Grupo Financiero Galicia SA Q3 FY2025 Earnings Call

Good morning, ladies and gentlemen, and welcome to Grupo Financiero Galicia Third Quarter 2025 Earnings Call. This conference is being recorded, and the replay will be available at the company's website at gfgsa.com. Some of the statements made during this conference call will be forward-looking statements within the meaning of the safe harbor provision of the U.S. federal securities law and are subject to risks, uncertainties that could cause actual results to differ materially from those expressed. Investors should be aware of events related to the macroeconomic scenario, the financial industry, and other factors that could cause results to differ materially from those expressed in the respective forward-looking statements. Now I will turn the conference over to Mr. Pablo Firvida, Head of Investor Relations; and Gonzalo Fernandez Covaro, CFO. Please Mr. Firvida, you may begin your conference.

Thank you. Good morning, and welcome to this conference call. According to the monthly indicator for economic activity, MI, the Argentine economy recorded a 5% year-over-year increase during September. In year-to-date terms, the economic expansion reached 5.2%. During the third quarter of 2025, the primary surplus reached 0.5% of GDP, and an overall surplus of 0.1% of GDP was reported. This result was explained by revenues increasing by 32.8% year-over-year, whereas primary spending rose 30.6%. During the first 10 months of 2025, the primary balance stood at 1.4%, while the financial balance amounted to 0.5% of GDP. The national consumer price index accumulated a 6% increase during the third quarter of 2025 and a 24.8% year-to-date increase as of October. After four consecutive months below the 2% mark, headline inflation was 2.1% in September and 2.3% in October, accumulating 31.3% in the last 12 months, the lowest level since July 2018. The third quarter was marked by high volatility in the months leading up to the midterm elections. The exchange rate came under pressure at times nearing the upper limit of the floating band which prompted the Central Bank to step in with foreign exchange sales. Nonetheless, the exchange rate averaged ARS 1,400 per dollar in September 2025, a 15.6% devaluation compared to June 2025. Meanwhile, peso-denominated interest rates saw sharp swings, reflecting increased uncertainty and liquidity shift. In fact, the average rate on peso-denominated private sector time deposits for up to 59 days averaged 48.7% in September 2025, up 16.5 percentage points from June 2025 levels. Private sector deposits in pesos averaged ARS 94.1 trillion in September, increasing by 5.6% during the quarter and 53% in the last 12 months. Time deposits in pesos rose 13.1% during the quarter and 76.3% in the year. Peso-denominated transactional deposits decreased 2.4% during the third quarter but increased 31.5% in year-over-year terms. Private sector dollar-denominated deposits amounted to $32.6 billion in September 2025, increasing 7.2% during the quarter and rising 38.9% in the last 12 months. Peso-denominated loans to the private sector averaged ARS 79.3 trillion in September, showing a 9.7% quarterly increase and a 105.4% year-over-year increase. Private sector dollar-denominated loans amounted to $18.3 billion, recording a 15.8% quarterly growth and a 153.4% annual increase. Turning now to Grupo Financiero Galicia. Net loss for the quarter amounted to ARS 87.7 billion, due to losses from Banco Galicia for ARS 104 billion, from Naranja X for ARS 6 billion, and from Galicia Seguros for ARS 12 billion, partially offset by the profits from Galicia Asset Management for ARS 25 billion. This loss represented a minus 0.8% annualized return on average assets and a minus 4.7% return on average shareholders' equity, while accumulated annualized figures for the fiscal year reached 0.9% and 4.7%, respectively. The quarter includes extraordinary restructuring expenses associated with the merger with HSBC for ARS 105.3 billion net of income tax. The quarter ROE without the extraordinary expenses would have been 1%, and the nine months ROE 6.9%. The result from Banco Galicia included ARS 101.1 billion of extraordinary expenses, and in addition, were negatively affected by the increase in the cost of risk associated with the growth of the loan book and the increase in the nonperforming loans in the retail segment, particularly in personal loans and credit card financing. Together with a decrease of the financial margin associated with an environment of high-interest rates and regulatory increase of reserve requirements. It is also worth noting that most of the comparisons will be made against the second quarter of this fiscal year as figures for the third quarter of 2024 do not include information about the acquired business of the former HSBC Argentina. Net operating income decreased 23%, and net interest income decreased 10%. Net results from financial instruments were down 89%, and loan loss provisions increased 26%, which were partially offset by a 9% growth of net fee income and a 12% increase in profit from gold and FX quotation differences. Average interest-earning assets reached ARS 22.7 trillion, 8% higher than in the previous quarter, primarily due to the increase of the average portfolio of loans, 5% in pesos and 27% in dollars. In the same period, its yield decreased 259 basis points, reaching 30.1%. Interest-bearing liabilities increased 27% from June 2025, amounting to ARS 19.9 trillion, primarily due to the increase of time deposits in pesos and savings accounts in foreign currency. During this period, its cost increased 88 basis points to 16.5%. Net interest income decreased 10% when compared to the second quarter because of a 35% increase in interest expenses due to a 36% higher interest rate on time deposits, partially offset by a 7% increase in interest income, mainly due to a 12% higher interest on loans and other financing to the private sector. Net fee income increased 9% from the previous quarter due to a 6% higher income from credit card fees and up 19% from fees on deposits. Net income from financial instruments decreased 89% due to an 88% lower result from government securities. Gains from FX quotation differences were 12% higher than the year-ago quarter, including the results from foreign currency trading following the lifting of exchange restrictions. Other operating income increased 11% in the quarter, mainly due to a 45% increase in other income, primarily corresponding to credits recovered. Provision for loan losses increased 26% due to the growth of the financing portfolio and to an increase in delinquency that is limited to personal loans and credit card financing to individuals in pesos. Personnel expenses were 83% higher than in the second quarter due to the voluntary retirement program recorded in connection with the restructuring plan following the acquisition of HSBC's business in Argentina. Administrative expenses were 11% lower than in the previous quarter due to a 32% decrease of expenses for maintenance and repair of goods and IT, and to a 14% decrease of higher administrative services. Other operating expenses increased 5% due to a 7% higher turnover tax. Results from the net monetary position decreased 9% from the second quarter following the declining evolution of inflation. The income tax charge was positive as the pretax net income was a loss. Demand financing to the private sector reached ARS 20.4 trillion at the end of the quarter, up 14% in the last three months, with peso financing increasing by 5% and dollar-denominated financing growing 35%. Net exposure to the public sector was down 3% compared with the previous quarter, primarily due to a 38% decrease in government securities in pesos measured at fair value through OCI, offset by an increase in government securities in pesos at amortized cost. Deposits reached ARS 22.9 trillion, 8% higher than the quarter before, mainly due to a 26% increase in dollar-denominated deposits, primarily time deposits that were up 72%. The bank's estimated market share of loans to the private sector was 14.8%, 30 basis points higher than at the end of the previous quarter, and the market share of deposits from the private sector was 16.4%, 40 basis points higher than in the second quarter of 2025. The bank's liquid assets represented 94.5% transactional deposits and 59.2% of total deposits compared to 94.3% and 65.2%, respectively, from a quarter before. As regards to asset quality, the ratio of nonperforming loans to total financing ended the quarter at 5.8%, recording a 140 basis points deterioration as compared to the 4.4% of the second quarter. As I mentioned before, the deterioration is limited to the personal loans and credit card financing portfolios. At the same time, the coverage with allowances reached 105%, down 16.4 percentage points from the 117.9% recorded a quarter ago. As of September 2025, the bank's total regulatory capital ratio reached 22.1%, decreasing 160 basis points from the end of the second quarter, while the Tier 1 ratio was 21.8%, down 140 basis points during the same period. In summary, the third quarter was marked by high political effects and monetary volatility and negatively affected margins and asset quality. In addition, the results were affected by very high one-time expense due to the restructuring of the merged banks. Despite this, Grupo Financiero Galicia was able to keep liquidity and solvency ratios at healthy levels, and we expect an improvement in profitability during the fourth quarter and next year. Now Gonzalo Fernandez Covaro will make some additional remarks.

Hi, everyone. Well, continuing with what we see for the future. I mean regarding how we see the rest of the year, October continued with low margins due to the high interest rate that we saw in the third quarter. But we are already seeing a vast improvement in margins in November. We are really seeing margins at the same level as the second quarter and the first half of the year in November, and we expect the same for December. Portfolio performance still needs some time to get back on track, so we still see a deterioration in the fourth quarter at a lower trend than before, but still some. Overall, the bank will be better, will improve returns, mainly due to the margin improvement. But Naranja X, we have some headwinds in terms of portfolio performance. With this mix, we are seeing the ROE for the full year 2025, around 4%, the reported one. If we exclude the nonrecurring integration costs that we mainly booked in the third quarter, we should be around 6%. Talking about 2026, we're expecting an ROE in the low teens range, I would say, between 11% and 12%. Of course, a lot of moving targets for next year. We will be updating this guidance in further quarters. But this is our best-case scenario to be around 11% to 12%. Margins, we'll see improving them in the first months of the year, together with what we are seeing in November and December, then some kind of light reduction as a consequence of the rate reduction, but not really high. So we'll still see healthy margins next year, I would say, in the levels of the second quarter. NPLs, we expect a peak on NPLs in March of next year, but then improving as a good portfolio that we are originating is gaining weight in our mix, and we will end the year with NPLs better than the run rate that we are having now. Regarding costs, we are also seeing a reduction year-over-year in costs because of all the restructuring we have done. You saw the restructuring costs we booked in the third quarter, and that generated a 1,000 heads reduction in the group quarter-over-quarter. If we add up all the year, we have a headcount reduction of 2,000 heads for the year. So that is, of course, generating cost reductions for next year. We are seeing already fourth quarter of next year, our projection shows, our fourth quarter of next year ROE run rate already at the 15% level. So that puts us with a solid base to start 2027 and deliver ROEs above 15% is the target ROE that we are aiming for the longer future. So with that, I mean, we are also open for any questions you may have.

Our first question comes from Daniel Vaz from Safra.

I'm looking at your capital ratio for 2021 at the group level, which decreased by 120 basis points from the second quarter. You mentioned that your new cohorts of origination are improving, and you anticipate a peak in non-performing loans in March. However, your return on equity remains significantly low compared to your targets. How do you foresee capital levels in this situation, and what is the minimum capital that you consider acceptable for risk-taking within the group? Additionally, if you need to cut back on origination, how are you managing new originations in relation to the beginning of the year? Earlier this year, I believe the longer duration affected your margins and cost of risk. In comparison to other fintechs, their duration adjusts more quickly. Both of these aspects relate to each other: how is your capital position and how does it relate to your origination compared to the beginning of the year?

Our capital was affected by the reserve in other comprehensive income due to bonds valued for hold to collect and sale, which creates a reserve in equity that fluctuates between accrued income from the bonds and market value changes. In the third quarter, bond prices significantly declined, resulting in a negative reserve in equity of ARS 160 billion that impacts the equity ratio. As of October, our Tier 1 ratio stands at 24.5%. This improvement is partly due to the OCI reserve being slightly positive, leading to a recovery of ARS 160 billion in just one month following the election-related bond rally. With the current capital levels, we feel secure, setting our minimum operational appetite at around 13% to 13.5%, but we don’t foresee approaching that level soon. Based on our forecasts, we have sufficient capital to sustain our operations until at least the end of 2027, and we see no restrictions on growth. We will keep monitoring this closely, but as the bond reserves are now stable, we don’t anticipate needing additional capital or facing limits on our loan book growth through next year and into 2027. Regarding your second question, I think I didn’t fully understand it.

Origination of loans and maturity of the loans.

I mean, we continue to originate, both commercial and consumer lending. We have put some slowdown in consumer lending due to the portfolio quality, as you have seen. First mortgages will significantly reduce the origination of mortgages. In that case, not because of quality, but because there is no securitization market. And as you know, we cannot be putting 30-year lending without any securitization market where we can offload those loans. But in the consumer sector, we continue lending personal loans that may have a duration of 2 years, 2.5 years. We are also increasing car loans or auto loans with the same duration. Talking about commercial financing, we are still originating with a very short duration. Slightly, we start to increase duration because the demand was not there. Now after the elections, with Argentina stabilizing and with a lot of growth potential in the country, we are expecting to see more projects for our clients to finance longer terms, but that's not yet happening. However, we are expecting that next year, the duration in commercial lending should be getting longer than what we have today.

Our next question comes from Ernesto Gabilondo from Bank of America.

My first question is about your loan growth expectations for next year. Can you provide some details on expectations for each segment? I've also heard about new private investments in Argentina. Have you estimated an amount or can you share which regions and sectors you are seeing these investments? How do you plan to engage through corporate loans, particularly for SMEs? That information would be very useful. My second question relates to asset quality. I believe you mentioned that the NPL ratio might peak by March next year. If that's accurate, could you provide a possible range? Are you also observing a similar trend in the cost of risk? Do you expect the cost of risk to peak by March next year as well? How should we view the overall cost of risk for next year?

Thank you for the question. Regarding the outlook for next year, we anticipate a growth in lending of about 25% in real terms. Our goal is to continue increasing our market share, so this projection will adjust based on market growth, which we expect to be around 20% to 22% in real terms. In terms of sectors, we are seeing significant growth, particularly in commercial lending, with many projects and investments in the country that our customers are discussing. We also aim to grow consumer lending by utilizing new origination tools and adjusted models to secure better credits. We will start off slower in the initial months, maintaining a cautious approach, before ramping up growth as quality improves. In commercial lending, we are observing a considerable investment focus on the oil and gas sector, especially in Vaca Muerta, as well as within the mining sector for copper and lithium. Additionally, we are actively involved in agri-business, which is expected to have a strong harvest this year and attract good investment opportunities for the upcoming year. We are also beginning to notice movement in local mergers and acquisitions, along with some potential privatizations on the horizon, though we don’t have specific names yet. The intention is to remain close to our customers, as many of these projects may be too large for the local financial market to finance entirely. We can participate with smaller amounts or in transactions that local balance sheets can manage. We sense strong investor appetite for Argentina, which suggests we will outpace market growth next year, ultimately enhancing our year-over-year returns. Regarding non-performing loans, we project that the peak will occur around March next year, estimating it to be about 6% to 7%. We also expect the cost of risk to peak around the same time, forecasting a range of 9% to 10%. Following these peaks, we anticipate both ratios will decline by the end of the year. We're already seeing improved performance in the new consumer lending portfolio compared to older ones, but it will take time to evaluate the older portfolio's results, with more clarity expected during the second, third, and fourth quarters of next year.

Super helpful, Gonzalo. Just another question in terms of this potential growth that you can see for the loan book next year. Another competitor just mentioned the possibility to tap the markets next year. Is this something that you are also exploring?

Bonds or equity?

In bonds.

I mean, yes, of course, it's something that we have in our alternatives. We need to see how the windows are starting to open. Really, we don't see the need right now. But of course, that's something that we are always evaluating. We need to see, of course, the equation, the profitability equation of the cost that the market could offer at some point. But yes, mainly considering larger tickets for the projects that may come, yes, definitely, it's an alternative that we consider very seriously.

Our next question comes from Brian Flores from Citi.

I just wanted the first question to be a clarification on the ROE trend because you mentioned the peaks of NPLs and asset quality, as you mentioned, cost of risk by March, right, of 2026. So you mentioned 11% to 12% in terms of real ROE for 2026 with reaching the 15% in the fourth quarter. So would that mean we should see a mid- to high-single digit in the first half? Just thinking about the speed of the recovery, right? Apparently, it seems to be a very gradual recovery. Just wanted to check on that trend, Gonzalo. My second question is perhaps a follow-up on Ernesto because I think we're all thinking about external funding, right? But you have perhaps one of the best franchises in Argentina, meaning that deposits are very, very relevant. So I just wanted to understand if the visible funding cost advantage that you have demonstrated in previous quarters should continue? Or do you think the funding cost war should, I would say, increase or deepen in 2026?

No, thank you. To address your first question about the trend in ROE, we did notice a slowdown in the first quarter. The figures you mentioned might be accurate. We are anticipating a recovery in the first quarter, where we expect margins to be at good levels. However, we still face a significant burden from non-performing loans in the last month of that period, before starting the recovery in the third quarter. Therefore, I would expect lower ROE in the fourth quarter, followed by a gradual improvement into the following quarters and continuing into 2027. Your assumption regarding ROE evolution is indeed correct. It's important to note that the group includes Naranja and the bank, and Naranja also needs to enhance its portfolio performance. Consequently, we will require a couple of months next year to achieve ROE above two digits. Regarding funding, we are currently evaluating potential market debt, but our primary focus is on our deposit base. We see some opportunities in deposits for next year as market liquidity is returning. With the reduction in interest rates, we expect the market to become more liquid. Additionally, changes in regulations affecting mutual funds might allow banks to better access market funds. Currently, money markets hold substantial deposits from customers in Argentina, with some portion deposited in banks and the rest placed in the market. Moving into next year, we expect these entities to deposit more funds in banks, enhancing liquidity for lending. Our goal is to increase our deposits to capture greater market share. All our business lines are aligned with this strategy, as we view deposits as more stable and cost-effective funding. However, depending on the pace of credit growth, we may need to turn to the market. If necessary, we will pursue that option, but our top priority remains deposits, and we believe deposits should start to grow more significantly next year compared to this year. While it may not match the pace of lending, it will improve from this year’s levels.

Perfect, Gonzalo. I think the last guidance you mentioned in the second quarter on deposit growth was around 35% in real terms. So I just wanted to check if you are revising this number. And I also understand, as you mentioned, the deposit growth for 2026 is going to be lower than the portfolio.

Yes. For this year, we are maintaining our growth guidance. For next year, we expect around 20% growth in real terms for deposits and 25% for lending. However, this is subject to change as there are many factors affecting next year's outlook. We will adjust our guidance as we assess the situation. Just a month ago, there was a lot of volatility, but now we are seeing stabilizing interest rate reductions. We need to observe how everything develops, but so far, we are assuming around 20% for next year.

Our next question comes from Tito Labarta from Goldman Sachs.

A couple of questions also. I guess, just on the Naranja, you mentioned Gonzalo that needs to recover as well. Do you think NPLs peaked there also in 1Q? Or how do you see the evolution of asset quality in Naranja and then also your ability to resume growth at Naranja? Then second question, just on margins. Do you think we saw some pressure this quarter, I mean, just given all the liquidity issues in the quarter. Do you think this is the bottom? Should that already begin to recover in 4Q? Or will that take a little bit longer until you start to get the loan growth and asset quality under control, just to think about the evolution of margin in the short term and I guess, thinking about 2026 as well?

Yes, regarding Naranja, we are observing a similar timeline for their peak, which we expect will occur in March or April next year. They have been implementing several initiatives, and their turnaround is happening faster due to a shorter lending duration. As a result, they are able to resolve issues in their portfolio more quickly than the bank. We anticipate that March will be their peak as well, and we expect to see an improvement in non-performing loans for Naranja for the remainder of the year. Their focus on lower segments does lead to greater fluctuations in bad loans, but it can also result in more positives at the same time. Concerning net interest margins, we noted that the lowest point was in October. While the fourth quarter includes October, which had low margins, we expect one-third of the month to be impacted by that margin dip. October was particularly challenging due to it being the worst month before the election. However, November has shown a rapid and significant recovery in margins, and by December, we anticipate margins to return to levels seen in the second quarter. In summary, the lowest point was in the third quarter and October. Looking ahead, margins should remain strong for the next year, although we may see a slight decline in the second half due to ongoing rate reductions. Currently, our funding costs are decreasing significantly, but we will see slower reductions in lending rates as we have already locked in longer-term loans at higher rates. Therefore, we expect to maintain higher margins in the first half of next year, with only slight reductions anticipated thereafter.

Okay. No, that's helpful, Gonzalo. And just on the reserve requirements, I mean, they've been reducing a little bit. Do you think it's enough now that liquidity is less of an issue? Do you think that they need to reduce the reserve requirements further from here? Or how do you think about that and the impact on your liquidity?

I would say that so far, things are okay. The Central Bank made some changes to liquidity requirements last week, which improved the daily calculations and reduced cash encashments with zero interest by 3.5%. This change will enhance EBITDA margins for banks starting on December 1. While this isn't a significant move for injecting liquidity into the market, it's possible that the Central Bank may reconsider this policy next year if necessary. I wouldn't suggest that a revision is needed right now or in the next three months, but depending on market conditions, there could be an opportunity for changes down the line.

Our next question comes from Camila Azevedo from UBS.

My question is a follow-up on Ernesto's question on asset quality. I would like to get a better view of the asset quality dynamics during this quarter, mainly between segments. You said we should end the year with better NPLs than current levels. Could you please share more details on that? So, in general terms, what should we expect? And with this, what coverage ratio would be comfortable going forward?

Sorry, Camila, there was a noise in the middle of the conversation. We couldn't get the first part. Yes, the part of ratio, not the other part.

Sure. So I'll repeat the entire question. I would like to get a better view of the asset quality dynamics during this quarter and which levels should we expect for the end of the year? You said that we should expect better NPLs. So, in general terms, at which levels? Did you get the coverage ratio part?

When we discuss NPL for the end of next year, not this year, the peak will still be in March next year. Looking ahead to the end of 2026, I would estimate NPLs to be in a range of around 4.3%, give or take, likely between 4% and 4.5%. That seems to be the expected range for NPLs by the end of 2026.

And the coverage?

I didn't get that.

And the coverage, the last number is 101.5%. Really, it comes from the model of expected losses, talking with the credit department. They say that the coverage is beginning to grow, and it's likely that at the end of next year would end up at 110%. But really, it's...

When we grow your book, we create a significant amount of upfront reserves, which increases your coverage. Currently, we are in the process of utilizing those reserves that were established when the portfolio experienced significant growth. We aim to stabilize the portfolio, which should lead to growth again. However, right now, we are using what was previously booked because new loans haven't added much to the upfront reserves. It’s a mathematical situation, but we are confident in our current level.

Our next question comes from Pedro Offenhenden from Latin Securities.

I wanted to ask if there are any remaining integration costs from the HSBC acquisition that could impact results in the coming quarters?

No. I mean, I would say the restructuring, nothing big. I mean we may have some small thing in the fourth quarter, but regarding systems that we are shutting down, but not restructuring costs, which is the big portion, we booked everything in the third quarter, not just the people that left in the quarter, but also what we plan that the ones start leaving until the end of the year. So everything is booked there. Something very small, not related to restructuring may happen in the fourth quarter related to systems, but really small, nothing important.

Our next question comes from Carlos Lopez from HSBC.

First of all, congratulations on how brave you are because you are giving predictions for the ROE for the middle of the year and for the end of the year. I hope that those forecasts are actually achieved. More concretely, I realize that we have gone through three conference calls, and I don't think anybody has told us what their economic assumptions are. What do you expect for inflation, interest rates, and the currency for the end of next year? Maybe you have said it, and I missed it, sorry for that. Second, in terms of liquidity, your LDR in pesos is around the 100% level and more partial it is because of Naranja. Is there an absolute level beyond which you would rather not go and therefore, you might be willing to restrict your loan growth until deposits catch up?

In terms of macroeconomic assumptions, I will share the latest estimates from our Chief Economist for this year and next year. GDP growth is projected at 4% for this year and 3.7% for next year. Inflation is expected to end this year at 30% and next year at 18%. The foreign exchange rate is predicted to be 14.10 by the end of this year and 16.10 by the end of next year. The loan-to-deposit ratio in pesos is also relevant.

Yes, we have our liquidity coverage ratio exceeding 180%, indicating strong liquidity. Our loan-to-deposit ratio is around 99% to 100%, but we are confident that our peso deposits will continue to rise. In the third quarter and October, we experienced significant economic activity due to the elections, with pesos converting to dollars. Recently, we have noticed a trend where some individuals are selling dollars to convert back to pesos for operational needs, as they do not foresee a devaluation in the near future. We are exploring various strategies to grow our deposits and remain focused on this area, believing this will support continued growth in peso lending. We do not perceive any constraints on our growth at this point.

You don't see that as a constraint?

No.

Okay. And in terms of the dollarization, you're completely right. There has been dollarization both of loans and deposits. You and the other banks are mentioning demand for dollar loans. Should we expect, therefore, further dollarization of the banks on the asset side? And have you started to see or not yet a reduction in demand for dollars? Have you seen actual dollar sales back to pesos?

Lending in dollars is strong, with high demand in commercial lending continuing. We have seen significant growth in our dollar deposits since the implementation of the tax amnesty, and our share of dollar deposits exceeds what we would consider fair. We are leveraging this growth while adhering to our internal limits for dollar lending. Additionally, we have access to the market for dollar debt. We anticipate continued growth in dollar lending, but at a moderate pace due to liquidity limits on our dollar deposits. While the demand for dollars from our customers has decreased after the elections to normal levels, there is always ongoing interest in purchasing dollars in Argentina. This demand could rise again if any political instability arises, but we expect the situation to remain stable in the coming months. Currently, the demand levels are consistent with those seen at the beginning of the year, and we believe this trend will continue for now.

Can you give us an idea about the levels of your dollar purchases that you are seeing from your customers? I mean other banks have told us they went from $5 million or $6 million to $30 million or so per week. Where are you and where are you relative to, let's say, the second quarter or the first quarter?

We used to have around $50 million per day, and now we are at about $15 million. These are lower levels, but they are similar to what we experienced at the beginning of the year.

Our next question comes from Yuri Fernandes from JPMorgan.

I have a follow-up question. Most of my inquiries have already been addressed, but regarding asset quality, there have been many discussions about the peak. What gives you confidence that the first quarter will be the peak? In the past, we heard that the second quarter would be the peak, then the third quarter, and now we're discussing the first quarter of 2026. I understand it's challenging, but what leading indicators are you monitoring? What makes you believe there will be improvements? Are factors such as lower loan yields, economic recovery, or lessons learned in underwriting from the past year influencing your perspective? I’m trying to grasp what underpins your confidence in better asset quality. Additionally, how should your expected loss model function in this scenario? Should we anticipate lower provisions now that you are predicting improvement, or do you still need to account for some incurred losses? Please clarify the distinction between incurred losses and expected losses in your context.

When forecasting future non-performing loans, there are two factors to consider: what we can control and what is beyond our control, such as market conditions and the economy's impact on families. We always focus on our actions and the changes we anticipate. However, external factors like elections and rising interest rates can create unexpected challenges, especially in a developing country. For instance, just a month ago, we faced a potentially hyperinflationary situation that could have altered the election outcome. Currently, we are navigating through significant volatility, which is affecting interest rates and families, making predictions challenging. We base our estimates on the controllable factors within our influence. As for our lending practices, we have improved the credit scores of our borrowers, moving away from lower scores. We've also reduced lending limits in some cases and are closely monitoring performance by vintage and through our harvesting process. New loan origination appears to be performing better than older accounts. Regarding credit cards, we're noticing that longtime customers are beginning to show signs of distress due to changes in their financial situations, despite no direct actions from us. We see slight enhancements in personal loans, while credit cards still pose challenges. As we approach the third quarter, we understand the difficulty in predicting return on equity; we rely on our current assessment and recognize that while we estimate an 11% ROE, it remains an educated guess based on prevailing conditions. We've made strategic changes in our lending assessments that we believe will lead to improved performance, but unexpected economic shifts, such as those affecting household incomes, could arise. Based on current assumptions, we anticipate positive outcomes.

No, it's very clear. I understand it's challenging. Good examples. I was just trying to grasp what has changed.

We are discussing the dynamics of the models. The key point is that with new lending, you need to account for new losses or reserves based on past lending behavior. Thus, you likely won't see a significant reduction in the cost of risk initially, even if we're originating better quality loans. Reserves must be set aside based on historical portfolio performance, meaning you can't simply assume that newer borrowers will perform better right away. It takes time to show that these new loans or customers are outperforming the older ones, which is when you can start to reduce your cost of risk. This process is gradual and not immediate, which is why we anticipate higher provisions in the first quarter of next year.

Our next question comes from Santiago Petri from Franklin Templeton.

I understand you mentioned you're expecting return on equity by the mid-teens by 2027. I would like to know what loans to GDP assumption you are assuming for this achievement. Is this return on equity the sustainable steady state that you are aiming at, or are you aiming at a higher return on equity? What will be the steady state loan-to-GDP penetration in Argentina under this assumption?

Currently, loan to GDP is around 10% to 11%. Our projections suggest that this could improve by 2% each year under favorable conditions. We aim for a sustainable return on equity between 15% and 20%. With the right circumstances, we expect to reach approximately 15% by the end of next year. Starting at 15%, we could potentially reach that range by 2027, although we cannot predict Argentina's situation by then. If improvements continue and loan to GDP grows by at least 2% per year, we believe this could be achievable by 2027 and beyond. In 2027, we might still be in the mid-teens, but by 2028, it could be in the higher teens. Our long-term objective, assuming the country stabilizes and we adapt our operating model to reduce costs and compete with fintechs like Mercado Pago, is to achieve a return on equity between 15% and 20% after 2027.

Okay. I think that was the last question right?

Right, Pablo.

Okay. Well, so thank you, everybody, for attending this call. If you have any further questions, please do not hesitate to contact us. Good morning, good afternoon.

Bye-bye.

Bye-bye.

Grupo Financiero Galicia is now closed. We thank you for your participation and wish you a very good day.