I'd just like to mention that earlier this morning, we announced that we are addressing the bond market to refinance our outstanding bond, ACR02, maturing in January 2024. More details will be shared with the market in due time. This presentation will focus on our Q2 report and is, as always, divided into four parts. First, I will take you through the highlights of the quarter. Our CFO, Nina Mortensen, will present Q2 financials before we give an updated outlook and round off with a Q&A session. You may ask questions live after the presentation or through the available chat function. Now, let us move to slide three and have a look at the second quarter highlights. In our opinion, Axactor delivered a solid second quarter in what is normally also a seasonally strong quarter.
Gross revenue grew 5% year-over-year, and measured in constant currency, the growth was 8%. The increase in gross revenue is achieved despite macroeconomic headwinds in Germany and in the Nordics. Cash EBITDA was up 4% year-over-year, and in constant currency, 6%. The EBITDA margin has stabilized at a high level, reaching 50% for the quarter. Actual EBITDA ended at EUR 33 million, up from EUR 30 million same quarter last year. Again, despite somewhat more challenging collection environment in certain countries. We delivered double-digit ROA of 11%, despite the steep increase in cost of funding that both we and the rest of the industry are experiencing.
All in all, we are satisfied with the Q2 key financials, and we believe that this is yet another confirmation of our stable performance that we have seen since the start of 2022. Please move to slide 4 for a short recap of our strategic direction. As mentioned several times before, we stick to our successful strategy that we developed in 2020, which consists of the following three pillars. Axactor has a strong focus on doing accretive portfolio investments, and as we have seen over the last couple of years, we have done so successfully. Axactor's single most important profit improvement initiative is to buy portfolios at a satisfying Gross IRR level. Secondly, we always seek to improve our cost position. Axactor was incepted to disrupt the industry on Cost to Collect, and we have obtained a superior cost position.
A natural result of this is that our EBITDA margin is high compared to peers, especially when you consider that we have a meaningful share of 3PC business that has a lower EBITDA margin than the NPL business. Thirdly, we are pursuing a niche strategy in terms of industry, segments, and markets. Priority is on the bank and finance segment, which counts for the vast majority of our business. We have chosen the six markets in Europe we believe to be the most attractive over time. All of our markets are mature in terms of NPL transactions, has a stable legal and political environment, and has proven to provide attractive returns over time. Axactor has shown a healthy ERC growth since year-end 2021. Please move to the next slide for a graphic overview of our ERC development.
With increasing investment levels, as we saw from 2021 to 2022, our ERC will grow as a natural consequence. Q2 showed a continuation of the growth we have seen throughout 2022. Since Q4 2021, Axactor's ERC has grown by 20%. For us, ERC growth through CapEx deployment is not the most important KPI. However, when the investments are done at the attractive Gross IRR levels we have achieved over the last two years, we are very pleased to see that the ERC continue to grow. Now, let us move to next slide, where we will give a more detail on our accretive investments in NPL portfolios. In two years, we have managed to increase the Gross IRR backbook by 1.9 percentage points. It might seem moderate, but actually, it's not.
Please bear in mind that the one percentage points increase in the total Gross IRR equals approximately two percentage points improvement in Return on Equity. In the Q1 presentation, we reported that the last 12 months' investment had been done at an attractive level of 22.2% Gross IRR, which was 4.6 percentage points higher than our average NPL book. If we look at the most recent transactions from the beginning of December and throughout the first half, the Gross IRR has gone dramatically up to 35%, which is approximately twice as high as our total NPL book.
The reason for this is mixed, but the imbalance between supply and demand for NPL volumes is currently substantial, and due to a more challenging funding situation for the industry, combined with a more uncertain collection environment, there are no short-term triggers to change this imbalance. It is true that the volumes are still relatively low, but Axactor has invested EUR 56 million at this level, and sellers are working to absorb the new price level. The speed of the Gross IRR blending will obviously be influenced by the annual CapEx level. Please move to the next page for more details on the matter. As announced earlier this year, we expect to invest between EUR 100 million-EUR 150 million in 2023.
Our replacement CapEx is estimated to be approximately EUR 114 million, the investment level does imply a constant or moderately increasing NPL book by year-end, depending on where in the interval we end up. NPL investments for Q2 reached EUR 40 million, investments in the first half ended at EUR 73 million. We have committed another EUR 20 million for the rest of the year, we have already EUR 93 million of investment plan for this year. Depending on the price level going forward, in combination with access to funding, Axactor will be moderate in terms of new portfolio investments. Please move to the next page for a short comment on our cost development. Q2 was another confirmation that Axactor continues to have one of the best cost positions in the industry, particularly within NPL Cost to Collect.
NPL Cost to Collect year to date was 38%, which is marginally lower than for 2022. It is true that the Cost to Collect level can vary a bit between quarters, but the trend is important for us to underline. Axactor's strong cost discipline is now materializing in an EBITDA percentage that is among the best in the industry, 50% for the quarter. We believe there is still improvement potential as we continue to do accretive portfolio investments, which will lead to further margin expansion over time. Additional scale effects and more data-driven operations will also be supportive to the ambition of improved margins. I will now leave the word to Nina for a financial update, starting on page 10.
Thank you, Johnny. Let's start with the development in gross revenue, where we also this quarter continue to see a healthy growth with an increase of 5% compared to the same period in 2022. Adjusted for NOK/SEK currency effects versus the euro, the estimated gross revenue growth was 8%. The NPL segment reported a growth of 8% this quarter, supported by a good collection performance and a solid investment level in 2022. The 3P C segment had a more challenging quarter and experienced a decline of 10%. Let's look a bit more into details on each of the business segments, starting with NPL on the next slide. Total income for NPL segment ended at EUR 52 million in Q2, up from EUR 46 million in the second quarter of 2022, with a satisfying growth in total income of 13%.
The contribution margin for the segment was at 76%, the same level as the previous quarter. The collection performance ended at 102% for the quarter, largely driven by strong collection performance in Italy and Spain. Debtors in the Nordics and Germany continued to opt for longer payment plans with lower monthly installments. Bailiffs are offering debtors more flexible terms due to higher living costs. This includes higher reservation amounts and payment-free months in several markets. On the more positive side, we see that continued strict cost control in all countries helps to secure a healthy margin. We expect the NPL segment to continue to see revenue growth, given the high investment level last year. Please turn to the next slide for comments on the development in the Triple C segment. The Triple C segment has been under pressure from market competition.
This quarter, Triple C has also been negatively affected by delayed implementations of new customers in the Spanish market. Axactor has initiated several targeted measures to address this. The company is currently going through all Triple C contracts and will terminate those with lower than acceptable margins. One of the main short-term initiatives is the company's decision to exit the Triple C segment in Sweden this coming fall. The Triple C revenues ended at EUR 13 million for the quarter, 10% below Q2 last year. The contribution margin ended at 33%. Let us move on to the next slide, where I'll present more details on the reported financials. Total income at group level ended at EUR 65 million in Q2, up from EUR 60 million in Q2 2022. Q2 is a seasonally strong quarter.
The total income level in Q2 was especially supported by good growth and strong collection performance in the NPL segment. The reported EBITDA came in at EUR 33 million, corresponding to a healthy EBITDA margin of 50%. The EBITDA margin is supported by good cost control in all countries. Cash EBITDA ended at EUR 60 million for the quarter, compared to EUR 58 million for the same quarter in 2022, equal to a growth of 4%. In constant currency, the estimated growth was 6% in Q2. Moving on from reported Cash EBITDA to some comments on the funding situation on the next slide. We are pleased to see that the renewed RCF was officially in place by the end of June. The facility has been structured with satisfactory terms and a three-year maturity, with the option to extend the maturity by an additional two years.
As Johnny was saying, we announced earlier today the plan to refinance the ACR02 bond, maturing in January 2024. The company has repurchased EUR 44 million out of the original EUR 200 million outstanding loan. Axactor has also repurchased an additional EUR 19 million of the ACR03. The ACR03 bond is maturing in September 2026. Let's move on to the next slide and the development in Return on Equity. We are pleased to see that Return on Equity for continuing operations for the last 12 months is stable at 10%. This underpins the consistently solid financial results in 2022 and in the first half of 2023. The Return on Equity ended at 11% for the quarter. I'll now hand it back to Johnny for an updated outlook.
Thank you so much, Nina. In our updated outlook, we have four items that we would like to emphasize. Firstly, we expect to continue to do accretive investments on NPL deals with a minimum of 30% Gross IRR compared to the back book, which is at close to 18%. Secondly, we expect a minimum growth of 10% on interest income for 2023. On 3PC, we see competition puts pressure on both total income and Contribution Margin, as more players are focusing stronger on 3PC due to reduced investment capacity. Thirdly, regarding funding, we have an interest hedge that secures partial protection of financial expenses for 10 more quarters. As you saw earlier today, we are currently in the market to refinance, ACR02. Regarding collections, we expect continued negative macroeconomic impact in the Nordics and in Germany.
Our collection curves are adjusted to reflect the current macroeconomic conditions. With that, I suggest that we open up for questions.
Thank you. If you would like to ask a question, please press star, then one on your telephone keypads. If you do change your mind, please press star two to remove your request to speak. As a reminder, that is star followed by one to ask any questions today. We have our first question on the line from Jan Erik Gjerland of ABG. Please go ahead when you're ready.
Hello. Yes, Jan Erik Gjerland from ABG here. You have previously said that 3PC connection is very interlinked with the NPLs, and buying into the different markets. Now, you're seeing margin pressure, et cetera, in the 3PC market. How do you think the connection is for your Swedish operation, and how it's important is it for you to continue with the 3PC in each market to make sure that you have the NPL input for future participant? Thank you.
Yes. Thank you, Jan Erik. First of all, I think, if, if you look at the total 3PC, remember that we still have the vast majority is in Spain, and we are still continuing to deliver solid 3PC performance in Spain, and in Italy, we're actually growing the 3PC part quite substantially. When it comes to the Nordics, I think we just have to admit that firstly, we have a subscale operation, and we have lost money in Sweden in particular. We have access to a lot of data already, and what we said when we decided to really go for 3PC in Sweden was that we the strategy was to have it on 3PC and then try to buy portfolios.
Well, now, after three years, we, we, we know the answer, and we have not actually been able to buy any portfolios from the partners that we have been doing 3PC with, because the market prices in Sweden have been far off compared to, to what we have been willing to pay. You could say that, well, okay, maybe we have saved ourselves for some money because we have had the, had the good, good data, but I think we are now ready to, to, to try to bid on portfolios without actually having 3PC. You are, you are pointing at something that we. It might be that we will have some weaker data for valuation, but it's the, basically the same data that all other players are using when they are doing valuation of the portfolio.
It will only be the one with 3PC that has a small advantage, but also the sellers have been much better at providing data in the sales processes. We will get a lot of collection data every time we go into a process. The, the part that we are missing out is that by having it on 3PC, you could, you will know if there's an up or downside to your own collection curve. I think for us now, what we are saying that we will prioritize profitability, and if we are not able to do 3PC in a profitable way, we will actually not do it in the country.
Okay, that's perfect. Thank you for that. Secondly, on, on the Gross IRR, which is, was 34.5 on average. I look into your different kind of markets when in your appendix here, and seeing that you have a broad base of investment during the first half of this year. Is it so that, could you give any sort of range for this? Is this, is it all around the 35%, roughly, or is it so that some, some areas are 15, and some areas are 50% Gross IRR, or how should we read it?
Yeah. Without being too specific about it, I would say that Spain and Italy offers definitely the highest gross IRR, and that it's two explanations for it. In Spain, we have invested in some secured portfolio, where the legal costs are a little bit higher than for unsecured collection, and that will naturally drag up the gross IRR somewhat, but that is not the main reason for the steep increase. That is one part. Also, generally, collection costs in Southern Europe is a little bit higher than in the Nordics, where you have a very efficient bailiff system and so on.
If you look down to net IRR, I mean, not reporting on net IRR, but we see that the net IRRs in the markets are still, it's still a little bit more attractive in, in the south, but we see that the Nordic prices and the German prices are also adjusting. Gross IRR will be a little bit less for the Nordics and Germany because of somewhat lower costs in collection.
Okay. Thank you. Finally, from me, my side, the leverage ratio of 2023, your target, is that still 3.5, or, or how should we think about that after you sort of jumped now to 3.8, was it? Yes.
No, it still stands. It does, and we have the opportunity to relatively quickly adjust the leverage ratio because we could just stop investing. As you have seen, we have very few forward flow agreements left. Hopefully we will reach the 3.5, but it will always be when you see the Gross IRRs touching 35%, I have to admit, it's always tempting to look into new deals. Since the covenant is four, we have an internal target, and we have 3.5, and now we are at 3.8. We will stand for the target of 3.5, and let's see if we can get all the way down.
That is still the, that is still our, our goal, yes.
Okay. Just one follow-up then, on the covenant side. You think in the fixed income information you gave this morning, that you would do some changes to the covenants. Was it for the bonds that is outstanding? Could you just shed some light into how and why, and how much you pay for it?
Yeah. This is for us, this is pretty undramatic. We have a covenant in the ACR03 on the interest coverage ratio. Because of the general increase in the interest level, we are, of course, moving towards that covenant. I think both us and others with the same covenant structures needs to address this. We are suggesting to take down, to change the covenant from four to three. On top of that, we are also changing another covenant, which is which is secured loan-to-value covenant, and that we are changing in the other direction. Actually providing more security for the bondholders, taking it down from 65 to 60.
We have offered what we think is a fair price of 25 bips for this for this amendment. We don't know the final result, of course, but we think this is undramatic. It has nothing to do with the actual business. It's just that we need to adjust the covenant structure to the today's interest levels, I would say.
Thank you much for your answers. That's all from my side.
Thank you.
Thank you. We now have Håkon Reistad Fure of Arctic Securities.
Good morning, thank you for taking my question. I was wondering about the net financial items. You had a positive gain from the refinancing of the RCF. Can you elaborate a little bit on what caused the gain? Also in relation to your hedging arrangement, does the refinancing change any of this?
Yes, I can elaborate a little bit around the net financial items, and you are correct that we had a positive gain this quarter of NOK 1.9 million relating to the refinancing of the RCF. This is in line with how we do accounting for our press, and this is accounted as not a new RCF, but as more extension of the previous one. In this respect, the NOK 1.9 is related to also improved terms on the new RCF. You also had a question also on the hedging arrangements. We still have in place a one-year hedge that is due end of this year. We have a hedging ratio of about 60%.
Also, as we stated in our report, we have a timing difference between when we take the cash impact and the PNL impact from the hedging. The PNL impact is taking over the previous 3 years duration of the hedge, while the cash impact is taking this year. I think as you can see from the report, that the cash impact is a little bit higher than the PNL impact, and this we will also continue to do then for the rest of the year, and we will also have the partial coverage in the PNL over the two next years. We can also enter in the new hedging arrangements this year.
Thank you very much. One more question from me, regarding REO assets, classified as held for sale. Just wondering if, if you expect to, to sell these assets, when during this year, and if you expect to, to be able to sell them at, book value, given the, the macro environment?
Yes. On the REO part, we have end of this quarter, NOK 4.6 million remaining, so it's getting quite small. We are continuing to sell the assets, and we expect to do so by the end of this year. As you also can see from the PNL, we need to give some discount, but some of the assets also still been sold close to the book value. This segment is expected to be finally closed by year-end, so we will not have discontinued operations going into 2024.
Okay, thank you. That's all for me.
Thank you. As a reminder, that is star followed by one to ask any further questions today on the phone lines. I can confirm we currently have no questions registered, so I'd like to hand it back to Johnny for any webcast questions.
Thank you very much. We have a few questions here. The, the first one is, can you please elaborate on the risk profile and Cost to Collect on new NPL investments, re the increasing Gross IRRs? Yes, I could, I could at least give it a try. The Gross IRR, the, the steep increase is, of course, a result of, the demand and supply in the market.
Given a bit restricted access to funding, we are, of course, pricing differently, but the pure technical way of doing it, when we look at the curves, is that we, we do, we do the curve with the best estimates that we have, taking into consideration all the new elements, for example, changes in the way the bailiffs are working in the Nordics and so on. Then, of course, we have taken up the requirements, the IRR requirements because of the new funding costs. You can say that there are several elements to the increase. It's, first of all, I think the industry has been overpaying in general for portfolios. If you look at the, especially the period from 15 to maybe 19 and 20.
The IRR has been too low, so it needed a base adjustment in any case. We need to adjust the IRR because of the higher funding costs. On top of that, we need to have a wider risk element, due to risk in collection, basically. As you know, there's no price list to this when you buy, bid for a portfolio. Normally, there are several bidders, and sometimes you win, and sometimes you lose. It's, it's... I think, I don't know if I am able to elaborate much more than that, actually.
I have said that the level of 35%, I don't think that is a necessarily a sustainable level when you start to deploy a lot of CapEx, but right now, this is the level we see. It might be that it will go a little bit down when you start deploying more, say, more sizable CapEx. As I also said in the presentation, right now, there's no triggers to change the imbalance between demand and supply. Next question is, what is your replacement CapEx at the moment? I think I also mentioned it in the presentation. I think it's still NOK 114 million. Or if it's, or it is very close to it.
It might be that it's plus minus NOK 1 million or 2, but that is the last calculation that we have done and that we have revealed to the market. We have one question, saying, you are getting close to bond doc leverage and interest cover covenants. How do you plan around this? Would contemplated bond include same covenant? Yes, the new bond will be not the same covenant as in as in the ACR03 that we adjust. We, we have a new set of covenants that is the same as we are trying to amend in the ACR03. I think I also addressed this earlier in in the Q&A, I, I hope this one is covered.
We have one question here: Can you please comment on the EUR 3 billion NPL portfolio acquisition from Sareb reported in July? How will this be funded, and how does it align with your full year NPL acquisition target, given the reported EUR 150 million deal size? There's a lot of things to clean up. First of all, no one has reported a EUR 150 million million deal size. I've seen it in a Spanish newspaper, but I can tell you it's far, far, far from reality. This is an old portfolio that we have acquired, and we don't give out the exact CapEx level, but it's included in the full year forecast for investments.
For this deal, in particular, I would just, I will suggest that you just forget about everything you read in the newspapers because it's completely wrong. It is an attractive portfolio, but it's much smaller in terms of CapEx, of course. Other than that, we, we don't want to comment more on that portfolio specifically. That was all the questions that we have received. If there are no further questions, I wish all of you a great day. Thank you for participating.
Thank you all for joining. I can confirm that does conclude today's call. Please have a lovely rest of your day, and you may now disconnect your line.