Azul SA (AZUL) 2020 Q2 法說會逐字稿

Hello, everyone, and welcome to Azul's Second Quarter 2020 Results Conference Call.

My name is Greg, and I will be your operator for today.

This event is being recorded.

(Operator Instructions)

I would like to turn the presentation over to Andrea Bottcher, Investor Relations Manager.

Please proceed.

Thank you, Greg, and welcome all to Azul's second quarter earnings call.

The results that we announced this morning, the audio of this call and the slides that we'll reference are available on our IR website.

Presenting today will be David Neeleman, Azul's Founder and Chairman; and John Rodgerson, CEO.

Alex Malfitani, our CFO; and Abhi Shah, our Chief Revenue Officer, are also here for the Q&A session.

Before I turn the call over to David, I'd like to caution you regarding our forward-looking statements.

Any matters discussed today that are not historical facts, particularly comments regarding the company's future plans, objectives and expected performance, constitute forward-looking statements.

These statements are based on a range of assumptions that the company believes are reasonable but are subject to uncertainties and risks that are discussed in detail in our CVM and SEC filings.

With that, I'll turn the call over to David.

David?

Thank you, Andrea.

Hello, everybody.

Thank you for joining us on our second quarter earnings call.

The second quarter was, without a doubt, the most challenging time in aviation history, and our team has done an incredible job taking care of our customers, managing our costs and, most importantly, preserving our liquidity.

I'm very thankful for the sacrifices they are making for Azul.

Our mission is to provide our crewmembers the best jobs of their lives and our customers with the best flight of their lives.

I'm happy to report that last month, Azul was elected the Best Airline in the World in the 2020 TripAdvisor Travelers' Choice Awards.

This is an incredible achievement and a strong testament to our crewmembers' dedication and passion in taking care of our customers every -- on every single flight on every single day.

As you can see from Slide 3, we are on top of some great airlines in the world.

Moving on to the next slide.

On the operational side, we have implemented a lot of actions to guarantee the safety and sustainability of our business.

We are the first airline in Brazil to introduce daily temperature checks for all of our crewmembers and a mandatory mask requirement for customers and crewmembers on every flight.

Our jets are fitted with state-of-the-art HEPA filters that remove at least 99% of the airborne particles.

We have also implemented an innovative boarding process called Tapete Azul or blue carpet, where projectors create a moving carpet image on the floor, guiding our customers to board when their specific seats are called.

On average, this process reduces boarding times by 25% while increasing NPS by 21%.

By the end of the year, we expect to have this system adopted on 70% of our flights.

Moving on to Slide 5. We're rebuilding our network.

We operated at only 70 daily flights in April.

But in September, we expect to operate over 400.

Azul's fleet flexibility is like no other airline in Brazil or probably the world.

We have aircraft ranging from 9 seats to 214 seats, which allows us to customize our network according to the demand using the right aircraft for the right flight.

During times of uncertain demand, the best thing to do is to bring in that demand at the lowest possible trip cost.

This is exactly what we are doing as we rebuild the network step by step.

As demand recovers, we will have more options to create our network faster and at a lower cost than airlines that fly a larger single-type aircraft.

It is simply not economical to fly a large plane in markets that don't have enough revenue to cover the trip cost.

Abhi is building schedules on a weekly basis, which allows him to constantly test demand, swap out aircraft at different times.

So if we can fill an A320, we fly an A320.

We can only fill an Embraer, we fly that.

If we only have an ATR's worth of demand, that's what we fly.

Another way Azul is fortunate is that we've been receiving support from our stakeholders, including crewmembers, aircraft manufacturers, lessors, banks, suppliers and the Brazilian government.

We have developed a management plan that includes a contribution from all of our stakeholders, and I am confident that we will come out of this crisis stronger.

I have created many businesses and airlines over the years.

It's all about people and relationships.

I can tell you we have the best people in the business.

Everyone on this call is an owner and a founder of the best airline in the world.

I'm confident that we'll get through this.

With that, I'll pass the word on to John, and he'll give you more details about our results and our response to the crisis.

Take it away, John.

Thanks, David.

Hello, everyone, and thanks for joining us today.

I couldn't agree more with David.

We have a great team, and I'm very thankful for their dedication, and I'm proud of how together we have been able to successfully navigate through these unprecedented times.

During the second quarter, our main focus was to preserve our financial liquidity and reduce fixed costs.

Thanks to our fast reaction, we were able to end the second quarter with a higher cash position than in the first quarter, an achievement that not many airlines in the world were able to report.

We expect to have a net cash burn of only BRL 3 million a day in the second half of the year.

And as you can see on the right-hand side of the slide, we have already negotiated the rollover of all relevant debt repayment in the second half of the year.

As you can see on Slide 7, we ended the quarter with significant sources of liquidity and unencumbered assets totaling BRL 6.6 billion.

We have no restricted cash.

We also have our loyalty program, TudoAzul, which can sell points in advance and is wholly owned and unencumbered.

We reached an agreement with the Portuguese government that was approved overwhelmingly by our minority shareholders this week to withdraw our conversion rights of the TAP bond in exchange for the financial support of EUR 1.2 billion to the airline.

That puts the current value of our TAP investment at BRL 754 million, equivalent to BRL 689 million in principal plus accrued interest and BRL 65 million for the sale of our equity participation.

We feel confident that with the aid provided by the Portuguese government, TAP will come out of this crisis and that our investment will continue to be protected.

As a reminder, we will maintain our status as a senior creditor at an annual interest rate of 7.5% and collateral, including a lean on TAP's loyalty program.

Our commercial relationship with TAP also remains intact.

During our last earnings call, we presented a management plan to maintain the liquidity required to confront this crisis and optimize the airline for the future.

The management plan counts with the support of all our stakeholders, including crewmembers, lessors, aircraft manufacturers, suppliers and the Brazilian government.

I'm happy to report that the implementation of the management plan has already generated over BRL 7 billion in working capital savings until December of 2021 resulting from the negotiations on several fronts you can see on Slide 9. We negotiate with labor unions, bringing flexibility to our operation and job protection for many of our crewmembers.

We also implemented other initiatives, including voluntary programs and involuntary furloughs, resulting in a projected salary expense reduction of 40% in the second half of the year compared to pre-COVID levels, a key component of our fixed cost reduction.

We continue to make progress in negotiations with our financial partners and expect to roll over all of our short-term debt amortization as it becomes due.

We've already rolled over all Q3 and Q4 payments.

In addition to the deferral of 59 Embraer aircraft announced in May, we also negotiate with Airbus the deferral of 23 aircraft to 2024 and beyond.

We've negotiated with Breeze Aviation to take delivery of 12 E-Jets over the next 18 months, including 3 still this year.

In August, we concluded the sale of 2 E1s generating additional cash after paying off the debt associated with these aircraft.

We were also able to accelerate the exit of 3 E-Jets this year previously scheduled to exit the fleet in 2021.

We successfully negotiated aircraft lease terms resulting in BRL 3.2 billion of working capital reduction, as we demonstrated on Slide 10.

As a result of the new lease agreements, we will follow an adjusted payment schedule based on a conservative demand recovery projection.

As you can see on the left side of the slide, this is much better than a power-by-the-hour structure because there's a buffer between the demand recovery projection and the rent payment schedule.

We also keep the upside if the recovery happens faster than expected, which is currently happening in Brazil.

This new payment profile gives us BRL 3.2 billion in working capital relief until December of next year.

In addition, as a result of the successful negotiations with our lessors, our lease liability is expected to decrease BRL 3.4 billion from the end of March to December, reaching BRL 12.5 billion by the end of the year, reflecting the present value of the newly negotiated lease payment under IFRS.

This is an increase in equity value for the company.

Moving on to Slide 11.

As demand recovers, we continue to rebuild our network.

We remain true to our network strategy by being the only carrier in over 80% of the markets we serve and by maximizing passenger connectivity through our main hubs in Campinas, Belo and Recife.

In June, we signed a historical codeshare agreement with Lat Am, increasing even more the capillarity of our network, reducing market risk and benefiting customers from both airlines.

This is truly a one-of-a-kind partnership agreement in terms of scale and connectivity.

We started selling tickets yesterday, and we look forward to offering the best travel network, service and options to our customers.

We've been carefully monitoring the demand recovery to gradually increase our network and schedules.

By September, we expect to offer over 400 daily flights, as David mentioned, representing 43% of last year's total ASKs.

By December, we expect to reach 60% of last year's total demand as we continue to see a positive trend in the demand recovery.

It's important to highlight that we only operate flights that generate enough revenue to comfortably cover its variable costs.

This is why having a flexible fleet with aircraft that have lower trip cost is a key in this current environment.

A couple of years ago, we invested heavily in our cargo business, and now we can clearly see this investment has paid off.

Azul Cargo continued to present an outstanding performance, with the second quarter of 2020 revenues essentially flat compared to the same period last year, even with a dramatic reduction in the size of our network.

Since then, the performance has been even better, and we've been able to increase our cargo revenue once again.

In July, for example, revenue increased 48% year-over-year.

Most of this growth was driven by broad expansion in all segments of the cargo market, especially our e-commerce logistics business, which represents just about 20% of our cargo revenue in the second quarter.

Given the fleet flexibility, breadth and connectivity of our network, we are uniquely positioned to take full advantage of the rising e-commerce demand.

Every asset we hold is available to our cargo team.

In summary, we had 3 major objectives since we talked to you last: first, react to the immediate crisis by preserving cash; second, implement the Azul management plan, which now gives us more than 24 months of liquidity by raising BRL 7 billion in working capital through our stakeholders; and third, create market opportunities such as our cargo e-commerce solution and the codeshare with Lat Am, which further allows us to develop our unique business model.

Now our focus is on utilizing our entire toolkit to bringing back our network in a safe and responsible way.

Crisis brings opportunities, and we are clearly taking advantage of the period we are now facing to search new opportunities.

With that, David, Alex and I will answer any of your questions as I turn the call over to the operator.

(Operator Instructions) Our first question comes from Josh Milberg with Morgan Stanley.

My first -- the first question I wanted to ask, about your lessors agreement.

Alex had suggested to us some time ago that you'd have this NPV benefit from the late payments, but that the nominal lease amounts would be honored.

And I just wanted to check if that is, in fact, the case or if there was some haircut.

And then a related question is whether there was or will be a change in the discount rate used to calculate the liability.

Sure, Josh.

Thanks for the question.

Yes.

So I think the majority of the benefit comes from the reprofiling, but there were discounts and haircuts, right?

But with the -- like, it's kind of in the low to mid-single digits across the whole lease liability.

It was a big discount on some aircraft but kind of diluted across the whole fleet.

We're talking about kind of low to mid-single digits.

So the vast majority of the NPV benefit comes from the reprofile.

IFRS requires you to use our discount rate, right?

And I think there is opportunity for the standard to be improved.

I don't know if it's the ideal way to do it.

But yes, there is a benefit from the reprofiling for sure.

Obviously, if you keep the same discount rate and you're just paying things a lot later than you would originally pay, you see a big NPV benefit.

But there is also the impact from the discount rate.

Okay, Alex.

That's great.

And then I also wanted to ask if you guys could comment a little further on your year-end capacity guidance and just talk about what you're assuming both as far as the evolution of yield over the remainder of the year and what you're assuming for recovery of corporate demand.

I ask that in part because we know that in the past, we've had a bigger dependence on some of your local peers, on business travelers.

So I just wanted to get some color around what you're seeing and what you're anticipating there.

Josh, Abhi here.

Yes.

So I think, first of all, we've been very pleased with the evolution so far from what we've been able to do from April onwards, to now and then on to September.

We're well ahead of what we had thought on our first quarter call, which was 40% by December.

Now we think it's at least going to be 60%, which is a huge improvement.

There's no question that -- and this is probably global, that leisure demand is so far ahead in the recovery.

We can see that in our booking channels, for example, the site and the app are much more ahead on a year-over-year basis than corporate is.

But we are seeing nice improvements in corporate as well July versus June and August versus July.

So I expect that this sequential improvement to continue.

And I think that right now, leisure is, it's safe to say, probably well over 50% of last year's levels.

Corporate is behind but catching up.

And so I think being at 60% overall, that's going to be more biased towards domestic, less towards international.

I think you're going to have leisure above that number, maybe in corporate, a little bit below.

But everything that we are seeing so far supports the sequential growth that we have been doing.

Most importantly, we've been very measured in how we're ramping up this capacity.

So it's ramping up smoothly, week by week, month by month.

We're not having any big jumps in between giving time for demand to catch up, giving us time to build the curves as well as we get bigger and bigger over the next couple of months.

So I think that even by the end of the year, leisure is going to be ahead.

Corporate will be behind leisure.

But we're seeing both of them recover.

And so far, demand has kept up to the capacity that we are doing, and I feel very, very good given the financial metrics that we're holding ourselves accountable for.

As John said, this flying has to cover the variable margin in a comfortable way, has to help in generating cash for the airline.

I expect that this sequential growth to continue.

And to the extent that we can do better than 60%, we're ready for that.

I think that we're sort of being flexible and taking advantage of the opportunities as they come.

And Josh, I think it's important to note, the deal we cut with our lessors, and when we did that, we had a more conservative assumption in place.

That's to our benefit now, right?

And so we obviously have less payments going forward than we believe we'll see in terms of demand.

So even if the leisure business mix isn't where we expect it to be or want it to be, let's say, we're still going to be much better off.

No.

John, Abhi, that makes a lot of sense, what you said.

And do you think that it does make sense or a reasonable assumption could be that yields stay at the same level over the remainder of the year?

What are you thinking there?

I think there's going to be a steady improvement.

We've already seen average fares come up very nicely over the last 3 to 4 weeks.

And the very good news is that the volume of demand has held up as those average fares have come up.

Average fares are still down as they are all over the world, but they are coming up.

And so I expect those yields to continue improving as the corporate mix gets a little bit better, as the customer gets more confidence in the health and safety on board, in the traveling experience as hotels open up, experiences open up over the country as cities open up.

So I think the yields are going to continue to improve together with volumes.

And I think that's going to drive the revenue improvement month by month, and that we're going to try and keep pace with capacity.

So I expect yields to continue improving sequentially every single month.

Okay.

That's great.

And if I could just squeeze in one more quick one.

You guys mentioned the E1s that Breeze is going to take.

Are they going to take those on the same sublease terms that were initially established?

Or was there some adjustment there?

Yes.

Our next question comes from Savi Syth with Raymond James.

If I might just follow up on the fleet question here.

Could you provide what you expect in terms of delivery timings over the next 18 months?

I think there were a few still that you're planning on taking delivery, but I might be wrong there.

And then also kind of tied to that, do you expect any more E-Jet removals in 2021?

Yes, Savi.

So we have an agreement where we don't have any CapEx requirements over the next 2 years.

And so obviously, there were some aircraft that were built by Embraer.

But they were all lessor-financed.

And so we will be taking a few Embraers and a few Airbus aircraft over the next, call it, 12 to 18 months, already built aircraft.

But the payment profile starts with what we did with the other lessors.

And so we have a significant runway of reduced rent in a few months with no rent at all, similar to what we did on our other deals.

And so that was kind of part of the overall agreement to ensure that we have the appropriate runway going forward.

There are a lot of aircraft that are exiting the fleet.

I mean we still have a deal with a lot in Poland.

But I think everything is kind of on pause.

Even Breeze delayed a little bit their entry into service, but you'll still continue to see E1s exiting the fleet over the next couple of years.

And it's really important to highlight that we pushed off all other deliveries to 2024 and beyond, right?

And so this is -- we have helped from an aircraft standpoint in 2020, 2021, 2022 and 2023.

And so that provides us a significant amount of runway from a fleet perspective.

And we have no financing risk going forward because the deliveries are coming from the lessors.

Yes.

And when we developed the management plan, we assumed the worst, right?

We assumed that no aircraft would leave the fleet, and so everything that we are able to do now is generating upside, right?

So the 2 aircraft that are going to Breeze this year, there's a third one that leaves the fleet this year and kind of delivers early next year, those were not sort of in our management plan and our ASK to our stakeholders, right?

We didn't expect them to leave the fleet when we were building the management plan.

The sale of the 2 aircraft that we've done, we sold 2 aircraft in this environment, right, which was also a positive surprise.

It was not something we were -- when we built this in April, it wasn't something that was on the cards.

But it's a few tens of millions of reais of incremental cash that is going to come through the door in Q3, right?

So a lot of positive news but all consistent with the concept that we always talked about of hoping for the best and preparing for the worst.

So all of these positive news from Breeze and from aircraft sales and early redeliveries are all positive based on -- compared to our original forecast.

That's super helpful.

And Alex, maybe if I can ask just on that.

There are a few moving parts on the kind of the cash side as we head into 2021.

If you can assume things plateau at this 40% of demand by year-end level that you're expecting, what could cash burn look like in 2021?

I know I touched unrealistic, but I'm just trying to get a sense of once the relief on taxes and fees kind of get reversed, then maybe you have a little bit more on rent to pay or maybe less.

So what does cash burn look like in 2021?

I think what we have now is a road map of -- we know that we can sustain the cash burn that we guided to of BRL 3 million in the second half.

There's some seasonality in that.

Third quarter is a little bit higher.

Fourth quarter is going to be lower.

But we kind of know the road map, and we have an understanding with all of our stakeholders that this is the plan for now, right?

If things change, the plan may change.

And even our agreements with them may change, right?

I think we've gotten to a very constructive collaboration where everybody has signed up to helping Azul, and everybody has the awareness that it's the best thing for everyone for them to support Azul and get us to the other side of this crisis.

So if the cash burn accelerates, right, I mean, there are a lot of ifs there, we may have to sit down and renegotiate.

It all will depend on what our cash position is, what potentially we may raise additional capital.

We've gotten this -- again, it's all -- the management plan is built under this very strong foundation of all stakeholders pitching in.

And if we can raise additional capital under the right conditions, we will, right?

And that may allow us to sustain a higher cash burn if that becomes a reality, right?

So there are so many ifs there that if our current cash position can sustain a higher cash burn, that's great.

But if it can't, we have a road map, and we have an understanding that we will have to negotiate again, and we're confident that we will have the support that we need.

So the uncertainty that we had maybe back in April, right, of are we going to have enough cash?

I'm very confident now that we're going to have enough cash, right?

And that may be under different scenarios for daily cash burn, but we're not worried about having enough cash.

But Savi, I think it's really important.

The deal we cut with the lessors is through the end of 2021.

We have significant rent reductions through 2021, okay?

And so I don't know of another airline that's been able to negotiate a deal like that, that you talk about if demand is at 60% next year, we are not paying full rents until 2022.

And so we have a significant rent reduction for the first 6 months and the second 6 months of 2021.

And that's the plan in place.

And so obviously, things can change.

There could be a second wave, but we've prepared the airline for a worst-case scenario, and we're already seeing things much better than that.

And so you've seen that, and we try to guide to that to what you're seeing in the third and the fourth quarter.

But that does not go away in 2021.

And then even in addition to that, what we pay in rent in 2022 is flat rents to what we were paying pre-COVID.

So it's not like we snap back in 2022 with higher rents.

We've negotiated with our lessors that, that only goes into effect in lease extensions and repayments starting in 2023, and many of those over the term of the lease, so that will be paid over a 10-year period in many instances.

And so we've created a significant amount of runway through our stakeholders because our stakeholders understand that this is a highly profitable business.

We have a sustainable network, and we've got a great franchise here.

And I think that that's the significant difference from what you're seeing from other airlines in the world.

Our next question comes from Victor Mizusaki with Bradesco BBI.

I have 2 questions here.

The first one is just a follow-up on Josh's question about lease liability.

So to double check here, I mean, Alex, so there's no change in the discount rate to calculate the NPV, right?

Sorry, can you repeat the question, Victor?

Because you mentioned about the discount rate to calculate the NPV of the leasing contracts.

So just to double check, I mean, so there's no change or no increase in the discount rate?

No, there is.

I mean the discount rate follows our cost of capital, right?

And so our cost of capital certainly was much lower pre-COVID than it is right now, right?

And IFRS essentially requires you to recognize the lease liability when at the start of every lease or at the modification of any lease, and so that's what we're doing, right?

It's not the same rate across the board.

But like I said, there is -- when you look at the NPV, there's a benefit from later payments.

There's a benefit from the higher discount rate, and there's a benefit from reductions in rent payments, in discounts and also in early redeliveries of aircraft, right?

Everything helps.

I think the smallest one -- the smallest effect of the 3 is the discount, but it's there, and the rest is fairly even.

And Alex, when I take a look on the presentation, I mean, you also mentioned about these -- I mean the change of some leasing contracts to power by the hour.

So I don't know if you can comment if this is material to cover this NPV as well.

Victor, we have no power by the hour contracts.

Yes.

This is -- yes, so -- no, let me just kind of clarify to make sure.

I think the majority of the airlines in the world are probably going to do power by the hour, right?

That's kind of fairly easy and fairly standard across the board.

This is a much better deal than a power by the hour.

Like John mentioned, it's a set profile with a cushion to what the recovery is that we're seeing, right?

And so we keep that upside.

Obviously, if we are swimming in cash, right, there are provisions.

And this is a partnership.

We work under a win-win concept.

And so obviously, we're not going to try to do great while all of our suppliers and partners do badly, right?

But everybody is in this together, and everybody has been very supportive.

But unless we're swimming in cash, right, this is a set rent profile with a cushion to what we're seeing in terms of demand recovery and to what Abhi is able to utilize the aircraft, right?

But it's not a power by the hour.

Okay.

Okay.

And just last one.

Now that you basically sold the -- your stake in TAP, is there any risk that maybe TAP you try to early return some of the aircraft to Azul?

We have a contract in place, right?

And so in fact, 2 of the aircraft that we just sold to a leasing company were on lease to TAP.

And so we have a contract in place, and so there's no...

Yes, the market saw value in the fact that these aircraft have a sublease to TAP, and so they were willing to pay for it.

So I don't think there's any expectation on any kind of disruption there.

And I think if this crisis has taught us anything, is that if you want a early redelivery on aircraft or reject aircraft outright, there's one way to do it, it's under Chapter 11.

If you're not under Chapter 11, you can make marginal changes, right?

But nobody is going to be able to deliver an aircraft years and years in advance to a lessor unless they go through a very dramatic and disruptive approach because, like we said in the previous call, right, the asset is no longer the aircraft, right?

The aircraft used to be seen as an asset, but who wants to -- who wants an aircraft today?

No airline in the world wants additional aircraft.

The asset is a sustainable business model, is a constructive long-term partnership that can generate cash.

And that's what we have.

And Victor, TAP just received an injection of capital of EUR 1.2 billion, right?

And the state took over the ownership from David and Azul.

And so it's a sovereign deal, right?

And so I think that that's something that's very encouraging going forward for our leases, for our debt and certainly for our partnership for the joint venture.

Our next question comes from Mike Linenberg with Citibank (sic) [Deutsche Bank].

Actually, Deutsche Bank, but Germany versus New York, right?

Anyway, couple of questions here.

I guess to Abhi, just looking at your progression and how your capacity recovers through the year and you have it getting to 60% of 2019 by the month of -- by December, I think you've also indicated that demand by December will also be at 60%.

So presumably, we're looking at flat load factors year-over-year by the month of December.

This quarter, we were down, I don't know, close to 10 points.

When we look at July, August and September, where the capacity goes from about 23% to 43%, are you seeing meaningful convergence on the load side such that you feel comfortable that supply and demand will be back in equilibrium by year-end?

Mike, yes.

So I mean, if you look at our July traffic, we flew a 78%, 79% load factor, right?

So I think that you kind of had the initial shock of the crisis.

You also had very high no-show rates initially.

And now you had very high change/cancel rates in the month of April, for example.

Those things have stabilized.

And so you also now have time with higher booking volumes to build a booking curve that we didn't really have for the month of April.

So yes, I think for us, load factor is more of an output.

I'm not really targeting a load factor.

But yes, I think load factors are going to be in the high 70s for the foreseeable future if not a little bit higher than that.

And so I think we've already reached that position.

We were at 76% in June, 78%, 79% in July, and I think that's where we'll be going forward.

What has improved quite significantly in the last 3 weeks is average fares.

And that's been a very positive sign for the market overall, especially for us because the fact that we're alone in 80%, 85% of our routes, we're always pushing the envelope on that.

We always have been historically.

And we continue to do so.

And it works for us.

It works, especially with close-in demand and close-in revenue.

And so to the extent that we feel like the revenue-maximizing solution is a load factor of 76% versus 78%, I'll certainly take that just to avoid dilution in our routes, which obviously we're alone.

So yes, I think it's going to be in the range of 76%, 77%, 78%, 79%, and again, the fact that we're ramping up capacity very steady week by week, month by month also gives time for demand to catch up, gives us time to build the booking curve and gives us time to make adjustments as needed.

So far, we've been rewarded in the increases that we've taken.

Initially, I would say it was more on the volume side.

And now lately, it's come on the average fare side, and volumes have held up with those higher average fares.

So certainly looking positive for this continued sequential increase.

And I think we have all of the toolkits, given our network, given our connectivity and all the different fleet options that we have to take full advantage of the recovery that continues over the next couple of months.

And Mike, I just want to highlight that Brazil has been shut down for the last 3 or 4 months, and it just started opening up in the month of July.

If you think about the State of São Paulo, which is the big economic center, it's just started easing restriction.

They just lowered kind of the rating over the last 2 weeks, kind of allowing for more and more people to go out and about, go to restaurants.

Schools still have not opened here yet.

That's now projected to open in October.

And so the fact that we're seeing this demand now as the economy slowly reopens, that's what gives us kind of hope that you're going to see a sequential improvement as we move forward.

Yes.

I'm just -- in the U.S., we're not even getting the volumes.

And the fact is you're getting the volumes and then, to Abhi's point, both just recently and even earlier on the call that you're seeing this sequential improvement in yield.

So if you can get volumes and that type of improvement on the yield side, you could have one of the better fourth quarters or at least one of the better months of December than any airline when you look around the world.

So that's why -- that's kind of what drove my question there.

Just maybe a second question on Lat Am Brazil and yourselves, I know that, that turned on in August.

And I think, John, you did make that point that there's no sort of local JV or codeshare arrangement like that, really, I think, anywhere in the world when you think about depths and breadths and from a domestic market perspective on the fact that you were able to do that.

I think it's going to be -- it's going to have -- it's going to show up in your numbers.

And I'm just curious, as you look out over the next, say, quarter or 2, as you think about possible contribution, is it a couple of load factor points on Azul, whether it's through connections or people who sort of cross-utilize the carriers?

Is it a few points there of load factor?

Is it a few points of margin?

Sort of how do you think about it?

Because what you have is very unique and powerful.

Yes, Mike, I think one thing is -- the market has not understood this partnership.

This would never have been possible 6 months ago.

We never even thought of it.

And so this is just a way that we've taken advantage of the market opportunity.

And so we probably wouldn't have gotten antitrust approval to do it 6 months ago.

And so it's a way for us to increase demand, and that's a great thing.

And when you think about the power of our network and the city served and the size of their network, the size of their loyalty business, right, it brings a tremendous amount of value to customers and to both networks.

And it's a way to get both of us flying a lot faster.

Yes, Mike, the answer to your question is yes, right?

It's going to be some form of incremental revenue.

Probably more on the load factor side, that's probably fair.

But it's going to bring connecting traffic.

Of course, we combine the power of our distribution channels as well.

We're both obviously very significant here in terms of distribution and, of course, accrual possibilities for our loyalty members.

And so my hope is that, yes, we absolutely utilize each other's connecting networks to deliver the best possible options for the customers, and they don't have to go anywhere else.

And I think if we can do that, both of us will be able to recover better with this partnership than without.

Great.

And then just one quick last one to Alex.

Just the TAP bond, when does that mature?

2026.

Our next question comes from Steve Trent with Citibank.

Just a few quick ones for you.

First, when we think about government policy, I don't mean the funding, but the other things that the government has done, some breaks on landing fees, I guess, there's been some adjustments on cash refunds to people that cancel trips.

Do you see any longer-term changes that the government is looking to make that could facilitate competition, this business about airlines having to pay customers lodging if there's a thunderstorm and the flight's canceled, stuff like that?

Yes, Steve, I think one thing that hasn't been highlighted enough in the new law, 925, that was just recently approved is moral damages can no longer be claimed for lawsuits.

And so that was a big problem.

You saw that Brazilian airlines paid 20, 30x what a U.S. carrier would pay.

And so that was embedded in our cost structure.

And so that law changed that.

And so that's a great long-term benefit for the aviation industry overall.

You highlighted the reimbursement.

There's other smaller things like the deferral of the landing fees and -- during the pandemic.

But I think the biggest one that I'm probably most proud of is the consumer protection has changed because of the moral damages no longer exist.

And that's a big long-term benefit.

And I think this government is very sensitive to aviation.

They understand the importance of it.

I've talked several times to senior people inside the government.

If you think about the tax revenue that aviation brings, the tax revenue that just Azul brings, Azul brought BRL 2 billion in tax revenue to the government last year.

And so that's important to help protect this franchise and to protect this industry overall is very important.

So I think you'll see more movement from the government on additional provisional measures, kind of a restart plan to kind of help this industry overall because this industry helps the taxi drivers, the hotels, it helps business in so many cities in Brazil.

And so I think that you'll consider the government make moves to kind of help this industry going forward.

Obviously, Brazil does not have the resources that the U.S. has, right?

And it doesn't have the resources that Europe has, but they do have the ability and the willingness to change the laws to kind of help make Brazil a much more competitive place.

Okay.

Very helpful, John.

Appreciate that.

And just one quick follow-up.

Could you remind me sort of on a normal cycle basis, approximately what percentage of Azul's ticket sales are refundable tickets?

For refundable -- hey, Steve, it's Abhi here.

It's actually pretty small, fully refundable.

We have different fare products, and most of them have change in cancel fees associated with a fully, fully refundable.

We only sell actually through our call center, so it's a very small percentage.

If you're asking about what sort of outstanding in terms of refunds, in terms of all the changes to the network, it's been very positive that over 80% of the outstanding tickets out there that we've changed are actually in the forms of credit shells, not refund.

So the future refund liability is actually pretty low, not something that should change our cash or anything like that going forward in a meaningful way.

And we're actively trying to get those customers to change over to credit shells.

But the good news is that the vast majority, over 80%, of the customers that have been affected by the changes in the network over the last 3, 4 months have opted to leave their money with Azul, and we'll potentially use that in the future.

Many -- much of that would break, to be honest with you, but the actual refund component is very small.

Our next question comes from Pablo Monsivais with Barclays.

I have 2 quick ones.

The first one is, why is demand picking up so fast while the COVID case in Brazil keep rising?

Are you seeing that the population is less fearful of getting infected?

Or what's going on?

Why we're not seeing this in other regions in Latin America?

And the second question is on the cash burn guidance.

Are you incorporating already the buildup of deferred revenues for the summer season?

I can start with the demand question.

So I think that -- at least, I can speak only about Azul and what we are doing to capture this demand.

One of the things that's unique to us is the connectivity in our network and how we utilize different aircraft in our fleet to maximize the capture of this demand.

So I'll give you one example.

We used to fly from Campinas to Santos Dumont in Rio, 14 times a day on Embraers.

Now we don't have the corporate demand for 14 flights a day, but we have leisure demand that wants to connect, so we fly 5 times a day on A320s.

Each one of them comes to our banks in Campinas and then distributes to over 50 cities.

So I think we're -- with our network, with our fleet flexibility, we're able to provide really convenient options to customers that want to go for leisure, they want to go visit friends and family or take care of a loved one, or small and medium businesses that need to go visit a customer, restart their shop, restart their business in different locations.

So I think there is an aspect of Brazil opening up in general.

That's helping overall.

But I think there is a very unique aspect to how we build our network.

Our hub in Campinas, for example, is very close already to being at what it was precrisis.

And that's because of the connectivity that we're delivering with the different fleet options that we have and through all the different destinations that we have.

So I think that it's -- Brazil opening up is one aspect, but I think that we're disproportionately able to capture a lot of that demand through the breadth and the unique connections that we offer in our network.

Yes.

And on the cash burn, so we gave you guidance for second half, right?

Obviously, there is seasonality in Brazil.

Q4 should have better cash burn than what we guided to.

Q3 should have worse cash burn that we guided to, but that's all embedded in the projections that we gave.

So our expectations of how the booking curve will behave and the way we're seeing the advanced purchase time line and the natural seasonality that we see in Brazil with some quarters being great in terms of generating cash and maybe not so good revenue, some quarters, you have great revenue, but then you don't generate cash as much, right?

That's all embedded in our projections.

And just on -- in the third quarter that Alex highlighted, we made some moves that were difficult to do but will pay back over the next 3 to 4 months.

And so obviously, letting crew members go, and we said we're going to hire everybody back when we get back to full speed, but some of that cost will be embedded in the third quarter, but you'll see the savings by the time you get into the fourth quarter, right?

And so I think that those are some things.

But when you think about the ATL balance and the forward booking of sales, if Abhi continues to add to the network and continues to forward-sell much more, there's the ability to kind of generate cash through that channel as well.

And so that's something that you're literally adding 50% more flights month-over-month, and that could be very positive from a cash flow perspective.

Great.

Maybe if I can squeeze one follow-up question to your comment is, what are you seeing guys in terms of cost and the reduction on labor from temporary initiatives?

And how are we seeing costs from permanent initiatives?

Head count was significantly reduced when you're building and adding construction capacity back.

How your cost will move?

And what maybe is the headcount number, I don't know, in 2021, 2022?

Yes.

I think the important accomplishment here was to make our salary expense variable, right?

Normally, the majority of salary is fixed.

It's mainly pilots and flight attendants that have variable pay that go up and down according to capacity, according to how much they fly, right?

But through our negotiations with our unions and through the headcount reductions that we've made, many of the volunteer programs that we offered, and we saw a big number of crew members signing up, we were essentially able to make that salary expense variable.

So if we -- we're seeing that the recovery is happening faster than what we originally projected.

So obviously, there will be more labor expense than originally projected, but that's more than compensated by the revenue that we're getting, right?

We continue to only fly what covers variable cost with a healthy margin, right?

And if, for example, on the other hand, if the recovery maybe slows down a bit, then the salary expense will go down as well, right?

So that was, I think, a very positive aspect of the negotiation and the development of the management plan that we were able to make this variable.

But it's important to note, we will be a much more efficient airline in 2021 and 2022.

There is no doubt about it.

Our headquarters is leaner.

Our airports are leaner.

We will be a much more efficient airline going forward.

And so once again, crisis brings opportunity, and we need to make sure that on the backside of this crisis, we're a better airline.

And so you've seen the technology that we've employed at the airports, right, and everything that we're doing to become a much better airline.

I do believe, over time, we will hire people back, but we're going to grow again.

And we'll have the opportunity to make room, but we will be much more efficient from an FTEs per aircraft on a going-forward basis.

Yes.

We're looking at this as a reset and as an opportunity for us to start up again.

We're still a young company, a young airline.

We started up less than 12 years ago.

But especially on the overhead side, right, as John is mentioning, we're seeing new ways to work, right?

We're seeing that maybe there are some structures that we had in the past that we don't need to rebuild, right, that we've learned and adjusted to.

On the operational side, the physical distancing is leading to a lot more utilization of technology to check-in.

So yes, I don't think you're going to ever get back to the same level that you had before.

We'll be able to get to a postcrisis efficiency level.

That's a lot higher than what we started with, and that's certainly a goal that we have here and that we're already managing through.

Great.

So reading between the lines, the percentage of temporary initiatives on your labor costs might be not as big and then -- because we are seeing a permanent change on a more efficient airline, I guess, for the future?

Yes.

We talked a lot about the BRL 7 billion of working capital that we created through the management plan.

But some of that -- that's just through the end of 2021.

We will be a more efficient airline in 2022 as well, right?

And so some of that savings is going to be permanent.

And so we've just sized for you what we've created over the next 18 months because that's what we're projecting when we will be at almost pre-COVID levels in terms of line.

Our next question comes from with [Florencia Torres] with [MetLife].

Congratulation for the good news in this challenging environment, and I have a couple of questions.

The first (technical difficulty) of that.

And the other one is regarding capital structure, just thinking in 2021, what's the expectation regarding leverage for this year?

It's expected to increase, and it's not a surprise, but if we try to think in 2021, what's your expectations?

So I'm going to answer the capital structure question first.

We couldn't hear your first question.

So if you could, later after I answer, maybe you could follow-up and repeat it.

So the way we're managing through this crisis, obviously, with the huge drop in revenue, right, we are adjusting capacity and reducing expenses, but no airline in the world is able to reduce expenses as much as revenues have fallen.

So that is going to affect your net debt one way or another, right, either by burning cash or by taking additional debt.

And that will increase our leverage.

But when we forecast -- when we look at the next 5 years, we believe we're going to converge to an acceptable leverage probably in 2022.

So we're not -- and everybody, I think, understands that 2020 and 2021 are extraordinary years that don't reflect sort of the true performance of the airline or the true capital structure of the airline.

So 2020 and 2021 will probably be still years when you're attributing a lot of emphasis to cash, right, and making sure that the airlines that you're looking at have enough cash to get to the other side.

And then 2022 will probably be the year where we're under sub 4 in terms of leverage.

And that's, I think, a better reflection of what our capital structure will be.

Now in terms -- we saw -- you saw our release on cash.

We were very happy with the performance that we had in cash in the second quarter because it's all about making sure you have the runway to get to the other side, right?

Not all airlines in the world will have the runway to get to the other side, and they may have to resort to more radical approaches like some airlines have already done in the region.

But we're happy with the cash position.

And -- but -- and when we look at what -- like we said, the end of the crisis is expected to be kind of end of '21, beginning of '22.

When we look at the expected cash position that we will have, it's comfortably above our minimum cash position, but I think it is prudent of us to consider increasing that cash liquidity, right?

Because the trends are positive, we are feeling encouraged, but I don't think anybody has full visibility as to when this crisis is going to end, right?

We have an expectation, but we have no assurances that we're going to get there.

So we will be monitoring the capital markets.

We're still talking to BNDES.

We believe BNDES is still an option.

And when we see an opportunity to access the capital markets, we will do so.

Maybe there will be equity, maybe there will be debt, but we'll take that decision if and when it comes up.

Okay.

Perfect, very clear.

And the first question was regarding the headlines regarding the coupon payment of October.

That if you can provide any update or thoughts regarding potential postponement or -- I know that you already made a very good shot in conversations with lessors, with the others, with banks or other financial institutions, but what's your -- what's the guidance regarding the coupon payments or relationship with bondholders?

Yes.

The coupon will be paid, right?

The article, I think, is well written.

The headline, not so much.

I think there's a little bit of -- I think the whole industry is kind of chasing clicks, unfortunately, but the article is accurate, right?

Our bondholders are reaching out and we obviously want to have a channel and a communication with our bondholders.

And if there is a chance for us to raise additional capital through terms that are acceptable to us, that's all positive.

But there's -- there are no negotiations, and the coupon will be paid in October.

Our next question comes from Stefan Styk with Bank of America.

So it looks like in the second half of the year, you have lease payments of around BRL 235 million per quarter, which looks much higher than the second quarter payments.

So I'm just wondering what -- how steep the ramp-up is at the turn of the year into 2021.

Just to be clear, I mean, we had 70 flights a day in the second quarter, and we're going to be at 400 in September.

And so obviously, lease payments will go up as we ramp up, but it's a significant difference in terms of capacity.

Yes.

I think what's important here is that our expected ramp-up in capacity and revenue generation is going to be well ahead of the ramp-up in lease payments.

And that is the bonus.

That is the difference that we keep in terms of cash generation in terms of helping our cash burn as well.

So I expect the ramp up to continue December, January and onwards.

And we'll go as much as demand takes us, and it's going to be -- there are no artificial limits in kind of -- that we have to stop at a certain point.

But we are very, very confident that even with the discounts that John talked about, for the first 6 months of next year, we will be already at those levels of capacity this year.

And so for the foreseeable -- for the life -- for the next 18 months of these lease agreements, we expect to be comfortably well ahead in terms of capacity, in terms of revenue generation.

And that is the buffer, that is the upside that we get to keep.

Yes.

And one thing that I just want to comment on, the comparison of Q2 to Q3, our expectation -- our original expectation was that the cash burn in Q2 was going to be of up to BRL 4 million per day, and we expected to end Q2 with BRL 300 million fewer reais than we actually ended up, right?

The fact that we over-delivered in Q2 should not be read as Q3 getting worse, right?

Q3 is going to be better than what we expected Q2, it's just that we did, if I say so myself, a great job in Q2.

But that shouldn't be seen as a negative.

It's certainly a positive.

The company will now answer questions placed on the webcast platform.

We have one question from Bruno Amorim from Goldman Sachs.

Can you please provide the percentage of passengers and revenues that comes from business versus leisure in '19.

Besides being a headwind on passenger volumes, could eventually structurally lower business-related demand also weight on margin in the future?

Bruno, so I mean, we always said precrisis, we were around 60%, 65% corporate, and that was the nature of our network.

We flew to many small cities alone.

I think that right now, obviously, leisure is much ahead and corporate is coming up behind.

But I think that we have all of the tools to adapt to whatever the new demand environment is.

So the example that I gave of Campinas to Santos Dumont is perfect.

We used to fly 14 times a day with an Embraer.

And now, for the next 3 months, maybe we don't have that demand.

But we definitely have demand to fly 4, 5 times a day with A320s, which deliver 174 passengers each time to our banks in DCP in terms of connecting traffic.

And so I think we have the flexibility in our fleet.

We have the breadth in our network, already at 88 destinations by September.

So to the extent that you don't have the high frequency in markets for the next 3 or 6 months, maybe the demand is going to require you to have more breadth, more destinations, more connectivity.

We are easily able to accommodate that uniquely much better than anybody else in the region.

And in addition, we have other tools in our toolkit.

We've just adapted the world's first Embraer airplane into a cargo airplane.

It's a 2-day conversion.

All you do is remove some seats.

And so if we don't need some Embraers over the next 3 to 6 months, I know that there's e-commerce demand to medium-sized cities that can deliver next-day and 2-day products at the low trip cost of an Embraer, significantly better than a 737 freighter that somebody else could offer.

So we have all those tools in our toolkit to adapt to whatever the demand environment is going to be.

I do expect corporate to come back and -- but maybe in the transition period, we'll have to structure our network accordingly and use other things in our toolkit.

Yes.

And one other thing, too, Bruno, is Brazilians don't travel enough.

Right?

And so maybe the U.S. has a fall in travel demand going forward.

Maybe other parts of the world doing so, maybe our growth curve is not as steep as we had thought, but Brazilians are traveling almost 1/3 of what Chileans are traveling.

Absolutely no reason for that.

Brazilians are traveling less than what Colombians are traveling.

Obviously, no reason for that.

With the size of Brazil, the opportunity is huge, and we will continue to find pockets of demand throughout the entire country.

And you think about everybody doing these Zoom conferences, that's not happening in the agro business, that's not happening all throughout Brazil and the small cities that we serve.

And so I think that's a strength of our network that nobody can really replicate.

Okay.

We have another question from investors.

So what happens to the BNDES funding now?

You have generated such good liquidity.

Yes, I think it's a good liquidity on a relative basis, right?

But I don't think -- we need to be prudent.

Like we said, we need to hope for the best, but prepare for the worst.

And I think preparing for the worst is increasing our cash cushion.

I think that's still recommended, that's still probably the right thing to do because we don't know what could happen, right?

We're still waiting for corporate to come back.

There are some green shoots from corporate, but we can't guarantee.

I think if somebody could guarantee to us that the trends are going to continue, then I think you're right.

Maybe we don't need it.

But we don't know that.

And we could be, quickly, if there is a second wave, if this pandemic takes longer to end, right?

We still remember how quickly things evolved in March and April, right?

This could get ugly very quickly.

So I don't think we can discard the opportunity to take additional cash.

And I think the BNDES is an option that is out there, but I think even the BNDES would like us to see if we have other options out there.

And we will be monitoring the capital markets to see under which conditions capital could be available to us if and when we see the opportunity.

But the BNDES is still an option.

It's still out there, and we continue to have regular discussions with them because it may be still -- BNDES is a lender of last resort, and I don't think anybody can guarantee that they wouldn't be needed.

Okay.

One more question.

How much liquid cash do you have?

So we -- we have no restricted cash, which is an important concept here, right?

I see some analysts and some airlines kind of including cash -- restricted cash in their cash position.

So I think it depends on what your definition of liquid cash is.

I don't think restricted cash is liquid cash, but I see some people calling at that, right?

If you just call it pure cash at hand, we have BRL 1.6 billion.

In Brazil, receivables are cash, right, and including receivables, we have BRL 2.3 billion.

And we have our TAP bond, which I normally wouldn't call it cash, but it's certainly a heck of a lot more liquid than restricted cash, right?

So if you include restricted cash in your cash position, you certainly should include the TAP bond in our cash position.

Especially now that there's basically a sovereign guarantee on.

And one more question.

Could you talk about what makes your Lat Am codeshare difference?

Is it size, nature, revenue split?

Yes.

The Lat Am codeshare is definitely the largest in terms of breadth that I know of.

Most codeshares tend to be international partners connecting with domestic partners.

You do have 2 examples of domestic codeshares, in the U.S. at least, but they're geographically focused on each coast.

This one is the entire domestic market in Brazil.

And so combined, precrisis, that's 50 million passengers, 60 million passengers, which falls within the scope.

So it's certainly unique in the scale.

It's certainly unique in the breadth, and they're certainly unique in the opportunities and the benefits that we can provide in terms of connecting our customers, in terms of the portfolio of destinations and the loyalty program benefits that we will provide and want to keep adding to over time.

So it certainly encompasses a lot more and provides a great platform for us to recover each of our networks going forward.

Ladies and gentlemen, this concludes today's question-and-answer session.

I would like to invite John to proceed with his closing statements.

Please go ahead, sir.

I'd like to thank everybody, and feel free to reach out to any of us after the call if you have any questions.

I just want to remind everybody, we have a great business, we have a great team, and we're taking advantage of this crisis to look for opportunities that will make this airline a lot stronger in the long run.

We appreciate your support, and we look forward to seeing you in the future.

Thank you.

Ladies and gentlemen, that does conclude the Azul's audio conference for today.

Thank you very much for your participation, and have a good day.