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

Okay. Good afternoon, ladies and gentlemen, and welcome to BPI's second quarter earnings call. I'm Chinky Lukban. I'm the Head of our Corporate Strategy and Investor Relations. I am pleased to introduce this afternoon's speakers. With us today is BPI President and Chief Executive Officer, Cezar "Bong" Consing; and our Executive Vice President and Chief Financial Officer, Tere Marcial. We will start with a few words from Mr. Consing on the second quarter results of 2020, and Ms. Marcial will briefly go through the presentation deck that was sent out earlier. Thereafter, we will have a Q&A session. [Operator Instructions] Bong?

Thank you, Chinky. Good afternoon, everybody. The first quarter -- the first half, and I prefer to speak of business in the first half, is really one where the core business, especially our core intermediation business, did pretty well, where our trading businesses did very, very well and where our fees and commissions lagged the last year's results. Based on the strong saving businesses and the results therefrom allowed us to increase our provision to what is more than 4x the amount of provisions that we had in the first half of last year. And we're asking ourselves, is this the right level? Could we do more? And that is something we are assessing now. This is pretty much like the calm before the storm. We're expecting a super typhoon. We wonder whether it will hit us head-on or whether it will be less damaging than we initially feared. Nevertheless, we want to get ahead of it, and so you saw the provisioning we took. So that in a nutshell is our first half results. Again, strong core intermediation income, strong trading results, weaker fees and commissions, primarily because of the lockdown, and much heftier provisions.

All right. Thank you, Bong. Good afternoon, everyone. I will go through 2 parts of the presentation. First is the macroeconomic overview and then I will proceed with a presentation of our financial and operating highlights for the first semester of 2020. First, on the macroeconomic overview. The Philippine economy contracted by 0.2% in the first quarter of 2020. The last time our GDP shrank was as far back as 22 years ago in 1998 during the Asian financial crisis. Another contraction in the second quarter, which is widely expected, will push the country into a technical recession. For the rest of the year, the country's GDP may further deteriorate. Based on our projections, the economy may contract by at least 5% this year. The last time we had this magnitude of contraction was back in 1984, '85. The Philippines jobless rates also hit a high -- a record high of 17.7% as of June, reflecting the adverse effects of the COVID-19 induced economic shutdown on the Philippine labor market. With both supply bottlenecks and slump in demand, imports continue to decline, which underpins a stronger Philippine peso, and our Philippine gross international reserves hitting a new record high of $93.3 billion in June. This is 8.4 months' worth of imports and 7.3x short-term external debt. Next slide, please. With benign inflation environment, the BSP aggressively implemented measures to increase domestic liquidity, bringing M3 growth year-on-year to 16.6% in May. Among these monetary policy measures were a series of policy rate cuts from 4% at the beginning of this year to 2.25% currently, including the recent surprise 50 basis point rate cut. We saw a cut in RRR, reserve requirement ratio, from 14% to 12%, and the BSP actively purchasing government securities in the secondary market. Inflation print was as low as 2.1% in May, picking up to 2.5% in June. And our outlook for full year inflation is about 2.3%. Domestic interest rates went -- declined across the entire quarter, down by an average of 150 basis points year-to-date. Meanwhile, the national government has implemented various fiscal measures to support the Philippine economy. So far, the government's fiscal response, including indirect measures, is less than 10% of the country's GDP. As to the country's budget deficit, the government expects this to expand from 3% to 8% of GDP by the end of this year. As a result, the country's government debt to GDP ratio will most likely increase from 40% to 50%, but still a fairly comfortable position looking at the country's overall credit metrics. As to banking industry updates. Looking at the latest banking industry statistics, we are seeing early signs of a slowdown, with customer loan growth softening to 8.5% in May, this is for the industry, from 9.4% at the end of 2019. This reflects the impact of COVID-19 to the economy, particularly on the spending and investment decisions across both households and firms. Moving on to our financial and operating highlights for the second quarter of 2020. As we operate in the macroeconomic environment, as I have described briefly, the bank's second quarter performance can be summarized in 5 key points. First, revenue. We're seeing higher net interest income and higher net interest margin from lower funding costs, strong securities trading income. Asset quality and provisions, we are seeing higher credit costs due to higher expected credit losses. However, so far, we benign NPLs given the regulatory relief measures of the government to cushion the impact of the pandemic. Third, on earnings quality and cost discipline, we are showing strong preprovision operating profit and lower cost-to-income ratio. Fourth, on capital position and liquidity, we continue to maintain comfortable loss absorption buffer, and at the same time, we saw record high liquidity ratios. And lastly, the rebalancing of traditional and digital channels. The observed acceleration in digital adoption of our customer base, coupled with operational limitations brought about by the lockdown, strengthened our view to continue to rationalize our branch network. Moving on to profitability. Our net income for the second quarter reached PHP 5.29 billion, bringing our first semester net income to PHP 11.68 billion. This is down 15% compared to the first semester of 2019. For the first half of 2020, total revenues increased by 14.8% to PHP 52.69 billion. Net interest income grew by 12.5%, reaching PHP 36.4 billion, on 5.9% increase in our average asset base, supported by an 18 basis point expansion in our net interest margin to 3.55%. Noninterest income was PHP 16.29 billion, an increase of 20% versus 2019, primarily from higher securities trading gains. First semester pre-provision operating profit was up 32% year-on-year. Higher earnings allowed us to accelerate provisioning in Q2, sustaining the bank's reserves buildup since Q1 2020. First half provisions reached PHP 15.01 billion. Operating expenses for the first semester 2020 totaled PHP 24.19 billion, slightly down by 0.3% from the previous year, on lower expenses related to premises, technology, and marketing and product-related expenses. Comprehensive income was down 14.5%, mirroring the trend in net income. Moving on to revenue mix. On a quarter-on-quarter basis, we see asset yields have already shown a decline following the recent policy rate cuts. This brings first semester asset yields to an average of 4.93%, down 33 basis points year-on-year. Cost of funds also continued to trend lower, in line with the fall in interest rates and relatively strong CASA growth, thus improving cost of funds to 1.56%, down 56 basis points year-on-year. Year-to-date, net interest margin is higher by 18 basis points on lower cost of funds. Moving on to noninterest income. Total noninterest income increased by 20.3%. This is boosted by higher securities trading gains of PHP 4.7 billion in the second quarter, bringing first half trading income -- securities trading income to PHP 6.8 billion. Meanwhile, first semester FX income was flat year-on-year, as the higher corporate and retail flows were offset by weaker remittance volumes. Third, fee income for the first semester from our Asset Management and Trust business, Investment Banking business and Digital Payments registered a strong growth year-on-year. That's 14%, 61% and 307%, respectively. However, total income for the first semester was down by 14% year-on-year due to lower transaction volumes and fee waivers since the ECQ. And these are from our wireless businesses, including credit cards, branch service fees, retail loan fees, ATM fees and securities brokerage fees. Moving on to operational efficiency. This pandemic has impressed upon us the importance of continued focus on operational efficiency. During the second quarter lockdown, BAU spending for branches, such as utilities and marketing expenses, were channeled to pandemic-related expenses, such as meal and transportation allowances, implementation of health and sanitation protocols, testing and social distancing measures, which includes [indiscernible] multiple work locations. We have also reduced various discretionary spending related to domestic and international travel, advertising and publicity. We also slowed down on branch-related projects and the similar expenses, as further branch rationalization becomes more compelling given present market conditions. Total OpEx for the first semester reached PHP 24.2 billion, slightly lower by 0.3% year-on-year. Cost-to-income ratio improved sequentially to 42.5% from 49.6% in Q1 and well below the 52.8% cost-to-income ratio registered in Q2 of 2019. Early investment in technology has enabled the bank to pursue its digital transformation, while maintaining moderate expansion in our branch footprint. This pandemic has further accelerated digital adoption. During the ECQ, the number of digital transactions peaked at 90% of total transactions and before settling to 81% as of June 2020 as the economy -- as the quarantine measures continued to ease up. Also, digital channel registration has been growing steadily since the beginning of the year, with as much as 85,900 new enrollments for the month of June. This has grown at a compounded rate of 23% monthly since the beginning of the year. Moving on to balance sheet. Our net loans grew by 6.7% and deposits up by 6.3%. Both to be discussed in the next -- I'll discuss it in further detail in the next couple of slides. We show also that securities was lower year-on-year as we took profits on our securities position, allowing us to generate significant trading income. Year-on-year, loan-to-deposit ratio level was maintained at about 81%, and we are seeing higher CET 1 and CAR. Thus, our loss absorption buffer further improved. Next slide, please. Our loan levels and year-on-year loan exhibited declining quarterly trend in line with the industry. However, the bank continues to maintain its market share. That said, this year's focus is on asset quality and net interest margin accretive loans rather than aggressive loan volume. Despite our loan mix being 73%, 23% in favor of corporate loans, there is no change in the strategic direction of the bank in pursuing our high-margin businesses of SME and consumer, but understandably, the growth trajectory of our consumer and SME business will slow given the significant impact of COVID-19 on these segments. Next slide, please. Next slide. On funding and liquidity. Our total deposits grew modestly at 6.3% year-on-year. Among the customer segments, retail deposits, in particular, were up almost 11% year-on-year, with notable growth in our personal segment as well as overseas Filipino segment. Our CASA ratio was higher at 11.8% year-on-year, as BPI is a beneficiary of flight to quality during this pandemic. Net clearing wins were noted during the ECQ. Similar to our loan portfolio at almost 90% peso-denominated deposits, there is no exposure to currency volatility. Our LCR and NSFR were -- are well above regulatory requirements and are now at record highs. Total customer deposits, which account for almost 90% of our total funding. And despite this pandemic, BPI continued to be present in the capital markets, having issued 3x peso bond, contributing to our overall level of liquidity. Moving on to asset quality. We still see benign NPL levels and NPL ratio given the regulatory relief measures. Q2 2020 NPL ratio was at 1.83%. However, moving forward, we expect that NPLs would be higher as we assume the borrowers who may have available payment relief measures will have difficulty servicing debt and some of them might show NPL formation in the coming months. Reflecting the bank's aggressive provisioning stance, our NPL cover increased to 141% in June 2020 compared to 97.5% as of June 2019. And from pre-COVID credit cost level of about 40 to 50 basis points, our first half credit cost is tracking 200 basis points. On capital and earnings. This will show that we created capital at the end of June 2020, as we increased our qualifying capital coming from reasonably good profits, net of dividends that we declared in the first semester, combined with slightly lower risk-weighted assets relative to the end of 2019 in the first quarter of 2020. Total equity amounted to PHP 278.8 billion, with an indicative core equity Tier 1 ratio of 15.63% and capital adequacy ratio of 16.52%, both well above regulatory requirements. Profits are driven by aggressive provisioning stance taken by the bank, bringing down our second quarter ROE to 8.56% and ROA to 1.08%. We declared dividends of PHP 0.90 per share for the first semester of 2020, and this was paid out in June. Going forward, we expect to be able to continue our semestral dividend payout. And as a final point, COVID-19 has also given BPI another venue for stakeholder engagement, particularly on a segment that needs further support during this crisis. We launched the BPI Care bonds to support micro, small and medium enterprises who have been significantly affected by the global pandemic. BPI recognizes that these enterprises prices, which account for a significant percentage of our country's employment, are critical to the growth and recovery of our economy. I will stop there, and we will open the floor to questions.

Yes. Thanks, Tere. [Operator Instructions] We have a first question from Elizabeth Santiago of Abacus Securities. How much of loan growth was from extension of loan terms and grace period? How much was from new loan releases?

So this is a year-on-year loan growth. Bulk of this came from our Corporate business. Some coming from term loans that we have provided for the top most corporate names. On the other hand, for our consumer business, we actually saw much slower loan releases, particularly in the months of March, April and May. In the middle of May, we saw some recovery in new loan releases coming from our auto as well as our mortgage businesses. But still, these are lower than pre-lockdown levels. The latter part of June is already showing some further pickup in new releases. But again, total loan book for this segment is expected to be down for the rest of the year.

Okay. Thanks, Tere. Our second question comes from Harsh of JPMorgan. Harsh?

3 questions, all on asset quality. Could you please clarify what is the total cumulative provisions that you are thinking regarding COVID in 2020, '21 combined and the distribution between the 2 years? That's the first. I'll ask a couple after this.

Harsh, thanks for the question. To be honest, our 2021 expected provisions is still very difficult to quantify at this point. We have provided PHP 15 billion for the first half, and we don't think we will stop there. We will continue to build on provisions for the balance of the year, but definitely not as aggressively as we have done in the first half. Now the best indication I could give you with respect to how we will -- if further provision in 2021 is our expectation of NPL formation, which we expect to peak in 2021, possibly in the latter part, and it could spill over to early 2022, if you will look at the NPL number that we have reported, it's 1.83%. It's still very benign. And a large part of that is because of the implementation of the Bayanihan Act. So there are clients who actually -- there's a part where certain names are probably going to have difficulty servicing debt. But there are certain names, on the other hand, that actually just took advantage of the grace period and showed -- they decided to defer payment. Internally, we track what is the NPL ratio assuming all who took advantage of the grace period were to go in default. And our sense is the 1.83% should be closer to 2% or maybe at much higher than 2%. At the end of the year, we expect NPL to continue to accelerate from the levels today, possibly reaching 4%. Now 2021, based on our models, we are looking at anywhere between 6% to possibly 8% assuming a more adverse economic scenario. And if we look at the worst-case scenario, it could potentially hit 10%. The actual magnitude of real NPL will dictate how we will further provision in 2021.

Harsh, to be perfectly frank, when the lockdown started, we thought we would get to PHP 15 billion in provisions by the end of this year. But as the lockdown progressed and as we look at our clients and saw what they were seeing in terms of sales, et cetera, we thought we should increase the pace of our provisioning. That being said, I think I'm comfortable in saying that the second half provisioning will be lower than the first half provisioning. I think PHP 15 billion, if anything, we were looking ahead, I think in the second half, that number will be lower. And then what we will do is we'll see where we are in the fourth quarter, at the end of the fourth quarter, and assess for next year. Tere had made some points about 1.83% NPL ratio. If you took -- if you look at the people that took advantage of the Bayanihan Act, and you simply assumed that everybody who took advantage of the Bayanihan Act went into default, that NPL ratio would be closer to 3%. Okay. And we're tracking that. Now what we also have to be mindful of is infections are going up again in Manila. We have to prepare for a scenario where there is a chance, I hope small, that we will go back into an MECQ environment. If we go back into an MECQ environment, that has also several knock-on effects that we have to prepare for.

Got it. Okay. That's fair. So is it fair to say -- I don't, sorry to put you on a spot, Bong, but that the first quarter number, is that a good indication of a quarterly run rate we should broadly think about, well, next 2 to 4 quarters?

I think that's fair. But remember, all of this thing will swing in our provisioning. And we get different senses every single week as we sit down with our clients. But I think that's fair.

Got it. Got it. Okay. Now so -- just so that I get it right. When you say everybody who took advantage of the Bayanihan Act, if you consider it goes from 1.8% to 1.3%, what it tells me is only 1.2% of your borrowers of your loan book equivalent took the advantage of moratorium. Is it that...

Harsh, it goes from 1.8% to 3%. But remember, not all amounts due -- fell due during the Bayanihan period, right? The Bayanihan period is a subset of everything, and then not everything fell due during Bayanihan. So if we simply say, "Ah, you took advantage of Bayanihan, we'll not tag you as NPL. So that goes to 3%. So to say that it's only 1.2% of clients is actually inaccurate.

Okay, okay. So then, let's say, if you look at your entire loan book and the proportion of loans which availed of moratorium, what would be that proportion? And then the follow-up would be, what portion -- I know Tere gave a very wide range. But how should we think about relapse into stage 2 and 3 of that entire portion of loans which did avail of some sort of moratorium?

Well, I don't have those numbers right now, Harsh. Do you have them, Tere?

Our estimates, Harsh, is, at least on the consumer side, towards the middle of May, we thought that approximately 80% of our clients decided to avail of the moratorium. There was still about 15% to 20% who actually paid. But since the -- no -- the moratorium has expired, we were actually able to resume collection from these consumer loans.

Yes. What happens is the moment the moratorium expired, Harsh, we start automatically debiting our clients again. And the amount -- and those -- the successful automatic debit rate is very, very high. It's very, very high, which gives us comfort that we have the right kind of clients.

Right. Any -- sorry, the last question. What is that success rate? Like what proportion of guys who availed of moratorium were -- have again started paying back in last 45-odd days?

Well over 90%, in the high 90s.

Okay. Our next question is from Robert of Citi. I'll read it aloud. His question is, with the post-ECQ grace period over [Audio Gap] expect to apply to the bank for loan restructuring? And what are the key stress sectors, noting that BSP data already shows [ May ] loans past due at 4.88% of loans?

Well, the -- it's hard to give an estimate in terms of the percentage and number of loans that were restructured. I can give you some indications of the segments. Understandably, these are companies that are in, what we call, the COVID-exposed industries, such as food, hotels, restaurants, some auto dealerships, property developers, transport. So we are seeing such discussions happening, but we will have to quantify what percentage of those names will actually be considered as restructured. The BSP has very specific definition as to when a loan is to be recognized as restructured. Among those where we have active discussions on loan payment deferrals, I would say less than half would qualify for the BSP definition of restructuring.

Just to add a little color. On the consumer space, our OFW and self-employed clients are the most affected. And obviously, on the corporate space, anybody related to hospitality or tourism are very affected.

Okay. A follow-up question from Robert. Could you provide updated guidance on credit cost, including going into 2021?

Well, as we have indicated earlier, for the balance of 2020, we will continue to build the provisions, but not anywhere near the PHP 15 billion level that we have within the first semester. In terms of credit cost, I would estimate this to be in the 120 basis point range, probably slightly -- probably even higher. I guess what's important is the actual amount of provisions because credit cost -- translating into the credit cost will depend on the level of our loan book. And just very abruptly, maybe something that is closer to PHP 18 billion, it really will -- it will be decided as we see the coming quarters with respect to how we are seeing the increases in our expected credit losses and after [indiscernible].

Okay. Thanks, Tere. We have a question from Fredrick of Maybank. Fredrick?

Can you hear me, guys, clearly?

Yes. Clearly. Go ahead.

I originally have 2 questions, but the first one is related to credit cost, so I won't repeat that anymore. So just my only question is, can you give more color on the effective tax rate in the first half of 2020? It appears that you actually booked a tax benefit in the second quarter. May we know what's the reason of that tax benefit? Is it mostly because of timing issues with the tax authorities with regard to front-loading of the provisioning? And do we expect this to normalize moving forward?

Well, 2 reasons. One, as you mentioned, is related to the tax advantage coming from provisions. And the second is because we registered a significant amount of trading income in the first and second quarters, and most of this trading income are exempt from taxes.

Okay. Maybe just a follow-up question. Just to be clear. You mentioned earlier, Tere, that provisions would probably fall somewhere around the PHP 18 billion level. You're already at PHP 15 billion. So for -- in the BIR's point of view, the provisioning in the second quarter would be booked in the third and fourth quarter. That's why we should expect effective tax rate to normalize back to the 20% level moving forward?

Yes. It should be higher in the second semester.

Our next question is from Aakash of UBS. Aakash, go ahead.

Can you hear me all right?

Yes, we can.

Okay. Great. My first question is again on restructuring. I just want to understand -- make sure that my understanding is correct. So what you have said is roughly 80% of your mortgage loans were under moratorium. And then after the MECQ ended, 90% of these started paying. So I think what you're saying is 8% or so of the mortgage loans are still in the restructuring categories. Could you share what is this number for corporates and SMEs? Just a [ rough side ] would also be fine.

Aakash, well, as far as we can see from the numbers, there isn't a significant amount that has already gotten into a restructuring classification. And since a greater number of clients actually settled amortizations upon assumption of -- basically, when we went into GCQ, when the Bayanihan Act expired, the NPL formation that we received [Technical Difficulty] and so [Technical Difficulty] because as you count NPL, which is 90 days plus 1, so it's -- as it falls due, we will see that possibly in July, August, September. So the third quarter will be critical in determining whether such those who availed of Bayanihan will actually become NPL. But as to the actual loan amount that's restructured, no significant amounts have fallen into a restructured bucket, at least that's as far as what we have seen in the last 2 months.

Are you okay, Aakash?

Yes. Sorry, just a follow-up question on that. So what will be the staging treatment of the restructuring that you do in the coming months? So for example, in the next couple of months, you might restructure some of your corporate or SME loans. What is the stage in treatment? Will they be considered as stage 2? I think you also said that BSP's definition is very different. Could you give us some color on what that definition is? And how is it different from the new accounting rules?

Well, the staging of loans is determined by the -- basically PD and LGDs. And we have seen some migration from stage 1 to 2. And those numbers are already incorporated in the increases in ECL that we have seen year-to-date. If we just look at all loan portfolios, across the different loan portfolios, across different stages, 1, 2 and 3, we recorded approximately 40-plus percent increase in ECL from year-end levels. Having said that, the actual provisions that we have taken is more than that level because we were given basically go signal by our external accountants that given the possible worst-case scenario, it provides additional management overlay to provision more in anticipation of the worst-case scenario. So if you look at ECL increase per se, it increased by less than the actual increase in provisions that we took in the first half of the year.

And just to clarify. When we went into this crisis, we went back and we did our ECL models, because it was very clear that our ECLs models needed fine-tuning. So we did it to basically allow for more ECL recognition, if you will. But even after that, we went ahead and took more provisions than the ECL numbers shown.

Got it. And then the last question I have is on the trading and trading income and costs. So on the trading side, you've had a pretty good momentum in the first half. How should we think about the second half? Can you sustain that kind of trading income momentum? And on costs, again, if I look at the 1Q, it was pretty impressive that your costs fell. When you look at across the categories, it was probably the other costs that really drove this decline in cost quarter-on-quarter. So what exactly went on there? What was in the others category which fell? And can this continue in the coming quarters?

I don't think we would plan on being able to replicate our trading gains in the first half of the year. We might get some, but I think the first half was unusual for reasons you well understand. Now in terms of costs, the second quarter cost-to-income ratio was way down. We do think in the third quarter and in the fourth quarter, the cost-to-income ratios might be a bit higher. A lot of it depends on what we actually would put back online. This crisis has had us reschedule a lot of our things we would otherwise be spending on and rethink some of our platforms. But if you want to say in terms of the second half expense base, we're looking at that increasing relative to the first half. Still, we're looking at an expense base that for the full year will be -- won't be higher than all of last year. It will come somewhere around that.

Yes.

Is that about right?

Yes. Just to add, Bong. It's really going to be influenced by the denominator effect. Our internal estimates show that our first semester revenue is higher than our second semester revenue. A lot of it is because trading income will not be replicated in the same magnitude in the second half. So by -- even if we maintain our costs in the second half to the same as the first half, cost-to-income ratio will deteriorate just because of the denominator effect.

We have a question from Melissa of Goldman Sachs. With the acceleration of digital registration and transactions online, what are your thoughts on branch expansion going forward?

Well, what I think that COVID has thought us is that the tipping point for digital, this will be brick-and-mortar as we move forward. We thought the tipping point was probably 2 or 3 years away. But it looks like the tipping point is about that. So I suspect we're not alone amongst the banks in looking at our branch network. Today, less than 10% of all our transactions happen in the branch. More than 90% happen in through our digital channels or through ATMs. But in fairness to our branch colleagues, the majority of the value of the transactions still happens in the branch. And the trick is to try to get to a stage where the number and the values kind of equal, but that's going to take a while. So if we're obviously looking at our platform, our brick-and-mortar platform, I don't think we're the only bank thinking about it, and we'll figure this out. But we have the advantage, I think, of probably more digital transactions happening every single day now than any other bank out there, and that gives us a good momentum as we go forward.

A follow-up question. Can I check if the assumption on which you provided for this credit cost is an NPL of 4%, as you mentioned? And what is the loan loss coverage you should expect the bank to have once NPLs are recognized?

Well, a lot of this depends on how long this crisis happens, right? What typically happens in a crisis like this is NPL coverage ratios go in excess of 100%, well in excess, as banks like us take a lot of provisions. And then over time, that comes down. It's hard to say that we're going to try to maintain 100%, because that may not be the most efficient response. So what I think will happen is, early in the crisis, which is now, you'll see banks like us building up our coverage ratios well in excess of 100%. And then when the NPLs hit, you'll see those numbers coming down.