afternoon to everyone joining us on the call today. Welcome again. Thank you for persevering through the rain for those of you who are here physically. Last week, we disclosed that our first half earnings was PHP 32.96 billion. That's up about 7% versus last year. This was driven by continued growth in our revenues, which is up over 14%, which in turn was driven by strong Net Interest Income, marked by strong loan growth, as well as continued expansion in our Net Interest Margin. Our operating expenses also grew slower than revenues. That's contributing to positive jaws. OpEx, while still at 11% growth, was slower than last year's growth rate. Our provision expense jumped in the quarter and also jumped versus comparable periods last year. Our NPL ratios remain stable.
Our increased provisioning reflect our continued shift in our loan portfolio, where non-institutional loans now comprise 29.7% of our total portfolio. Our increased provisioning also was due to increased write-offs this year compared to last year. My colleague, our CFO, Eric, will provide you all the details as usual, including granular details on NPL cover by product. After the financials, allow us to give you more details on our progress on our digital initiatives, our agency banking, as well as other sustainable initiatives. Myself and my colleagues will be here to answer questions after Eric's presentation. Then we will close after that. Thank you very much. Now allow me to present Eric. Eric?
Yeah, good afternoon, and thank you, TG, for that opening. I'll now report on the performance of BPI in the second quarter and first half of this year. We're happy to report another solid quarter performance, supported by volume growth in all of our businesses. On the profitability side, we are pleased to report that we have delivered a record PHP 32.96 billion in after-tax earnings for this first six months of the year, driven by strong revenues and positive operating leverage. Profitability remains strong with analyzed ROE at 14.9% and ROA at 2.01%. On the balance sheet, loans and deposits continue to grow ahead of industry at 14.1% and 6.5%, respectively. The capital position moderated slightly from last year, reflecting a strong pace of loan growth and capital distribution. However, it remained robust with indicative CET1 at 14.47% and CAR at 14.27%.
On asset quality, NPL ratios stood at 2.25%, flat versus last quarter and up 5 basis points from last year, driven by a focus on the high-yield segment. NPL cover slid to 97.05%. However, we believe that asset quality remains solid and that the ratios remain favorable versus industry averages and within our risk appetite. On growth, we continue to expand the client base to over 17 million customers and increased our level of client engagement. On our new initiatives, agency banking is emerging as the high-growth opportunity we've always expected it to be. While teachers loans and motorcycle loans are on the rise, signaling strong growth potential. For the first half of the year, we delivered a net income of PHP 32.96 billion, an 8% improvement year-on-year, driven by strong revenue growth that offset higher operating expenses and provision.
Net interest income rose 16.2% year-on-year to PHP 71.2 billion, supported by excuse me, robust loan growth and a continued improvement in NIM. Trading income increased 115% to PHP 1.6 billion, set against a backdrop of a volatile trading environment and the declining trend in interest rates. Forex income fell 42.7% due to tighter market spreads, despite higher volumes and an expanded client base. Fee income was up 9.1% to PHP 18.5 billion, reflecting transaction activity growth. Total revenues at PHP 92.55 billion was up 14% year-on-year, while operating expenses increased 11.7% to PHP 42.75 billion, driven by business volume growth, investments in technology, manpower benefits, and marketing initiatives. Pre-provision operating profit rose 16.1% to PHP 49.8 billion. Provisions, however, showed a marked increase to PHP 7.25 billion. We'll delve into this a little more later.
All of these contributed to the net income of PHP 32.96 billion for the period.
On the sequential quarter, total revenue reached PHP 47.85 billion, up 7%, driven by sustained loan growth of 3.1% and an 18 basis point increase in NIM. Non-interest income rose by 8%, primarily driven by strong growth in trading income. Fee income remained largely flat as gains in wealth management and insurance was offset by the impact of five fewer banking days in the second quarter, a reduction in IBFT fees from PHP 25 to PHP 10. Additionally, a significant property sale gain booked in the first quarter contributed to a high base. OpEx growth was exceptionally high as it was impacted by delayed booking of prior period expenses, mostly for branch transformation projects, utilities, and advertising expenses. On shareholder returns, we sustained strong profitability with an annualized Return on Assets of 2.01% and a Return on Equity of 14.93% in the first semester of this year.
Earnings Per Share reached PHP 6.24 per share, marking a 7.6% improvement year-on-year. Our performance continues to support strong and consistent dividend growth. For the first half of the year, the bank declared a dividend of PHP 2.08 per share, up 5.1% from the same period last year and up 131% from the fixed dividend amount paid until 2021. This corresponds to a payout ratio of 17.71% of 2024's full year net income. Moving on to the balance sheet, total resources reached PHP 3.4 trillion, reflecting a 9.3% year-on-year increase. Loans expanded to PHP 2.4 trillion, up 14.1% year-on-year and 3.1% quarter-on-quarter, driven by sustained credit demand. Deposits grew 6.5% year-on-year, largely from growth in CDs. Loan and deposit growth outpaced industry averages, contributing to continued gains in market share. The CASA ratio remained fairly stable at 62.5%.
Managed deposit growth combined with a 450 basis points cumulative reduction in the reserve requirement ratio since October of last year supported a continued improvement in the loan-to-deposit ratio, which rose to 91%, up 6 percentage points from last year and 1.5 percentage points from the previous quarter. Liquidity ratios remain robust, with indicative LCR at 171% and NSFR at 135%. Loan growth continues to show positive momentum. Maintaining the trend of previous quarters, NIM improved further, rising 4.67% in the second quarter, up 18 basis points quarter-on-quarter and 34 basis points year-on-year. This was driven by the rise in asset yields alongside a moderate drop in funding costs. We attribute the sustained improvement in NIMs to several factors. Shift in the loan mix towards high yielding segments, which now account for 29.7%, up from 26.7% just a year ago.
Improvement in the loan-to-deposit ratio, which rose to 91%, up 149 basis points from the previous quarter. Managed deposit growth, which complemented the use of borrowings as an alternative source. This trend reflects the bank's focus on optimizing its asset mix and funding structure to support profitability. Gross loans increased by PHP 292.7 billion or 14.1% year-on-year. Non-institutional loans accounted for PHP 149.6 billion of this growth. Rising 27% year-on-year and contributing 51% of the increase in total loans. The growth in non-institutional loans was led by business bank loans up 71%, personal loans up 36%, auto loans up 30%, credit card loans up 29%, microfinance loans up 21%, and mortgage loans up 19%. These reflect robust growth momentum, especially considering the high base following successive years of strong growth.
On funding, we continue to shift from time deposits to bond issuances, taking advantage of incentives for green and sustainable financing under BSP Circular 1185. In June, the bank issued the BPI SINAG Bonds, a 1.5-year peso-denominated fixed-rate bond amounting to PHP 40 billion. These bonds carry a 5.85% annual interest rate, paid quarterly and are listed on the PDEx. Deposits remain our primary source of funding, borrowed funds grew at a significantly faster pace at 54.8% quarter-on-quarter and 50.2% year-on-year. As a result, the share of bonds and bilateral loans increased to 8% of total funding, up from 6% last year. The loan-to-deposit ratio continued to improve at 90.9%. Loan-to-total funding ratio declined to 83.7% due to the additional bonds issued last month, which further expanded the funding base beyond deposits.
We remain focused on strengthening our deposit base, particularly in CASA. Our deposit mix is still predominantly retail, compromising 70% of total deposits, or 73% if we include the SME deposits as well. NPL levels stood at PHP 53.08 billion, up PHP 1 billion from the previous quarter and PHP 7.5 billion year-on-year. The NPL ratio was stable quarter-on-quarter at 2.25%, but up 5 basis points from last year. Credit write-offs for the quarter totaled PHP 3.2 billion. While the increase in NPLs, the NPL ratio and credit write-offs is largely attributed to the expansion of the high-yielding loan segments, which while offering better returns, inherently carry higher credit risk profile. NPL cover declined by 3 percentage points from the previous quarter to 97.05%, but remains adequate when viewed per loan product, which we'll detail in the following slide.
In addition, under BSP Circular 941, which includes surplus reserves allocated for general loan loss provisions, the coverage ratio improves to 23.8%, providing a strong buffer. For historical context, the bank's lowest NPL coverage ratio was in 2006 when it dipped to 47%. Looking at NPL per product, we continue to monitor NPL trends closely across our lending portfolio with a focus on segment-specific coverage strategies. Microfinance posted the highest increase in NPL ratio quarter-on-quarter and year-on-year, primarily driven by negotiable loans. Business banking saw a quarter-on-quarter increase of 130 basis points and a year-on-year increase of 202 basis points due to the new NPL formation from client base expansion test programs and a few accounts from the high ticket segment, where a small number of defaults had a higher impact on overall NPL. Credit cards showed improvements supported by provisions.
Auto NPL ratio is flat, and mortgage showed just a slight movement of 7 basis points during the quarter. As shared in our previous briefing, we assessed the NPL coverage per segment, factoring in risk levels and collateralization. Credit card NPL are unsecured, requiring higher NPL coverage due to the lower recovery prospects. Conversely, owing to their secured nature and lower LGD historically, mortgage and auto loans require lower NPL coverage. Microfinance and Business banking remain the books that we are watching more closely. For now, we believe, supported by our ECL models, that we are okay with these books. At our largest book by size, Institutional Banking, we maintain a coverage ratio of 127%, even though nearly 30% of these loans are secured by collateral, reflecting a conservative approach to provisioning even as this book has tightened up its asset quality slightly.
There have been some concerns expressed on the recent rise in provisioning levels. We're sharing with you some data to help frame why we believe we continue to be on the right track. On the upper table of this slide, you'll see the ending NPL levels for each of the years from 2017 to our current first half results, along with the NPL cover at the end of each year and the loan loss reserves, which is basically the NPL level times the NPL cover. Below that, we show the actual provision expense taken in each year. On the last line, the additional provision that would be required to bring the NPL cover back up to the starting level of the period.
In the case of the period from 2018 to 2024, the additional NPL cover we would need to get back to is 129.2%, and we would do that by adding another PHP 11.1 billion of provisions. On the lower table, you will see the average annual provision taken across 2018 to 2024, including a top-up provision to bring the NPL cover back up to the 129%. The total of those provisions is PHP 82.5 billion, which when divided over the seven years, averages out to be PHP 11.8 billion in provisions per year. You see that on the first line of the lower table. If you divide that over the average loan balance of PHP 1.5 trillion over the same period, 2018 to 2024, you would get an adjusted credit cost of 79 basis points per year.
Meanwhile, if you annualize the PHP seven and a quarter billion of provisions taken this year, and also topping it up to the same NPL cover that it was at the start of the year, we would come out with a credit cost for the year of 87 basis points. This is just 8 basis points more than the average credit cost over the period 2018 to 2024. Meanwhile, our average NIM from 2018 to 2024 was 3.6%, versus our 2025 average NIM of 4.58%, or an improvement of 98 basis points. You see that 98 basis points of additional NIM is what we see as paying for the additional credit cost generated by the shift in the book. I hope this helps clarify why we are still very comfortable with the choices that we've been making in terms of the loan mix of the book.
Moving on to fee income. Fee income reached PHP 9.29 billion, up 0.4% quarter-on-quarter and 3.4% year-on-year, with strong contributions from our core businesses. Credit cards was up 16.6% year-on-year, supported by a robust 19% growth in retail billings and a 9% increase in the card base. Wealth management was up 7% year-on-year, with an all-time high level of AUM hitting a record PHP 1.76 trillion, a 22.5% year-on-year and 3.2% quarter-on-quarter jump, driven primarily by client fund inflows and to a lesser extent, by asset valuation gains. Insurance was up 12.3% year-on-year, fueled by equity income from investments in BPI/MS and AIA, as well as branch commissions from new insurance product sales.
The above increases were partly offset by retail loans, which was down 22%, primarily due to the collection of housing loan penalties in the prior year. Service charges down 3%, primarily due to a shift in transactions from the branch to online channels, as well as a reduction in the number of banking days this year compared to last year. Remittance was down 1.8% due to lower transaction count amid increasing competition. Moving on to operating expenses. OpEx grew by 10.6% quarter-on-quarter and 10.8% year-on-year, primarily driven by premises and other expenses, which includes marketing costs, rewards, and volume-related expenses. Technology expenses grew by 8% year-on-year, driven by ongoing investments in digitalization and some volume-related expenses. Notably, this growth is moderated compared to previous years. Premises expenses rose by 12.1%, driven by higher costs for utilities, security service expense, and rent.
Other expenses up 21.5% included marketing costs, largely attributable to consumer-focused campaigns and rewards programs. The increase also reflects business volume-related expenses and third-party fees, with marketing costs accounting for a significant portion of the increase. Despite the increase in expenses, these investments delivered substantial value to the bank. We added 1 million new customers since the start of the year, bringing our total customer count to 17.1 million. We achieved strong volume growth across the bank and gained market share in key areas including loans, deposits, credit cards, and wealth management, and other significant businesses. We achieved operational efficiencies with cost-income ratio continuing its steady decline since 2021, adjusting for the one-off gain in 2022.
We don't expect first half CIR of 46.2 to hold out to the end of the year, as expenses for us typically accelerate into year-end, it's currently trending about a percentage point tighter than it was last year. Moving on to capital. Our CET1 capital was at PHP 383 billion, up PHP 31.5 billion from last year and PHP 1.3 billion from the previous quarter, driven by income accretion despite a dividend payout in June. CET1 stood at 14.47% and CAR at 15.27%. While both slightly declined quarter-on-quarter, they remain well above regulatory and internal thresholds, providing ample capacity to support continued loan growth. Now we'll move on to some non-financial updates. On establishing ourselves as a leader in digital banking, we continue to maintain our seven client engagement platforms.
Starting from the left, we have BPI VYBE, our e-wallet, currently has 2.04 million signups, with 86% being VIPRO users, allowing them higher usage limits. We continue to grow its existing functions, such as CenterBank and eWallet, add new billers from various categories, and seek more cash in via partner merchants and kiosks. BPI app, our main operating app for retail clients, enhanced existing services from bills payment and provided a more secured experience by web login via QR scan and cash withdrawal available on Mobile Key devices only. It also expanded the reach of core features such as new-to-product digital account opening for credit card-only clients, mobile check deposits for selected client segments, and in-app FX conversion. Next is BPI Trade app, for clients who invest in equities, with a significant increase in overall app usage.
Unique logins declined due to online account cleanup, it has improved since last quarter. Account opening has gone up with the help of e-registration, and we're seeing the payoff from last year's migration as users adapted to the new platform. The BanKo app is for microfinance clients with the relaunch of InstaCash Ko Loan, featuring better approval rates through enhanced credit risk parameters straight from the app and enhanced user experience. The BPI BizKo app for SMEs with a growing user base now at 26,900 users. Further expansion by e-payroll to bring additional new-to-platform users and drive usage. BPI BizLink app is for corporate clients. We launched the express check deposit facility and continued efforts to migrate clients to the mobile app.
Finally, we have BPI Wealth Online for high net worth individuals, which has over 2,500 active users, which has seen a 66% increase from last year. We continue to grow existing functions, increase capabilities in open banking, and improve the UI/UX. We now have 120 API partners, up from 74 in 2019, and offer more than 17,000 brands from only 749 in 2019. We are pleased to share that two years after the launch of agency banking, we are now seeing significant opportunities for growth be realized. 2023, our focus was on building strong partnerships and establishing partner stores. By 2024, we shifted towards activating these stores, enabling product sales and launching transactions to drive client engagement. Today, we have partnered with 25 brands, resulting in nearly 6,600 partner stores where customers can apply for BPI products.
Among these, 300 stores are now capable of processing deposits and withdrawals, or what we abbreviate as DEWI. With the tablet and app provided by the bank to its partner stores, customers can enjoy seamless service free of charge. Notably, several of these 300 DEWI-enabled stores are 64 standalone locations, meaning they operate in areas without a nearby BPI branch. One standout example is our DEWI store in Siargao Island, which serves a community with no existing bank presence at all, which is a milestone in financial inclusion. We are seeing transformative results that validate our strategy of expanding reach through agency banking. For the second quarter, we sold 123,600 products through this channel. Over 14 times the number of products sold in the first quarter of 2024. The majority of these are insurance and deposits products.
Deposit and withdrawal transactions surged from 2,800 in quarter three 2024 to 34,700 in quarter two 2025, a more than twelvefold increase. These are transactions that would traditionally occur in a branch, but are now being fulfilled through partner stores. The right chart shows the engagement level at the stores, with each store recording an average improvement in product sales as their personnel become more adept at promoting the BPI offering. Each store now sells an average of 14 products per quarter, nearly nine times compared to quarter one 2024. Moving to motorcycle loans. We acquired this product as a result of the merger, and it has since evolved into a high potential growth business. Since the merger in January 2024, motorcycle loan balances have grown to PHP 4.86 billion, reflecting a 25.6% increase year-on-year and 28.9% rise since the merger.
Prior to the merger, loan releases were on a downward trend. However, they rebounded strongly in 2024, increasing 24%, and continued to gain traction in the first half of 2025, with a 38% increase compared to the same period the prior year, highlighting both rising demand and operational scalability. Our market share now stands at 13.4%, up 265 basis points year-on-year and 208 basis points since the merger. Meanwhile, our client base has grown by 10%, reaching nearly 103,000 clients. The strong trajectory underscores a long runway ahead for this product as we continue to combine BPI scale with the business that was acquired through Robinsons Bank. We saw similar gains in teachers loans following the merger with Robinsons Bank, and Legazpi Savings Bank, which is now a subsidiary of BPI. We have seen strong momentum in its teachers loans portfolio, demonstrating early synergies from the integration.
The portfolio size and customer base has nearly doubled as a result of more teacher-friendly loan processes, enhanced engagements with the Department of Education, and increased channel productivity coming from the combined strength of BPI and BanKo's branch networks. Outstanding loans reached PHP 12.2 billion, marking a 72% increase year-on-year and a 105% jump since the merger in January 2024. Loan releases also rose significantly with figures up 109% versus releases last year and up 154% compared to first half of 2023. Among DepEd personnel, we helped 3% of them prior to the merger, which has since grown to 6.3%, reflecting the impact of our expanded distribution. We now serve approximately 56,265 teachers, a 79% increase year-on-year, and 108% growth since the merger. Growth is expected to accelerate further in the second half of 2025 as we fully leverage our branch network and digital capabilities for distribution.
A little on the sustainability side. BPI further boosted its sustainability achievements in the second quarter of 2025 on utilizing renewable energy in BPI's operation. BPI is the first Philippine bank to adopt a retail aggregation program, transitioning 70 of our branches to renewable energy in partnership with ACEN Renewable Energy Solutions. Additionally, BPI shifted our Binondo corporate office in Manila to renewable energy using the Green Energy Option program. BPI issued a PHP 40 billion sustainability bond, proceeds of which are allocated to projects with clear environmental and social benefits. BPI Capital acted as a joint lead arranger and selling agent of the transaction. BPI Capital also arranged a sustainability-focused deal for Giga Ace, partially financing the development and construction of Quezon North Wind Power with a 344.5 peak MW plan.
BPI partnered with The Corporate Treasurer for an exclusive event for treasurers and CFOs, influencing the integration of sustainability in their businesses, also sought to share its experience with others by participation in various events, including as the only Philippine private sector bank to speak at the Sasana Symposium organized by ADB, the World Bank, and the Asian Venture Philanthropy Network in Malaysia. BPI also celebrated the 2025 Sustainability Awareness Month themed Beyond Green, with over 14,000 participants from 24 events, both national and international, which showcased a wide range of ESG-driven activities of the bank, also conducted financial wellness sessions in Tanza, Cavite, in partnership with the Philippine Army to empower army reservists with financial literacy skills in budgeting, saving, investing, and planning for the future. BPI has also garnered a total of 20 ESG-focused awards in the last 12 months as of July 2025.
With five of these awards won by BPI Capital at Alpha Southeast Asia's 18th Annual ESG Green Finance Awards, four from Global Finance's Sustainable Finance Awards, three from Asian Banking and Finance Awards, two from International Business Magazine Awards, also two from the Asia Pacific Tambuli Awards, another two from FinanceAsia Awards, one from the ASES Awards, and one from The Asset Triple A Sustainable Finance Awards. Beyond just sustainability, we've also compiled a list of institutional awards and recognitions received by BPI from esteemed foreign and local award-giving bodies. Just last Friday, BPI was named the best bank in the Philippines in the Euromoney Awards for Excellence 2025. Recognized for its phenomenal fine growth, record-breaking financial performance, and several pioneering initiatives spanning sustainability and digital innovation. BPI also received the Best SME Bank award.
We're deeply honored by these and all the other accolades conferred on BPI. The awards can be seen on our website, I don't need to run through each one of them individually. Yet we still have a second page of these. Anyway, in summary, let me close with a few highlights. We delivered a strong first semester operating performance, reflecting solid execution through our business. Our balance sheet remains healthy with ample liquidity and strong capital levels. Asset quality continues to be robust with adequate allowance provisioning. We are now accelerating growth by unlocking synergies from the recent merger. Overall, we are very encouraged by our first semester results. Despite ongoing risks and uncertainties, we remain confident that the bank is well positioned to navigate the current environment, drive sustainable growth, and deliver long-term value to our shareholders.
Thank you. We will now open the floor to questions.
Thank you, Eric. Before we open the floor to your questions, please allow us a minute or two to set up at the venue. Just wanted to remind everyone, if you're joining us via Zoom, there are two functions at the bottom of the Zoom webinar screen, which you may use to queue. One is the raise hand function. The host will then prompt you and unmute your line for you to speak. Alternatively, you may type your questions in the Q&A box. We will read out your question on your behalf. For those on site, you may use any of the mics available on the floor. You may raise your hand and we will have someone hand the mic to you. Please also identify yourself by your name and company so we can address you accordingly.
For the benefit of everyone attending this call, whether in person or online, we'd also like to encourage you to ask your questions during this session. We will refrain from taking questions after we end this call. Now that we have set up, just wanted to introduce the panel. Joining us here in front with TG and Eric are the senior leaders of BPI. Tere Marcial, Head of BPI Wealth. Jenny Lacerna, Head of Mass Retail Products. Rico Pena is also here with us, Head of Enterprise Services. We also have Dino Gasmen, Treasurer and Head of Global Markets. Luis Cruz, Head of Institutional Banking. Boy Isarte, Head of the Payments Council. Joining us as well online is Ginbee Go, Head of Consumer Banking. Also representing Consumer Banking and here with us on-site are Dennis Fronda and Dex Quahotor.
To begin the Q&A portion, we can start with questions from those attending in person. Please go ahead.
Good afternoon to everyone. Liam from FIAP. I just have three questions for today. On the previous analyst briefing, you've said that your corporate clients, in light of the economic uncertainty, they are at a standstill. That's how you've explained it. Now, given the new developments and of course, a lot of things happened. We have had been imposed two versions of rate cuts already. I would just like to ask if you're already seeing a shift in sentiments towards your corporate clients o r perhaps are they seeing, or are they believing, or are they willing to set on the "added certainty," at least from everything that's happened? I share it. The second question is about remittance. I'd just like to request for colors on that 22% drop on the fee income. On the third is about your Artificial Intelligence.
I'd just like to ask, how much does it contribute to your operating expense, and what is your timeline for its full realization?
I'll talk then I'll pass it on to Luis on the sentiment of the corporate clients. I think there's still a lot of uncertainty. It's not so much about the domestic policies. I think there's a lot of uncertainty as to trade and the tariffs that the U.S. is imposing. Given many of our clients have relations not only with China but also the U.S., I think people are still waiting to see how that actually turns out for the last few week to give more color on the actual sentiment, given that corporate loans grew over 9% year-over-year. There seems to be some positivity there. Your second question is remittances. I really have no color why it would drop. You know, remittance fees. The volume's down. Okay. The last one is Artificial Intelligence.
I think there's a lot of talk about Artificial Intelligence. I think we're just beginning to play into it. There's a lot of ideas. We do have one AI, I assume when you talk about AI, you talk like everyone else, talk about generative AI. We've had AI and machine learning, pattern recognition for many years, including for fraud detection. When you talk about generative AI, we do have one that is in production now, but it's used internally. It is an internal tool where we've uploaded all our policies, procedures, and products into a system that's based off OpenAI, where any of the officers and staff, particularly on the branches, can query and with natural language and get the answers with consistency across. That's our first foray into it. That costs a trivial amount to develop. We are looking at other AI tools, particularly on the credit side, but nothing going on yet.
We're also looking in terms of customer service, but we haven't implemented any yet. Luis will give you color on what the corporate side is all.
Thank you, TG. The first question regarding the corporate sentiments. Start of the year, yes, there was some uncertainty. As you see, after the two quarters, the first half, the growth on the corporate side was almost double digit at 9.5%. I think what drove the growth really was coming from the project financing and acquisition, and we saw a lot of those during the first half. In fact, we also see some in the pipelines until end of the year. Most of them are coming from power infrastructure and some of the healthcare, and some consolidation on the logistics or the cold storage space. I guess the uncertainty really is on the working capital. Before we would see working capital utilization quite high or active. Right now we can see it at a level almost flat compared to last year.
I guess they're really managing their working capital versus whatever, or they maximize whatever cash they have in terms of cash flow. I guess that's where the uncertainty is happening. Overall, given the project finance, acquisition financing, and the opportunities that are brought about by, let's say, the green energy, the GEAPP opportunities, I guess that's where the growth is really coming. As a quick one on the remittance, the drop is basically in the volume given the uncertainty b ut I think you'll see some recovery towards second half.
I think also remittance is the way we book it because our remittance business is not only the traditional or what we know as the traditional remittance from other countries. Our remittance business is also partly responsible for the InstaPay and PESONet transactions, and we split the income with the digital services. Given that we cut our fee from PHP 25 to PHP 10, they bore half that cut also.
Thank you.
Thank you for your question. Wanted to check if there are any more questions from the audience. If none, I think we can take on some of the questions that came through Zoom. The first question is actually from Yong Hong Tan of Citi. He actually has a few questions, and I will go through them now. The first question is, ''With LDR benefiting from triple R cuts, do we still see more room to manage liability as the loan yields get repriced in subsequent periods? Second, where should NIM stabilize and how does that compare with your NIM sensitivity guidance? Third, he noted in his calculation that provisions hit about 74 basis points this quarter. Did this come in higher than expectations?'' He wanted some color on NPL formation this quarter. Fourth, there was a question about OpEx.
It has been well managed and growing at a slower pace versus past years. What drove the slowdown in OpEx growth and how sustainable is that?
Sorry, what was the third one?
The third question was about provisions. It's actually a two-part question, Eric. One is, did it come in higher than expectations? His calculation was that it came in about 74 basis points. He wanted to ask about the NPL formation.
Let me manage that. The first question is on our ability to manage our liabilities, I'll start off and let Dino talk about our funding strategy. Given that about 30% of our deposits are still time deposits, any cut in reserve requirements actually translates to lower funding costs on the time deposits. We're also making a very deliberate strategy not to compete aggressively for time deposits, as we have other alternatives for funding. You've seen us go to the bond market, where the bond market actually provides us more efficient means of funding given some of the regulatory arbitrages there. Also, we are turning to the swap markets a little more. I'll let Dino finish that statement.
That's right. Although our deposits are not growing as fast as it used to, what we're actually doing is substituting that by borrowings, inter-bank borrowings. That's composed of bond issuance, repos or clean borrowings from other banks. A lot of this is in foreign currency, majority is in foreign currency, so we swap it into PHP if that's what we need. What we're seeing is it's still cheaper than raising or competing in the market for time deposits. It's cheaper by up to 80 basis points on the average for us. I think we'll continue to rely on this, but not in such a way that when interbank liquidity closes, we find ourselves having problems. Right now, it's only a small portion of our liabilities, and we want to keep it that way. Not rely too much. Yep, thanks.
I'll take the question on the NIMs. Basically, with the question on where would NIM stabilize, I think what I can share with you is, if you look at our presentation, we showed that NIMs in the first quarter was 4.49. NIMs in the second quarter was 4.67. So right there you'll see that was an 18 basis point increase in the NIMs, of which about 3 to 4 basis points can be attributed to the triple R cut. A lot of the NIM improvement continues to come from the way the book is moving through time, right? And where that's going to stabilize is going to depend on where we continue to move the book. At this time, we continue to see there being enough room for continued growth in the consumer book.
I believe that's going to extend even beyond the end of the year, right? There is no point that we're looking at now that the NIMs are going to basically flatten out, right? That's based on where directionally the loan mix is heading. There was also a question on the provisions and whether it was within expectations, right? I think it was, again, there was some formation that was a little unexpected, but it was not out of range, right? I wouldn't say it was really out of what was expected, right? If you look at the book as a whole, right? In certain areas, maybe a little more than what we're expecting and t hat's why you'll see the business bank and microfinance books.
We saw a little more formation there than we were expecting, but kind of balanced by actually on credit cards, actually some improvement on that side. Therefore, kind of balancing out. Finally, there was a question on OpEx. I think on that, a lot of stuff that we've been doing to try and manage OpEx, trying to manage that. We've always said that we want to manage OpEx in the context of our revenue growth, right? We're always going to try and shoot for that foster job. We're going to always try to be keeping OpEx well managed so that it stays within our revenue growth.
I think to add some color, when you also look at the NPL formation in the quarter, our NPL grew about PHP 1.3 billion in total amount, right? We wrote off about PHP 3.3 billion. That's about PHP 4.3 billion, PHP 4.4 billion that you would normally add to provision. Was it a surprise? Not really. It's just the way we operate the book. I think you have to understand that when you provision, it's both a science and an art. When you look at the ECL model, which is a completely independent group, the risk group basically put a tag on about, if I recall right, about PHP 400 million, PHP 500 million of additional risk due to change in macroeconomic variables. All of that we have to account for when we do our provisioning.
Thank you for the question. We'll shift gears a bit. There is a question from J.J. Chua. Given recent regulatory and public scrutiny of online gambling in the Philippines, how much of an issue do you think online gambling is to your loan book?
Maybe if there is any exposure to people who are using the loan proceeds for online gambling, that is probably in what I would call the unsecured. I will ask Jenny to take a crack at that. Jenny?
Thank you, TG. There are concerns on online gambling. We actually hear it from the media, and we hear it, and we read it. For BPI, our exposure to online gambling, as we see it in the unsecured loans, would be very minimal, if ever. For example, for our loan products like credit cards where you have SIP or installment loans or cash advance, the composition of that in our portfolio is very, very little. In fact, for cash advance, the average loan size is just about PHP 10,000. For our SIP loans, it is about PHP 70,000. If you look at the total portfolio, the contribution is 0.3% and 0.7%. Largely the SIP loans, which is probably the bigger size in the credit card books, is actually a targeted campaign that cannot be availed by anyone.
We score the clients, we pre-select the clients, and we offer the clients that have good credit standing and offer the cash option as a way to be able to address other financial needs like home renovation. It can also be travel or education. It is a targeted campaign that is not open to anyone. While we watch it, we know that we have the guardrails to be able to determine that it is only being granted to customers who are responsible in using those products. Thank you.
You said the cash advance, which is the one on the card.
Is only 0.3%.
Of your total books.
Of the total transactions. The average loan size is only PHP 10,000. Yeah. Thank you.
We'll now go ahead with some of the questions from Zoom. We'll start of with the question, Dave Tan of JP Morgan has raised his hand, I'll now turn it over to Dave. Dave, please go ahead and ask your question. Thank you.
Hi. Good afternoon. Can you guys hear me?
Yes.
All right, great. A couple of questions from me. First, on loans, could you just update us on loan growth outlook across segments? Number two on loans, in terms of loan mix, it looks like we've reached your initial target of around 30% non-institutional. Do you have further targets from here where this segment or this part of the book will grow, and by when?
You're never happy when we hit our targets, David. What was the first question?
Loan growth outlook.
Oh, loan growth outlook. I think we're still looking at the 12% to 14%, and I know you're probably saying that's what we're always saying. I think we're looking at corporate, maybe 10% to 12%, and then we're looking at our consumer book, probably high teens to mid-20s, depending on the product. We're now at 29.7% of the portfolio is non-institutional. Our target that we set four years ago as a five-year target was 30%. As Eric said to me, we intend to continue to push that. Where that ends up, to be honest, we really haven't thought about it and to set a new target. We do know that we are putting a lot of emphasis on the consumer loan book and the non-institutional book, and we will continue to grow that as aggressively as we can without really sacrificing the growth of our institutional book.
Where it ends up really is dependent on how the two books grow, we would like to see it much higher. I don't really know if it's fair to set a target and say 35% in five years, which means you can meet the target by slowing down Luis book, right? We'll just try to grow both as aggressively as possible within the framework that we've always been, prudent lending, managed NPLs.
Okay, understood. In line with that, just to follow up on the credit cost side. Number one, do you have updated credit cost guidance for this year, given that it seems like we're trending a bit higher initially than what we're expecting? Two, as the book shifts more to this non-institutional segment, should we then expect further increases in credit costs? You did mention earlier, if you normalize for the coverage, it would have been closer to 90. Is that a level that you're also looking at?
That almost 90 included an adjustment to bring it back up to the level at the start of the year. We're not necessarily looking to bring it back up to the level at the start of the year. I think on an ongoing basis, we basically made provisions for seven and a quarter for the first half of the year. There's no indication on our part, at this time, that that provision level is incorrect, and that we will not be needing it. Therefore, in our minds, we are prepared to allocate as much as that in the second half of the year if the conditions warrant it. We'll continue to monitor. Again, we're driven by our ECL models as well as looking at the individual portfolios and trying to see whether they look like they're properly set up, right?
That they have adequate NPL cover, especially in the context of the collateral cover that we have. I think at this point, we remain quite comfortable with where we are, and our ECL models indicate that we have very sufficient cover, and therefore, there's no real need for us to bring this back up to a specific NPL cover.
All right. Thank you. Just to confirm, are you saying the first half run rate, more or less, that's what you're looking at for the year? I think Eric hinted at that. Two, I know you don't target NPL coverage, but last time you shared your coverage versus the ECL, how that ratio is. Could we just get an update on that as well? That's it from me. Thank you.
Dave, our policy internally is to really cover our ECL risk. That's the minimum provisioning that we should have, should cover the ECL risk. ECL is driven both by macroeconomic factors as well as the behavior of the book. As loans move into stage two or stage three, then we have to provision more. It really depends on those two factors. Right now, for the provisioning for the first half of the year at 7.25, we feel that that's the appropriate provision for the first half of the year. I don't see anything that should change my mind to say that that would be an inappropriate number for the second half of the year. Unless, of course, the macroeconomic variables change, then the model comes up with a different number.
One of the things that we have to be cognizant is that the ECL model is predictive and is forward-looking. So as we cover that, there could be a time going forward where the ECL model will predict a smaller number, right? For example, during the pandemic, the ECL model basically put out a number that said this is the expected credit loss in the middle of the pandemic, then two years later, dialed it back very significantly. Yet we continued to provide, but with a much smaller amount as we took that cover down. That's the way we will manage this book, because we don't think it's fair to investors to significantly over provide, sorry, more than what the ECL model says we should.
Okay, that's quite clear. Thank you very much.
Thank you for your question, Dave. We'll now hand it over. We have a question as well from Harsh Modi. Harsh, please go ahead and ask your question.
Hi. Thanks. Related question on risk-adjusted returns. Eric, you made a point earlier that the spreads have gone up and hence that extra spreads are paying for slightly higher provisions. Just wanted to understand that slightly better, that the spread increase, is it how much is due to rates? Also, you defined the benchmark, and how much is due to loan mix shift? Then I have a follow-up on that. Thank you.
Yeah. Okay. Thanks, Harsh. Actually, just to clarify, what I meant is that yields are going up. NIMS, if by spreads you mean loan yields over cost of funding, which is NIMS, then yes, that has been going up. Over the past few years, call it 2022 to 2024, I think the view was that, ''oh, NIMS are going up and it's all because of the rising interest rate environment.'' I think from our perspective, what we were seeing is, yes, we were seeing higher NIMS driven by the higher interest rate environment partly, but also partly because of this shift in the loan mix that we were doing.
Now that the environment is shifting back towards an easing environment and BPI's NIMS are still on an improving trend, I think it makes it clear that it wasn't purely driven only by the rising interest rate environment. The fact that our NIMS are still rising means that it was partly driven by the shift in the loan mix, right? That we continue to expect to be the case moving forward. That shift in the loan mix that is providing us with higher NIMS is also, by nature, leading to higher provisions, right? Because the consumer and SME books are books that have higher delinquency rates than our institutional book. What I was trying to illustrate is that the higher provisioning that we need to do, the higher delinquencies that we suffer are more than paid for by the higher NIMS that we're earning.
Therefore, it is a direction that we continue to move towards.
No, I get that. Thanks for that, Eric. I wanted to get a bit more granular into that. Let me spell it out. How much have the credit spreads gone up, if I may? Maybe there are some parts of the book where you have moved towards more fixed or more floating. Let's say if it's a card book, which is more fixed, personal loan, probably more fixed, especially on the unsecured and so on and so forth. That we understand it a bit better on the part. The reason I'm asking you this question is if, let's say, we do end up getting a sharper rate cut over the next 12, 18 months, is there a probability that we end up getting lower NIMs while the credit costs continue to remain high? Thank you.
Maybe the way to look at it is, I'll first say that if you talk about corporate spreads, they have definitely tightened, right? Despite every rational thing that should be happening in this market, corporate spreads continue to shrink. That's due to competition of the large banks and the medium-sized banks trying to gain market share. If you look at our historical asset yields, it's under deck. In 2024, the second quarter, our asset yield was 641. Today, at the end of the second half of this year, the asset yield was 670. That's an increase of 29 basis points. That's primarily due to a change in the mix of the book. I can assure you that the yields on corporate loans have come off within that period as corporate spreads have tightened and as the BSP has tightened their policy rate.
Does that help?
Basically what I'm concluding here, thanks for that, TG. What I'm concluding here is it's not necessarily the spread that I should look at. It's probably the total yields is how I should look at. In case over, let's say, next 12, 18 months, we end up getting much sharper rate cuts, it is not necessary that my credit costs will also come down. All of that will probably flow through to my risk-adjusted NIMs will come on over the next couple of years if rate cuts are much higher.
If rate cuts are more aggressive, right? I think you should have better credit performance on the corporate side, but not necessarily on the consumer side, because rate cuts don't necessarily flow directly into the consumer book. It's very slow.
Sorry, if I may ask one more question on wealth management and flows. That's one part, if I'm looking at most of the regional banks, there's been a phenomenal flow into non-dollar currencies. That is feeding through to wealth management, asset management. Are you seeing similar trends, especially for the affluent and mass affluent kind of customers? Anything there which is worth highlighting? Thank you.
Harsh, the question is whether we're seeing additional flows into non-
Non-dollar.
Non-dollar. Not really. We grew our AUM year-on-year by about 23%, and that's almost driven by, I would say, more on the PHP assets rather than the foreign currency assets. Only about 15% of our AUM is in non-PHP denominated assets. If you are trying to get insights as to whether clients are diversifying away from US dollar on the issue of reserve currency shift, I don't think that is the trend that we are seeing. In fact, during weakness or a temporary strength of the PHP, even during that period, we're not seeing any significant shift.
Okay. Thanks. TG?
To be fair, I don't think we have a very big offering for non-dollar assets also. Yeah. We have a very small offering for non-dollar assets. Most of our clients either stick with us for dollars, and to be honest, I think if they wanted to do a sizable trade out of the dollar into another currency, they'd move away from us, because we just don't have that sizable offering. Yeah.
Thanks, TG.
Sorry, Harsh. I just wanted to add one comment on your comment on the impact of a lower interest rate environment in terms of consumer delinquency. I think the one area in that will be positively be impacted in terms of consumer debt performance in a lower interest rate environment, is it tends to drive growth, right? When you have better economic growth, then people as employees, as self-employed, they tend to do better as well, and those are the things that will drive our delinquency performance better.
Fair point. Thanks, Eric.
Thank you, Harsh. We have a question as well from Akash of UBS. Akash, please go ahead and ask your question.
Great. Thank you for taking my questions. The first one is just on the NPL cover. Eric and TG explained why NPL cover below 100% makes sense for BPI, and it did fall to 97% this quarter. I was just wondering, based on your latest assessment, is there a minimum level that you will not let this group level number go below? Is that 95%? Is that 90%?
Hard to say a number, because it depends on the portfolio. If our whole portfolio was mortgage, then certainly below 50% would be fine, right? It depends on the composition of our portfolio and the performance of those portfolios, and that's why we pay particular attention to present it to our investor. One of the slides that we have is each product, what is the portfolio, the size of the portfolio, the NPL of that portfolio, and the NPL cover for that portfolio. As general rules, we have about 120% cover for institutional banking, and I think that's really driven because we need 1% general loan loss reserves. We have about 140, 150% for credit cards. We should be at about 100% for Business Bank. We're slightly below that. We should be 100% for personal loan. We're slightly below that.
For mortgage, which is all secured by real estate, maybe we should be at 20%, 25%. For auto, which is secured by chattel, we should be about 60%, 70%. It really depends on the composition of our portfolio as well as the performance of that portfolio. That's why we present it to show that this is why we're comfortable with this kind of provisioning and the resulting overall level, 97%, that it ends up. The 97% is just a number that falls out. You should ask us if we're comfortable with 100% provisioning for institutional loans. We should ask Kokok if he's comfortable with provisioning at 80% for Business Bank. That's the way.
TG, that's actually the question, which is if you apply the optimal level to each of the sectors.
Yes.
What is the group level number that you come up with? Is that 90%? Is that 95%? Because if you're optimal at every sector, you wouldn't go below that group level number, right? That you arrive at.
Yes.
What is that number?
It would depend on the percent of that product. Right? Then the NPL level of that product. If we have 50% or let's say we had a portfolio that was only two, institutional corporate loans and call it mortgage. We were 50% in each, right? One had 100%, sorry, one had 0% NPL.
No, I understand. Sorry, I'm just thinking based on your current assessment, right? Current loan mix, current optimal levels for each sector, what is the group level number that you come up with? That's the question. Not hypothetical, but just.
We'll get back to you because I haven't thought about what the optimal level is. Basically, what I want auto loans to be as a percent of my portfolio.
No, no. Currently, auto loans are X% of your portfolio.
Yes.
This is the optimal level for auto loans. Institutional loans are X% of the portfolio. This is the optimal level for institutional loans. Aggregate it all up, what is the high group level number? Is it 90%, 85%?
Eric will get back to you on that. Sorry.
Okay, no worries. Thank you. But I think all said and done, it looks like you still have room to be able to optimize your provision level across the sectors, which means that this 97% can actually go further down, right? It will still be in the comfort zone for the group. Just related to that, I think if you look at the credit cost, right? Credit costs have gone up from 50 basis points to close to 75 this quarter, 74, 75. Your NPL formation is still running close to 1%. Again, 90 basis points to 100 basis points, right? The gap is being funded by the coverage. Your coverage has come down, and it probably can support a few more quarters of slightly higher credit costs. Just what happens after that?
Once you hit that optimum level on the coverage where you cannot go below that, what happens after that? Like credit costs, do we change our expectation of the credit cost for the group? Does it go to 90 instead of 50?
It doesn't really change our credit cost expectations. We believe at this point, If you look at, for example, 2023 and 2024, especially in 2023, the pace at which NPL cover was coming down was very fast, right? Intentionally so. Then in 2024, it's continued to come down, but starting 2025, you see it hasn't completely, this kind of NPL cover hasn't eroded completely, but it's really leveled off. I think what you're seeing now is that we're really quite close to being at a neutral level, where you would see it plateau. I think that's what you should probably expect, right?
Of course, this is all subject to changes in macroeconomic variables that are going to drive changes in the ECL model, that I think we're coming close to the plateau, which means that, I think that probably you're looking at these levels as kind of the ongoing levels, right? Probably.
Eric, then in that case, I think your credit costs will have to go up, right? Because last quarter also, the NPL formation was around 1%, and you said that's the kind of new normal NPL formation based on the mix change on the loan book, right? This again, this quarter, I think NPL formation was close to 1%, which is why credit cost went from 50 to 75. If your coverage stays where it is, the credit cost will have to catch up with NPL formation.
If you believe that NPL formation will continue in a sense as it is, then that is probably true. As I mentioned, in a declining interest rate environment, we expect it to be good for growth, and good for growth means that it should be good for performance of the borrowers, right? That especially applies in the case of corporate borrowers that actually get to enjoy lower interest rates, right? As the interest rate environment comes down, their cost of funding comes down, and we're already starting to see some situations where corporate borrowers are able to kind of see themselves in a less precarious situation.
What about the microfinance and business bank borrowers? Because I think that's the two sectors which are driving the NPL formation, right? High.
Their NPL formation is above normal. If you look at the size of the portfolio, that's not driving in and of itself. Microfinance cannot really drive anything in the overall book because it's such a small size, right?
I see. It's corporate that is the bulk of it, which you think will normalize as interest rates come off.
Yeah. more or less, that's what we see.
Okay. in that case, I think the 50 to 55 basis points of credit cost guidance for the full year is still intact, right? You don't see any reason to change that at this point?
Sorry. I'm not saying that.
Yeah. Sorry.
Just to clarify, I'm not saying that 55 basis points is where we expect to end the year. Again, in the second quarter, we're running a little above that. I think there's certainly room for credit costs to move upwards, and it will be dictated by our models. When you say, "Oh, you're already at earlier showed number close to 90 basis points," and if you think it's going to go much up from that, I don't think that is the realistic expectations where it'll go much above 90 basis points, but that doesn't mean that it's going to stay at kind of in the mid-50s either, right?
Okay. Great. Just a couple more questions and quick ones. Tax rate was a bit lower this quarter. Could you share some color on what drove that?
That's probably more timing than anything else. There are some tax timing issues, right? When it gets paid, and it doesn't get paid immediately and some movements. I wouldn't look for that to be an enduring change.
The full year should still be similar to last year then, right? Is that fair to say? For the full year.
I think, at this time, you should still be looking at something comparable to last year.
Great. Then just at the system level, when we look at the loan growth and look at the breakdown across sectors, one sector that really stands out is the electricity, gas, steam, and air con. That's the name of the sector, right? Which has seen like 25% sort of growth year-on-year. I'm just wondering, what is going on here? Is it renewables that is driving it, and is it more one-off in nature or is it sustainable? Any thoughts that you have?
Sorry, which loan are you referring to specifically?
I think he's talking about the power sector.
Okay. The power sector, that's where we're seeing a start of the year as mentioned. There are a lot of those, especially because of the JF that's coming in, and the liberalization of foreign ownership. I guess that's what's really driving the power sector. It's allowing foreign investors 100% ownership, except for the land. That's something that we see it's an opportunity for the growth of the power sector.
This would be a multi-year thematic in your mind o r is it a one-year thing?
We expect it to be like a medium-term thing, where demand should be strong for the power sector. Based on the pipeline and the opportunities that we're seeing, we're seeing it until this year and next year and maybe into the third year. That's what we're seeing on the growth sector.
It's renewable, right?
Yes, mostly renewable.
Okay, understood. Could you be able to share what percentage of BPI loan book is renewable energy?
Sorry. What I have is about a little over half of our power generation loans are in renewable energy versus non-renewable energy. In the power generation sector overall, maybe I need to get back to you with that number.
Akash we'll get back to you on that.
Fantastic. Thank you very much for taking my questions.
Thank you, Akash. We have a question that posted in the chat, from Rafa Garchitorena of Regis. He had a question about credit costs already, but I think we covered the discussion on credit cost outlook. I'll focus on the other questions. On credit swaps, '' How big is this becoming and what would be the impact on NIMs? What would be an adjusted NIM?'' Then I think there's a question on the new tax, which I think refers to CEMPA. ''What is the impact of the new tax on bank deposits as well as bonds with a maturity of five years and longer? What do you think, Dino?
I think, Rafa, when you say credit swaps, you're probably talking about FX swaps, no? Which we use to fund the balance sheet. It's small. For BPI, we only have about PHP 60 billion in swapped funds. That I think is less than 1% of NIM if you add back the cost of that swap. With regard to CEMPA, the biggest increase I think is on interest income on dollar or foreign currency time deposits, which has increased from 15% to 20%. That's a very small portion also of the balance sheet. I think we're going to pass most of these to the depositor. Was there anything I missed? We haven't. I think that's anything else.
The five-year, the change in the-
Oh, the five-year deposits, we don't have much of those.
The bonds, I think.
It has no effect. CEMPA has no effect on bonds. I think the proposal before was for foreign holders of Philippine issuers will be subject also to 20% withholding taxes, but that did not push through. There's no change actually in the cost structure of foreign or foreign currency bonds that we issue.
Right.
There's also the question of the five years. Yes, five-year liabilities held by locals before enjoyed a no withholding tax, okay, on the coupon. That has been removed, but we don't have much of those anyway, so it doesn't really affect our liabilities. I hope that answers your questions, Rafa. Thank you.
Okay. Thank you, Dino, and thank you, Rafa, for the question. I'll shift gears again. We have a question from Priya Iyer of Consilium Investment. It's a question actually about the Robinsons Bank integration. Specifically, is the integration complete? There's also a question about concerns of NPL in the institutional segment.
I think the only thing left on the integration of Robinsons Bank is the remaining branches. We have about 80. I can't remember the number. 70 branches that we will be doing in two tranches, one in October and one early next year, first quarter next year. That's the last thing we have to do with the integration of the Robinsons Bank. The second question had to do with
Yeah. Hold on. Second question is about concerns about NPL of the institutional segment.
Oh.
Yeah, Luis had.
Luis can talk.
Yeah.
Okay. For the NPL on the institutional portfolio. Right now it's down to 1.2%, and it's very manageable at this point, given that it's also timing for some of the corporations. They've recovered. They finally firm up their cash flow, and we can do the full restructuring, and they're able to complete the payments. That's one. Second, some of the mid-market companies have started to sell some of their assets to really reduce the loans, which is a good sign for the companies to address the loans that they have. Most of them we are able to restructure the loan, and given that we're able to manage the NPL at a very reasonable and low level at this point. That's what we see so far, and especially the pipeline that we have that we're seeing, and companies are very much open in restructuring their loans. Thank you much.
Okay. Thank you, Luis. We have a question as well from Eunice from Security Bank. Guidance for ROE and NIMs for the full year, given the expected provisions and projected rate cuts, and how many policy rate cuts are we expecting for the rest of the year and as well as NIM sensitivity guidance?
I can do the NIM sensitivity. Per 25 basis point cut in the policy rate, we expect a corresponding mathematical computation of the change in our NIM should be about 7 to 8 basis points. However, in practice, only about 4 basis points of that is realized. What we see is some loans that theoretically can go down when policy rates go down, they don't go down all the way. In reality, what we see is about per 25 basis point cut in policy rates, about four basis point cut in NIM. Dino, do you want to comment on the policy?
How many rate cuts?
It's one or two. That's all I'm sure of. Our economist thinks there's probably going to be two this year. I think that the views change rapidly, depending on what their opinion is of what the U.S. Fed is going to do. Yep. I'll leave it at that. I think it's one or two this year, and that's also what the governor said. Thank you.
Okay. We also have a question typed in from Johan Chan. I think they wanted some clarification of, I guess, discussion on AI. When we say AI on the credit side, are we talking about AI performing credit evaluation and analysis?
We're talking about applying AI tools across our whole credit process. We could be using it to first gather the data for corporate clients, and putting it into a format that can be read by an analyst or a credit analyst. We are talking about agentic tools that potentially could score, and could pull data from information that is put in by potential consumer banking clients, particularly on auto, on cards, and mortgage. We're talking about agentic AI that actually could approve the loan or the card, and eventually begin a process to actually release the funds straight through. That's still something that we are looking at. It's a project that we're beginning to look at, and we wouldn't do it across all our credit products.
We would start first with a simple credit product, build the tools around it, and then if that works, then we could roll it out to other credit products.
Thank you, TG. I'll go back. Go ahead, Eric. Yeah.
Quick comment, too, for Akash. You were asking about, was it Akash or Anyway, there was a question on power as a percentage of the total loan book.
Yes. From Akash.
That is currently 11.3% of the total loan book. It's power.
Thank you, Eric. There's another question. I think this is from Eunice again of Security Bank. This relates to the NPL, I suppose the NPL ratio. Would it be possible for us to share what NPL levels we'd be comfortable with given the expansion towards consumer-based loans?
Like a single number, again, that I can talk about.
I guess more like an NPL ratio.
that we're looking at? It depends on the product, right? I think we'd be comfortable at cards at four and a half. I think that's the budget we're looking at. For business bank, we're looking at 8% NPL. It depends on the product. Again, it depends on the mix.
Thank you, TG. A question as well from Gino Rojas of Macquarie regarding ECL. Specifically, what key factors drive the ECL model?
Francis, here.
I think, Eric, I can take this.
Do you want to take it?
If I understand correctly, it's three things, right? It's probability of default, loan exposure, and loss given default. That basically is what drives the ECL.
Is there a quick one?
Basically, all of that is driven by the macroeconomic variables. Obviously, if the macroeconomic environment improves, for example, interest rates go down and the growth rates accelerate. All things equal, it should translate to lower or improved ECL levels, if you could call it that.
Okay.
Okay. I'll shift gears again. There's a question from, I hope I pronounce it correctly, Nisha of PAPA Securities. Can you give us more color on credit card spends? What's driving the credit card increase? Has there been a shift in usage? Jenny, please go ahead.
Credit cards is actually still growing at about 28% to 29%. What's driving it really are certain categories. For example, in 2024, cross-border is really very big. What we're seeing in the first half of 2025 is that domestic spending is actually growing more than cross-border. In the domestic spending category, what's driving it really is essential spending. You would see that people would spend more really on the basic and essential needs as opposed to discretionary. As I said, we're still growing healthy at 29%, but the mix of how it's growing has changed depending on whether it's discretionary or essential spending. Thank you.
Thank you, Jenny. We have another question again from Yong Hong Tan. It relates, I think specifically about NIMs. Just wanted to follow up an earlier discussion on risk-adjusted NIMs. Your cost of funds, your price first before your asset book. I know that NIM can change depending on loan mix, but based on your second quarter books, do you have the asset yield or NIM after repricing from the recent rate cuts? What would be the asset yield or NIM, I guess, and the resulting asset yield or NIM after, I guess, the effects of the rate cuts come in? I suppose it's more driven by NIM sensitivity, which I think we may have answered already earlier.
Actually, if you look at the presentation that we sent out. I'm trying to look for just which page it's on. Okay. We have a slide there, loans and quarter NIM. If you look actually on the second quarter versus the first quarter, you see the NIM of 4.49% versus 4.67%. Those actually line up quite evenly to the triple R cuts, which happened right at the cusp of the shift from the second quarter to the first quarter. If you're looking at the impact of, was it the impact of the triple R cuts or?
No, rate cuts.
Of the rate cuts.
Yeah.
Okay. Sorry. That's a little different then. The triple R cuts, we call it about 3 to 4 basis points. The rate cuts, I guess that one you can't see it exactly on the NIMs.
Yeah.
That's right.
I think typically, Eric, we were saying about 4 to 5-
Yeah. the impact-
Basis points.
Over time, right? That's over time. I think the question was immediately what is the effect? The reality is that even though, for example, all the new time deposits that we have should be based on the new policy rate, there is also some competitive factors. It gets a little more competitive, a little less competitive, and that drives how much of the policy rate reduction flows into the deposit reduction. Most of it will, but sometimes it's 1 or 2 basis points, more or less.
Thank you, Eric. Okay. We have a question as well from Melissa Kwan. It's a question actually on the retail book. What is the risk-adjusted NIM on retail, and how has this changed in the last year? As you expand more into retail, what income segment are you penetrating?
Let me start. I don't know the exact breakdown of what our NIM for retail, because retail is such a broad topic. It ranges from where we have personal loans, where our asset yield is probably 40%, right? We have credit cards where the yield on the revolvers is 36%. We have mortgage where the yield is probably eight or nine. Auto is probably at 11. Eric will try to figure out what the blended yield on that is. I think what we're trying to do is really focus on the consumer segment and really broaden our penetration. Over the last three years, we have gone what I would call below our usual market. We've targeted lower income segments. We have experimented with new scoring models. We have loan products that target actually lower economic segments like My Bahay.
Our card products, we have card products where we are taking lower scores but giving them lower limits and experimenting. Even in auto, we have very new programs. We really are trying to broaden the market, and that accounts for the expansion of the loan book. As to margins, I don't know, Eric, if you have it.
Yeah. The question is per segment? What is the-
It's retail in general. We can give the yields of retail now, I think it's yields in general.
The problem is retail is not one homogeneous. If you look at housing, it's very different than if you look at credit cards, right? But those are both retail. Actually, during our last investor presentation, I provided a modified risk-adjusted margin for each of the loan segments. That was for institutional loans, it was 1.2, business banking, 4.7, auto, 3.1, housing, 2.0, credit card, 8.6, personal, 12.8, and microfinance, 32.6. That's breaking it down on a per segment, and those were the risk-adjusted margins as of the first quarter of 2025.
Thank you, Eric. Just wanted to check if there are any questions from the audience. Go ahead, Gilbert.
I'm surprised nobody has asked about dividends. You seem to be underperforming when it comes to dividend payout relative to what you were paying a few years ago. Thank you.
The dividend payout ratio did dip a bit in the first half of this year. That was affected by the fact that at the time we fixed the dividend, I think our expectations or our outlook was a little more positive. We had come out of the first quarter, which was quite, I guess, pessimistic. Then we were kind of transitioning into this period where the outlook was starting to look like the Philippines wasn't going to be affected, and we thought loan growth was going to be very strong. Therefore, there was some thought about maintaining some of that capital in order to make sure that we had enough capital for growth. I think as I've mentioned previously, our goal is that we should try as much as possible to avoid doing rights issues.
It's really not that great for investors, we're going to try and avoid doing that. Leading up to this expectation of maybe a little more loan growth, we thought we might try to hold back on that dividend a little bit. How we perform, what we look forward to in the second half of the year, will be affected by these kind of macroeconomic variables and our outlook. Therefore, I don't think you should necessarily say, oh, because the first dividend was this, that the second half dividend will automatically be the same because that's what we've done in the previous years. I think we're going to be a little more flexible in that respect.
To put it bluntly, our board asks us the same question, we have agreed with the board that we don't need to keep the same dividend twice in a row. We can change our dividend every semester.
Thank you.
Any more questions from the audience? I think that about covers all the questions. I wanted to thank everyone for all your questions. Of course, we at BPI always welcome your feedback and take them into careful consideration. Before we end the call, may we call on TG for some final thoughts?
Thanks to everyone for participating in this very animated discussion, and thanks to my colleagues here for joining us. As usual, we really value your participation, we value your insights, and more than happy to pursue any questions that you might have. Please do reach out to Eric, myself, or Haj if you have any further questions or if we can provide any more clarification or if you are unsatisfied with today's answers. Thank you. Thank you for joining us today.
Thank you, TG and Eric. Thank you, TG, Eric, and the rest of the BPI team. Ladies and gentlemen, that concludes today's earnings call. I want to thank everyone for your participation. To those joining us online, you may now disconnect, and to those on-site, please do join us for some snacks. Thank you.