Ambea AB (publ) (AMBEA) Earnings Call Transcript & Summary

Ladies and gentlemen, thank you for standing by, and welcome to the Ambea Interim Report Second Quarter 2020. [Operator Instructions] I must also advise you that this conference is being recorded today. And now I would like to hand the conference over to your speaker today, Fredrik Gren. Please go ahead.

Thank you, and good morning, everyone, and welcome to Ambea's Second Quarter 2020 Interim Report Presentation. Speaking is Fredrik Gren, CEO; and presenting with me today is our CFO, Benno Eliasson; and Jacob Persson, Ambea's Head of Group Business Control and Investor Relations. Today's presentation will cover both the effects from corona in the second quarter but also share the latest improvement in the underlying business results. As usual, I will focus on the key value drivers, growth, both organic and M&A and our effort to improve margins in the former Aleris Care operation. Then I will take you through the financials for the group, and will also describe the financial development for the segment, and we will also be more specific around corona and its likely effect on the third quarter. After the formal presentation, we open up for questions, as usual. There is no question that corona have had a major effect on our operations and our financials in the second quarter. When we reported our Q1 results, we estimated a negative effect in Q2 with sales of minus 3%, and EBITDA effect, negative SEK 40 million to SEK 60 million. The negative sales effect were in line with that forecast. But the state support program in Norway came in higher than expected, taking the EBITDA effect down to minus SEK 25 million in the second quarter. Sales is down 4% in Q2 versus last year, driven by currency effects and lower occupancy in Vardaga, but also due to few contracted units came to an end. In the second quarter, we opened up 287 new beds and placements, which only marginally impacted sales, but start-up costs had a negative EBITDA effect -- impact. Despite both corona effects, low occupancy, especially in Vardaga and significant start-up costs, we managed to improve our EBITDA margin slightly. Underlying improvements in Norway, but also in Vardaga and Nytida, where former Aleris units are delivering stronger results versus last year were key improvement drivers of our underlying results. And finally, on a personal note, I have decided to step down as CEO of Ambea. It has been a fantastic 9-year journey to bring Ambea to the strong position we now have. I feel certain that Ambea will continue to deliver high-quality care and value to our shareholders. Of our top 40 line managers, 90% has been with Ambea Group for more than 5 years, actually, most of them much longer than so. So when I leave Ambea at the end of the year, it will be in experienced hands. A few comments on corona or COVID-19. As of today, all our nursing homes are free from COVID-19. And that is thanks to both lower infection rate in society in general and the strong routines we have put in place during spring. The second quarter has certainly been a turbulent time. During the end of March and April, we saw a rapid spread of the disease in society in general and also within Ambea. The effects were significant in elderly care in Sweden, but only marginal in our other business units, Nytida, Stendi, Altiden. We tracked development, learned and gradually strengthened our routines as the weeks passed and improved our ability, both to stop the virus from coming into our homes, but also limit the number of infected residents and staff. The sharper routines, continuous usage of protective material, staff education in combination with lower infection in Swedish society in general, took down the numbers of units with infection rapidly during the second quarter. Mortality rates in our nursing homes increased during April, but has during the quarter come down to normal level, but we still suffer from reluctance to move in to nursing homes. However, during the summer, we have seen an improving trend from move in, and we believe that occupancy will gradually strengthen during Q3 and Q4. We estimated the corona effect from Q2 to be minus 3% or minus SEK 80 million to SEK 90 million on sales and approximately SEK 40 million to SEK 60 million on EBITDA. Sales effects were in line with those expectations, but the EBITDA effect came in better, around minus SEK 25 million due to a high level of state support programs than expected. For the third quarter, we are likely to see a gradual occupancy improvement in Vardaga. Already in July, we see that mortality rate has fallen below the average mortality the last 4 years. And we also see that new residents are moving in at a rate around the normal levels. The occupancy level for Vardaga will strengthen gradually, and we believe there to be a negative effect on sales amounting to SEK 70 million to SEK 80 million in Q3. And given that most state support programs has ended, and we are cautious in our estimates regarding new support program, the effect on EBITDA is likely to be larger than Q2. We estimate a negative effect on EBITDA of around SEK 50 million to SEK 60 million. Turning over to our profitability improvement program. Our major focus area during 2020 is to bring up our margin towards our financial targets through both lifting performance in former Aleris units, a broader turnaround program in Norway, and ramping up our new greenfield units. Then corona came and has taken much of management's focus and certainly had a negative effect on profitability. But during the quarter, we have seen underlying improvements on many fronts. Both Nytida and Vardaga has continued to improve margins in former Aleris units, and in Stendi Norway, we are starting to see effects from our strengthened efforts to improve profitability. In Norway, we see an impact from the overhead savings program that we launched in Q1. We also see early effects from operational excellence program, including both procurement and scheduling, and we see top line improvement both in strengthened sales and sector mix. The impact will strengthen during the second half of 2020. In the second quarter, we continued our effort for future organic growth. We opened 3 new elderly care homes and 5 disabled care units in Sweden. It is short-term challenging occupancy situation for Vardaga due to corona and we are likely to see longer ramp up times than earlier expected. Currently, 20% of our beds in operations within Vardaga has opened in the last 12 months, putting a pressure on margin. We are cautious when opening new units and only open up and staff one department or one floor at a time to ensure cost efficiency. The underlying needs for nursing homes, however, are intact, and we feel comfortable in our longer-term growth strategy. Corona has had a negative effect on mortality of nursing home residents in Sweden in Q2 but mortality rates are already back on lower or normal level in July and likely to be so in the coming quarter. So as elderly and their relatives are overcoming their reluctance to move into nursing homes, we believe demand will come back. Contract management has seen a net loss of SEK 90 million in annual revenue in the quarter, related to contract losses for Vardaga, but also a decision in Denmark where one municipality is taking back a contract as none of the tender participants met their low price expectations. We see that as a positive sign of market maturity when no competitor is taking contracts at too low-priced levels. Acquisitions. There was no acquisition activity in the quarter and only 1 acquisition in Denmark year-to-date, Vivamus, which is continuing to develop financially as planned. Our leverage is coming down closer to our financial targets, so we are likely to be more active in the M&A market going forward. More on leverage and other financials in next section. With that, over to you, Benno.

Thank you, Fredrik. This quarter is actually the first quarter where we have full, you can say, post Aleris acquisition effect in the comparable numbers since the acquisition was made in Q1 2019, which means that the quarters are now fully comparable. So if we look how the different business unit areas have affected the group numbers, we can see, next slide, that Vardaga came into the quarter with a growth pace, but is the business area that is hurt most by the corona pandemic and reported in the full quarter a decline by 2%. Nytida sales declined by SEK 12 million or 1%. The comparison versus last year is affected by both the adjustments of the capacities after the acquisition of Aleris last year as well as opening of new LSS units, but sales -- and sales is only marginally affected by the corona pandemic in Nytida. In Norway, or Stendi, the rapid weakening NOK is -- of course, in Q1 this year, has affected the reported SEK numbers. And average NOK-SEK rate is down 11% versus Q2 last year, which means that the sales in local currency is more or less flat versus last year this quarter. In Altiden in Denmark, we made 2 acquisitions around New Year, which has added SEK 42 million in sales in the quarter. And that we -- is the effect on the sales numbers we see here. And last at Klara, who is hurt by the new Swedish VAT regulation for health care services from Q3 2019, which cooled down the market for these services. So that is the effect you can see here 21% down versus last year. So turning to the profit numbers. Overall, we are pleased that we managed to maintain our EBITDA more or less at the same level as last year in these challenging times. Vardaga, if we start there, was, as said, hurt gradually through the quarter by declining occupancy in our like-for-like units as well as much slower ramp-up pace in the newer ones, and we have 20% of our nursing homes started the last 12 months. The government programs that were in place were only compensating a minor part of these effects of lower revenues and additional costs to COVID-19. The Nytida comparison is benefiting both from the capacity adjustments we made in 2019 as well as the new government programs. A large part of these programs were -- funds were allocated to the Nytida units. In Stendi, we succeeded to meet rather tough Q2 profit last year. Q2 is normally the weakest quarter in Norway due to many bank holidays. And this quarter, we were helped by a good underlying performance in our restructuring program as well as a more positive sales mix and also contribution from the government program, which were decided in late June in the parliament. Altiden is in a buildup phase, where we are improving our capacity and competence, but we are still struggling with the margins in some of our segments, especially home care. And so that will still have [ to for time ]. Klara continues to deliver stable margins. But since the top line has declined by 21%, the EBITDA in absolute numbers is slightly down versus last year. If we look at how the new reporting standard, IFRS 16 affects the margin. We have been reporting since the beginning of 2019 through this new leasing standard. This means that all reported quarterly and year-to-date data 2020 are comparable with the reported number for 2019. However, data that includes quarters from 2018, like rolling 12 days up to Q3 in 2019 is still affected then by the different reporting standards. So in this slide, you can see the effect of this. We can see that the rolling 12 margin, excluding IFRS 16, is increasing the last 2 quarters as well. Cash flow. The operating -- the strong operating cash flow continued in the quarter, SEK 484 million versus SEK 338 million Q2 last year. If we exclude IFRS 16 effect, we are -- we're at the operating cash flow at SEK 285 million versus SEK 169 million last year. There is normally an underlying seasonality in the cash flow saying that Q2 and Q4 is a bit stronger, mostly due to the release of holiday provision in Q3. But having said that, the quarterly cash flow is always hard to analyze to this due to the dependence of customer payments on the very last day of the quarter. So the long-term trend is often more relevant. So if we look at the long-term or rolling 12 cash conversion rate, we see that operating cash flow versus the EBITDA is -- on a rolling 12 basis, is now above 100%, which is good. Financing. The deleveraging of the group is continuing, of course, in line with the good cash flow. If we look back at the second and third quarter of last year, those ratios were affected by the fact that we didn't have the full year of EBITDA from the Aleris unit. So this is a little bit higher than otherwise. The net debt, excluding IFRS 16, is down from previous quarters by SEK 193 million, almost SEK 200 million, and is now slightly -- just slightly above SEK 3 billion. Net debt-to-EBITDA ratio is down 0.4x to 3.6 and is now getting closer to our financial target of 3.25x. Previous quarters, we have been financed by 50% to 70% from our Commercial Paper Program and the rest to our committed bank facility. In the -- in late Q1, the market for unrated commercial papers like ours, more or less vanished, which means that we shifted to financing to more -- the financing to more traditional bank financing. But in the very last weeks of the quarter and continuing so far into Q3, we saw a gradual recovery of the market and spreads have come down, not to the levels before corona, but to a level where we can benefit partly from this financing source again. The increase in lease debt of SEK 482 million, which you can see here, reflects the fact that we have increased the pace and start-up of units under own management, and these units come with longer rental conditions than the average portfolio. And in this quarter, we have 3 new Vardaga units and 5 new Nytida units. Yes. So if we turn into the different business areas then, starting with Vardaga. The total sales reached SEK 862 million in the quarter, which is down 2% versus last year's second quarter. This is, of course, driven mostly by lower occupancy rates in the like-for-like units, but also a decline in contract management due to ceased contracts. New units in ramp-up have, of course, affected the sale numbers versus last year positively, but not as much as expected. We see a slower ramp-up pace in these units due to the corona situation. Own management sales in total grew with 5% versus last year, reaching SEK 544 million or 63% of the share of sales in the quarter. Contract management, however, sales reached SEK 318 million, which is down from SEK 362 million last year, and that is driven more from ceased contracts and from lower occupancy. EBITDA in Vardaga reached SEK 15 million versus SEK 34 million last year. Gradually, lower occupancy rates have, of course, affected EBITDA. We saw also higher sickness rates, especially in the first part of the quarter as well as extra cost or personnel protection equipment. Some of these extra costs were compensated by the different government programs. Of course, newly started ramp up units, representing now 20% of the total Vardaga. That is, of course, also affecting the EBITDA. We saw margin improvements in the former Aleris units as we are introducing our care concept to these units, but these improvements could not compensate for the effect of the lower capacity, of course. The EBITDA margin of mature units went from 10% last year, down to 5.5% in Q2 this year. The decline is both driven from the -- from low occupancy unit-by-unit as well as we, this year, are reporting the former Aleris unit as mature. So it's different units. And this came in with lower profitability in average than the former Ambea. Over to Nytida, where total sales reached SEK 928 million in the quarter. That's down 1% versus last year. Own management were down 3%, reaching SEK 776 million in the quarter. And this is an effect of the capacity adjustments we did in mid-2019 and that we have in the last year, relatively few new startups. Contract Management sales reached SEK 152 million, which is up 25% versus last year. Strong win rates in 2019 are now clearly turning around the negative sales trends in contract management. EBITDA growth for Nytida was 20%, reaching SEK 138 million versus SEK 150 million last year. We see effects from the Aleris synergy realization, but also from taking out this overlapping capacity after the acquisition. We also have fewer start-ups in the quarter. And last but not least, we are positively affected by the government program, which reduced social security fees, which were in place in Sweden from March to June. Totally, the adjusted EBITDA margin reached 14.9% in the quarter, and that is up 2.7 percentage points versus last year. Turning to Norway and Stendi. Sales decreased 11% and reached SEK 756 million. But currency effects had a major impact in the quarter, and sales in local currency were only down by 1%. In local currency, the own management sales increased by 4%, but in SEK, they were down 7% and reached SEK 689 million. Contract management sales reached SEK 66 million versus last year's SEK 111 million, and that decline was explained by a return of a nursing home contract during late 2019. EBITDA reached SEK 31 million or a margin of 4.1% in the quarter versus last year 3.9%. Given the fact that the second quarter is usually a seasonally weak quarter, we are rather pleased with underlying profitability development. We saw a better sales mix, some effects from the cost improvement program and a positive effect from the temporary government program in May and June. The announced program to reduce the administrative cost and strengthen the operational leadership in the organization continued in Q2 with annual savings of SEK 30 million. It should be realized by year-end. We saw some delays according to plan due to the corona situation. But looking ahead, Stendi will continue with ongoing initiatives during the third quarter. And we recorded SEK 9 million in realization costs in the quarter, with total restructuring costs still expected at SEK 45 million. Over to Denmark and Altiden. Sales amounted to SEK 170 million, which were up 32% versus last year, explained by the 2 acquisitions made around Christmas. The acquisitions, Vivamus and Casablanca, performed in line with expectations and -- in the second quarter and contributed positively on the profitability. In this seasonally week second quarter, Altiden had an EBITDA of minus SEK 7 million or an EBITDA margin of minus 4.1%. This quarter, we have to write-down a loss-making contract in the home care business, which were prolonged by the municipality, and this affected EBITDA by SEK 4 million. We have continued to invest in overhead, both for building up our support organization as an independent company after a carve-out from former Aleris organization in Denmark, and we have also strengthened both disabled care management and resources to support organic growth. We plan to continue our strategy to grow in more profitable segments of disabled care and own managed nursing homes. And last to Klara, we saw the total sales were down 21%, reaching SEK 60 million in the quarter, exactly the same number in the first quarter. Sales decline is predominantly in the staffing business towards private operators, and that's impacted by the changed VAT regulation introduced last summer. We continue to grow our Klara team services with favorable mix development and administrative savings. We continue to improve margin in the quarter, reaching 8.3% instead of last year's 7.9%. And with that, back to you, Fredrik?

Yes. Okay. To sum up our financial development, our growth target is 8% to 10% through a combination of acquired and organic growth. Given the size of Aleris Care acquisition, we exceeded that growth target in 2019. 2020 is likely to come in below as -- due to both corona currency effects and less M&A activities. Year-to-date, sales increased with 4%. Profitability-wise, we have a mid-term adjusted EBITDA target of 9.5%. The Aleris acquisition, of course, diluted our margin, and during the quarter, we had -- despite negative corona effect, seeing a small improvement in our rolling 12 EBITDA margin. And finally, regarding leverage, cash flow has been very strong in the last quarters, and our leverage has come down quicker than expected, reaching 3.6 in the quarter, close to our financial target of 3.25. So summarizing the second quarter, turbulent quarter with focus on handling the corona outbreak with major effects on Ambea operationally, especially in Vardaga. As I said, the infection spread quickly to many of our nursing homes in Sweden, but has during the quarter, come down significantly. And currently, all nursing homes are free from COVID-19. In Q2, we had a negative corona effect on EBITDA of approximately SEK 25 million. The negative effect on profitability is likely to increase during Q3, and we are -- estimate it to be minus SEK 50 million to SEK 60 million due to lower occupancy in Vardaga and less government support programs to mitigate these losses. Mortality rates are coming back to normal level, even below earlier years as we move into Q3, and the long-term care needs in society are intact. Ambea continues with growth activities in line with our strategy and opened several new homes in Sweden. Short term, we are -- we expect a challenging occupancy situation in Vardaga and we are likely to see longer ramp up times than earlier expected. Our Norwegian management team are doing a good job with the turnaround program, and we see some improvements already in Q2 but are likely to see continued positive trend in the second half of 2020. And finally, during the quarter, I decided to leave Ambea. So at the end of 2020, I will hand over the Ambea leadership to our experienced management team. The search process for a new CEO is ongoing. So with that, I conclude our presentation and open up for questions. Operator, please, could we have the first question?

[Operator Instructions] The first question is coming from the line of a Kristofer Liljeberg from Carnegie.

Okay. I have three questions. The first relates to occupancy rates in the Swedish nursing homes, if you could explain how much that is down this year and the assumptions for the third quarter? And related to that also, you seemed pretty upbeat about the current move-in rate. So with this level of old people moving in, when will you be back at normal occupancy rates again? My second question relates to the improvement in Norway. Is it possible in some way to quantify how much of this is underlying improvements versus the mix effects and also the state support program in the second quarter?

If I start with the occupancy rate. The drop in occupancy in the second quarter were around 10 percentage points. So that happened very rapidly in April, May and June. And what we saw in July was a situation where our mortality rate has been on average in the last 4 years, 2.9%, and that went down to 2.5%. We hope that, that will be a trend that continues, but that's July. And the move-ins came back to -- they were basically around -- the number in April, May, June was around half the normal rate of move-ins that we have but the move-in rate in July were back on a more of a sort of pre corona level. We believe that -- I mean, of course, it will not go back to normal as fast as we lost it. But we believe that September and October and onward is likely to gradually improve. We see that some sort of in -- some larger players in Europe in our industry are expecting us to be back at Christmas time. I believe it will take slightly longer and likely to move into 2021, early 2021 before we are back to normal level. That's our best estimate right now. And regarding Norway, there are a lot of different moving parts. And of course, adding to the ones you said is, of course, the currency effect. When we report a similar level of profitability in the quarter, it has a significant effect that the currency is -- Norwegian Krone has changed as well. So it is a little bit of all. I'm not sure we can quantify each one of them. But what we can say is that the effect on the overhead program has started to give some impact in Q2, but most of that will actually be seen not until Q3. And the same with the others. So I believe that there will be a larger impact in Q3 than what we see in Q2. I know that's not a full answer that you like but I think...

Fredrik, if you -- could you just -- the state support, how much was that in Norway in the quarter? If we just take that out of the equation.

But that also was sort of -- the state support program also came with extra cost because of much higher sick leave and costs for protective material. So the net effect of that were not that big in the quarter actually. So most of the effect is coming on better performance. Was there a third question there Kristofer? Or maybe that...

I think I'll stop there.

Your next question comes now from the line of Danske Bank.

So just one question for me. Looking at your -- yes, can you hear me?

Yes, we hear you well.

Okay. So just one question for me. Looking at Nytida which had a very strong development year-over-year. Is it possible to split out how much of the increase in EBITDA is due to realized synergies and how much is government support? And what do you think about the outlook for that area during the rest of the year?

Again, we probably can't really split the exact numbers because there's a mix of state support increased sick leave costs, protective material and some underlying improvement. But it's clear that the improvement rate that we have seen in the last 3 or 4 quarters in Nytida, much of that has been driven by the fact that we are seeing synergies from the Aleris acquisition, but also improving profitability of some Aleris units. We are now in Q3, and onwards, we are meeting quarters where many -- much of that effect has already sort of been captured. So you should not expect the Nytida margin improvement to continue going forward. It's rather to be sort of meeting the same levels.

Okay. But is it fair to say that the majority of the margin improvement year-over-year in Nytida is due to synergies and operational improvements?

Up until the second quarter, yes. In second quarter, there are some effect also on the state support.

From the line of Thomas Graf from Handelsbanken.

Yes. I hope everyone can hear me. I was just wondering about -- well, it's been turbulent times and so on. But if we look at -- you mentioned infection rates at the homes are now 0. If there should be a second wave and so on, are you confident -- how is your status there? What do you think about the possibility of the infection rate being none in the nursing house going forward. Are you worried about that? Could you give some status there? Would be great.

I think that much has changed since April, where we saw large rates of infection. First of all, the general infection rate in society is, of course, much lower. But I think in society in general, you are going to see -- continue to see small outbreaks. I think we are much better in sort of early identification of any infection and the fact that we do much more frequent testing makes it easier for us to -- strong routines makes it less likely that the infection will come into our nursing homes. But if they do, and unfortunately, that is likely to happen at some point, I think we are much faster in capturing the disease and making sure it doesn't spread wider. So it's now been many weeks since we had a last infected person. But I think also in June, we saw that we were able to stop infection very -- much more effectively than we were in early April. So I hope I can say that 0 will be the number also for the coming months. I think it's likely that we're going to see some instances, but much fewer than we've seen before.

Okay. And regarding the government support going forward, are you expecting any sort of government support to come in as in Q3 or Q4 because you have asked for support in all costs regarding protective materials and so on. Do you think any of that will -- of the support will come -- will come later on, so to say, any expectations in that for the government grant?

I think in our estimate, also in Q2, we always try to be very kind of cautious regarding state support program. So in our estimate for Q3, we are -- we only have a small number of state support program. Regarding the municipal program where we are asking for compensation for the increased cost of protective material, et cetera. That is not included in the negative effect. If we get money from those, that will have a positive impact versus our estimates. However, we are working with the municipalities to give them the kind of underlying material and all those, but we like to see that happening before we put it into our estimates. So there is an upside if we get more money from municipal.

And when do you expect -- if that comes, when do you expect those to come into the numbers, if you would get the state support?

I think that there's one payment term now in Q3, and there's another one in Q4. So we are likely to get feedback on that during the third quarter. And then we will get that into the third quarter result.

The next question comes from the line of Karl Bonnevier.

Karl-Johan Bonnevier. Just -- Fredrik, to start off to get a little better feel for your guidance for Q2 -- sorry, for Q3. When you look at the impact on the EBITDA line of minus SEK 50 million to SEK 60 million, and I understand from your previous answer, you are not hoping for much of state contributions compared to Q2. But then when you look at -- now you're coming into a slightly more aggressive opening up schedule for the Vardaga going into the second half, how much of that negative, so to say, is coming out of your expectation of those units ramping slower and how much is coming out of, say, the mature like-for-like base, if you put it like that? How is your thinking?

No, I think that the SEK 50 million to SEK 60 million is -- and we're trying to do -- it's our best estimate. And of course, the major impact is regarding how fast can we expect occupancy to pick up. So in that number, we have included both impact from sort of our existing units that are now -- that have seen sort of a lower occupancy than we have had for many years. That they've been very stable before they've been down, but also the fact that ramp-up times are likely to be slightly slower so it's a combination of both slower ramp-up and sort of lower occupancy on kind of existing units. The both of them are impacting -- adding up to the SEK 50 million to SEK 60 million.

And when you look at that -- when you now then are experiencing existing units having lower occupancy, is there anything you can do short-term to mitigate the cost impact of that? Or should we see your, say, cost picture basically as being fixed in the short term, and that's why you get this huge impact from a margin conversion perspective?

I think that we are -- what we are doing is, of course, to be very cautious on adding personnel costs to our -- I mean, when we get the keys and have to start paying rent, that cost is hard to do anything about. But we can be -- we can, of course, be cautious in hiring managers and staff. And making sure that those costs are kept at a minimum until we see occupancy picking up. So there are some things that we -- can impact, absolutely.

But I guess that must relate more to the new openings. What can you do in, say, the existing mature units to balance?

We can do a little bit there as well. And in some instances, we have been able to shut down 1 department. So with that, you can sort of move personnel or lower personnel cost also in existing units. So there are some ways to mitigate. We cannot mitigate to 100%, but we can do some mitigation on lower occupancy.

And when you're looking at new openings, I guess you already contracted the units that are in pipeline for the second half. So -- and as you put it, when you get the key, you just need to start to pay the rents. Is there anything you can do to, say, slide those kind of time schedules in this kind of environment?

Yes. We are in dialogue with many of our sort of property partners to see if there is a way to delay some of the startups. It's only a handful of them that is relevant for that kind of measures, but that -- and some of that has been sort of positive, that's where we've been able to push it back a couple of quarters. So a little bit of that is actually being in progress.

And just also on the Vardaga mature margins, those 550 basis points, could you help us see or understand what was the impact? So if you take the like-for-like old Ambea units, is there some sort of feel for what kind of margin headwind you saw in the quarter for those so we can understand the, say, the number impact of actually including Aleris here and the, say, the real-world challenge, if you put it like that?

Yes. And I don't have the exactly split on that. But what you also have is we have seen clear improvements on the underlying kind of cost KPIs from former Aleris unit. So -- Yes, so you have sort of a big negative, which is occupancy driven. Another one, which is sort of a lower profitability level of Aleris. And the third one, we're actually seeing some improvement versus last year. But I don't really have the split on that, sorry, for new one.

No problems, no problems. I can come back on that. And finally, Benno, when you look at the cash flow in the quarter, you obviously benefited from a good working capital release. Is there a lot of temporary things in that with delayed payment schedules for social security benefits and these kind of things that we should, say, expect to come back and haunt to you in the second half of the year.

There is some changed payment pattern from the customers. We see some municipalities are paying us earlier than previous. And we also have some of the government programs saying that we can pay in taxes later than otherwise. The total effect of this one, I don't have it exact, but there is some effects of that. And that will -- of course, it's hard to say if that will keep during the third and fourth quarter this year. We know that some of the government programs in Norway, for example, is over the year-end. But different -- the new payment pattern, if you call it like that, that the municipalities pay invoices before due date, that is -- hard to say if that is going to stay or not. But yes, we have some impact of -- the positive impact of COVID-19 on cash flow, for sure.

But if you would try to, say, give us a rough estimate is, are we talking say, rather SEK 20 million than SEK 100 million? Or what kind of measure -- what kind of level would you expect?

I would say something in between SEK 20 million and SEK 100 million, I would say.

And the next question comes from the line of Klas Pyk from Nordea.

Most of it have already been answered but I'll ask two. One very specific, looking back at your Q3 2019 report, you said that the seasonal effect of around SEK 40 million to SEK 45 million in Stendi. That is Q3 over Q2. And given that you are continuing to improve efficiency in Stendi, would you say that, that sequential or seasonal effect to be higher this year? Perhaps if you could split out how much of the annual SEK 30 million will be realized in Q3 compared to Q2? My second question is a bit broader one. Just given the micro environment -- macro environment with weak financial amongst municipalities, while at the same time -- politicians are speaking about improving quality in elderly care. Can you [Audio Gap] since you have had with the municipalities in terms of price pressure versus increased quality, et cetera?

Yes. Well, I'll start a little bit with the second question. I think that our dialogue with municipalities, I mean has lot of things happening here. I mean, first of all, we see a lower activity in -- or a dialogue regarding contract management or regarding new, that's been slower in the second quarter, of course, because of sort of other things or just that the politician has been busy with the whole corona situation. But we see indication that, that interest is taking up that dialogue is coming back now after summer. So there's clearly sort of appetite to kind of take-up the work on the old issues, so to say. So we think that, that dialogue will continue. Regarding the pressure on financials, I think that's -- there's been a very clear signal from our finance minister that all extra costs related to corona will be covered. So I think that many municipalities feel comfortable that they will not have sort of a short-term hit in -- from the corona. And regarding sort of the years ahead, that brings us back to close to a situation where we were 6 months ago with a higher number of elderly needing care and the difficulty of sort of balancing budgets in municipalities due to that. So I think that we will see that dialogue coming back to sort of the pre corona situation during fall. And I'm not sure, Benno, if you have a good answer on the first question on the...

Yes. Could you repeat the first question, please?

Yes. Of course. So I just look back at your Q3 2019 report and there you see the seasonal effect of around SEK 40 million to SEK 45 million in Stendi Q3 over Q2. And just to get a feel for how these operational efficiencies in Stendi are realized, could you just give a guidance on how much you expect that seasonal effect to be this year compared to the previous year?

Normally, we have these kind of seasonal effects. But now we also have the corona effect in Q2 this year, which has also affected the numbers. In Q2. And the underlying program is, of course, starting to benefit a little bit more during Q3. I don't want to give an exact -- how much that will affect the Q3. But I think that the seasonal effect is there every year. And the other things is going to gradually affect the numbers positively from more and more during Q3. So the seasonal effects are more lower staff costs due to vacation and like that. And if we see the same kind of demand from customers in Q3 versus Q2, then that staff cost will, of course, be the -- positively affect the EBITDA.

[Operator Instructions] And we have one more question from the line of Kristofer Liljeberg from Carnegie.

I was actually trying to remove that one. So no one more for me.

At the moment, there are no further questions, please continue.

Okay. So no more questions. Thank you all for calling in. The Q3 interim report will be published on November 5. And with that, have a nice day, everyone.

That does conclude our conference for today. Thank you for participating. You may all disconnect.