GATX Corp (GATX) 2020 Q3 法說會逐字稿

Good day, and welcome to the GATX 2020 Third Quarter Earnings Conference Call. Today's conference is being recorded.

At this time, I would like to turn the conference over to the Director of Investor Relations, Shari Hellerman. Ms. Hellerman, please begin.

Thanks, Casey. Good morning, everyone, and thank you for joining GATX's 2020 third quarter earnings call. I'm joined today by Brian Kenney, President and CEO; and Tom Ellman, Executive Vice President and CFO.

Please note that some of the information you'll hear during our discussion today will consist of forward-looking statements.

Actual results or trends could differ materially from those statements or forecasts. For more information, please refer to the risk factors included in our release and those discussed in GATX's 2019 Form 10-K and its 10-Q for 2020.

GATX assumes no obligation to update or revise any forward-looking statements to reflect subsequent events or circumstances. I'll quickly recap our third quarter financial performance, and I hand it over to Brian for a short discussion on Rail North America's maintenance operations as well as the Rolls-Royce and Partners Finance affiliates results.

Earlier today, GATX reported 2020 third quarter net income from continuing operations of $48.2 million or $1.36 per diluted share. This compares to 2019 third quarter net income from continuing operations of $37.2 million or $1.03 per diluted share.

Year-to-date 2020, net income from continuing operations was $132.4 million or $3.74 per diluted share. This compares to $138.7 million or $3.79 per diluted share for the same period in 2019. The 2020 third quarter and year-to-date results include a net negative impact of $12.3 million or $0.35 per diluted share related to the elimination of a previously announced tax rate reduction in the United Kingdom. The 2019 year-to-date results include a net deferred tax benefit of $2.8 million or $0.07 per diluted share related to an active tax reduction in Alberta, Canada.

These items are detailed on page 14 of our earnings release.

In the second quarter of 2020, GATX completed the sale of American Steamship Company. Accordingly, this business segment is reported as discontinued operations and prior periods have been recast to conform to the current presentation.

Now I'll briefly address each segment. At Rail North America, our fleet utilization remained high at 98.2% and renewal success rate was 58.1%. Although absolute lease rates for most car types were flat to slightly higher compared to the second quarter, we expect lease rates to remain under pressure given a continued oversupply of railcars in the market and carload volumes relative to 2019.

Third quarter renewal rate change of GATX's lease price index was negative 29.4%, reflective of the ongoing challenges in the marketplace relative to the strength of the expiring lease rate that generally commenced at the height of the market 6 to 7 years ago.

The average renewal term associated with the LTI is 29 months. As noted in our earnings release, despite higher fleet churn as a result of lower renewal success in the quarter, our maintenance cost performance was better than expected, which Brian will address further in his remarks.

We continue to successfully place new railcars from our committed supply agreements with a diverse customer base. We have placed our 8,950 railcars from our 2014 Trinity supply agreement and over 1,670 railcars from our 2018 Trinity supply agreement.

Additionally, we have placed over 3,470 railcars from our 2018 Greenbrier supply agreement. Our earliest available scheduled delivery under our supply agreements is in the second quarter of 2021. Remarketing income at Rail North America was $7.9 million for the quarter and $39.4 million year-to-date.

Within Rail International, both GATX Rail Europe and GATX Rail India saw steady demand for railcars during the quarter. GATX Rail Europe maintained high fleet utilization at 98.2%. The lease rate environment in Europe remains supportive of small increases in renewal lease rates for most car types. GATX Rail India grew its fleet to over 4,000 railcars while maintaining utilization at 100%.

Rail International's third quarter investment volume was approximately $45 million.

Turning to portfolio management. Results were primarily driven by a transaction at the Rolls-Royce and Partners Finance affiliates involving the refinancing and sale of a group of aircraft spare engines, which Brian will also cover in his remarks.

So with that, I'd like to turn the call over to Brian.

Okay. Thanks, Shari. Good morning, everyone. As Shari said, I want to provide some color on 2 items that have had a large positive effect on our earnings in 2020 and then we can go ahead and open up the line for your questions.

So the first one concerns Rail North America net maintenance expense, and coming into 2020, we expect the net maintenance to trend higher by about $8 million to $13 million. That's a 5% to 7% increase versus 2019. So the main driver of the increase was the commercial churn that we expected in the North American Rail fleet. Given the ongoing weakness in the market, and this was pre COVID.

By commercial churn, we mean a lower renewal success percentage on expiring leases. That results in more cars being signed to new customers to keep the fleet utilized and traditionally, that has meant more maintenance expense as expired cars often enter our maintenance network to prepare them for new customers and/or new service.

So as predicted, we have seen the higher commercial churn as we move through 2020, and this churn has been exacerbated obviously by the fallout from COVID-19. In fact, if you look at the third quarter, our renewal success was 58.1%. That's a full 17 percentage points lower than a year ago.

So however, year-to-date, net maintenance expense has remained flat for 2019. So there's a number of reasons for this favorable performance. Most of them are good. One of them, perhaps a little counterintuitive, but let me touch on those relevant reasons. So the first one, as we move through 2020, we continue to become more successful and increasing the amount of repairs done in our own network versus a third-party shop. So you've heard us talk about this before. We have this goal of moving as much maintenance work as is practical into our own shops. That's where we believe the safety, the cost, the quality and the delivery metrics are all superior. So as an example of the progress we've made, if you look at the third quarter alone, we set up 96% of our tank car work and close to 90% of all of our cars to run through the owned network. So that's steadily increasing volume has to drag down our unit repair cost in the owned network. We're going to continue to push on this initiative.

The second one is increasing the efficiency of our internal processes and our systems actually, and they continue to improve our ability to make sure that the type of work done on similar cars is consistent across the network. And when we can charge for that work, that we consistently build and collect the proper amount.

Now I've alluded to this before, these initiatives that have been driving down our costs for the last year or 2, obviously, we'll reach a plateau when we have achieved the complete uniformity across the network, but we're not quite there yet. Third, a little unpredictable, as always, are railroad repairs. They are lower-than-anticipated coming into 2020, but it appears that the railroad detention and manpower appears to be directed to other areas.

So the expense is down year-over-year. And lastly, on the maintenance side. And as I said, perhaps a big counterintuitive is the fact that sometimes the commercial churn in the fleet can cause maintenance expense to decrease versus expectations. So actually contradicting what I just said a little earlier, but there have been some cases in 2020 where the market is weak enough that we've made that economic decision to scrap older cars when customers return them at least then rather than incurred maintenance expense to prepare the cars for a new customer.

So why do we do that? We simply did not say a lease -- a new lease, being profitable enough to provide a return on investment in maintenance. So this is most common on the older cars and especially our boxcar fleet, which, as you know, is quite old relative to the rest of our fleet. So for example, coming into the year, we plan on maintenance expense to be incurred on target older boxcars because we anticipated being able to finally attract new leases.

Instead, the market weakened further due to COVID, and we ended up scrapping the cars when they came off lease. So that reduced maintenance expense versus expectations in 2020, but there's a downside to that and that it removes planned [boxcars running for] future years. So that's probably the best example of where reduced maintenance expense can be a little misleading, but nevertheless, I'm really encouraged by our maintenance performance, and I think that lower spending trend will continue in the fourth quarter and actually beyond.

The second topic I wanted to quickly touch on is a large gain on sale at RRPF, that spare engine leasing partnership with Rolls-Royce. We highlighted it in the press release that Shari said. Now we frequently realized residual gains in this business, but it's been in a variety of forms. So older engines have been torn down and profitably sold for parts. Excess maintenance reserves have been released at the end of leases and taken income, and we have sold engines on lease to third parties in the past. So similar to North American Rail, you can manage customer exposure, equipment exposure, renewal schedule exposure. You can optimize all that in the secondary market for engines.

It's quite liquid. The gain in the current quarter, though, was both large and unique, and I wanted to explain what it was and not let it mask otherwise difficult operating environment in that business. So as we've said in the past that RRPF, the portfolio consists of both of engines that are leased directly to airline customers around the world, but also engines that are leased back to Rolls-Royce. Generally, so Rolls can use them in support of their TotalCare program. In this particular instance, there was a large group of engines leased to Rolls-Royce where approximately $300 million of debt was coming due for refinancing in 2020 and 2021.

So obviously, refinancing rates have increased pretty dramatically for Air related businesses. So in this case, it made sense to restructure the current lease to Rolls-Royce into a new long-term lease, sell many of these engines with the leases attached to our third-party investors and then use the proceeds to pay down the vast majority of the related debt rather than refinancing it at higher credit spreads.

So once again, that addressed a number of goals for the JV. We reduced refinancing risk, we reduce some equipment risk, even lowered our exposure to Rolls-Royce from the JV. So -- but the value realized on the sale generated a gain of $0.68 per diluted share in the quarter, and it was actually an outstanding value, probably more reflective of a pre-COVID environment. So wanted to call it out because of a similar transaction on the engines lease to Rolls is unlikely to happen in the near future, but it does reflect the value embedded in the engine portfolio longer term. So I hope that helps explain 2 of these standout items in the third quarter earnings.

Operator, we can go ahead and open it up to questions now.

(Operator Instructions) We'll take our first question from Allison Poliniak of Wells Fargo.

Brian, I just want to go to your comments on moving the maintenance work into your own shops. I know it's something that GATX has been working on fairly successfully, but as you look at the environment today. Just trying to understand the dynamics, is that partly a function of the industry utilization of those cars that's giving you a little bit more, I guess, flexibility to bring them into house versus out in the field. Or is this -- I guess, is that irrelevant at this point? Just any thoughts there?

No. I think it's been an initiative of ours. Going back years, it was about 50-50 between the third-party network and the owned network. It just made more sense to us to leverage our investment, existing investment in maintenance facilities and bring as much as we could in. Historically, the vast majority of tank has been done in the owned network. Now we're doing a larger percentage of all repairs in the owned network. So I think, no, it's more our initiative to move repairs into the owned. And we do believe the safety, quality, delivery cost is all better. And obviously, we drive down our costs. I will say one of the things that actually increased maintenance expense over what we expected this year, there's a lot of ins and outs, but we did move a little more work into the contract network earlier in the year than we planned on, just to diversify that work in case we had a COVID outbreak that was severe enough to close down the facility for extended period of time. So actually, we could have done more, internally, if we hadn't done that conscious decision, but it's absolutely an initiative of ours.

And Allison, just to add to that, actually, the environment makes it more challenging to get the cars in because of the churn that Brian talked about. So you actually have more cars that need to get done.

Got it. Interesting. Okay. And then just -- I know 1 quarter doesn't make a trend, but the renewal success rate dropped a little bit from last quarter. Was that just the basis on what cars were coming in or something else going on there?

Yes. So it certainly is reflective of a challenging environment, but there is one item to call out there. So as you noted, the renewal success percentage was, what, 58% and it had been 70% in each of the first 2 quarters.

Part of that was driven by the Covia bankruptcy. So we expect ultimately to get about 500 cars back due to that, and then we'll have to remarket due to that bankruptcy. Have probably gotten around 300 back year-to-date. Without that, that renewal success percentage would have been more in the low- to mid-60s, which isn't too far off from long-term averages, but it's certainly below the last couple of years, which have been in the low 80s. It's important to note that about half the cars that we got back were immediately remarketed to other customers. The quarterly renewal success can move around quarter-to-quarter, as you noted. Because it can be influenced by a couple of larger contracts. However, in general, I would expect that the long-term average renewal success percent, which is sort of in the 65% to 70% range will probably be a better guide going forward than the 80% we've been at the last couple of years.

And we will take our next question from Matt Elkott of Cowen.

I wanted to make sure I understand the JV transaction. Brian, I think, did you say that all of the engines that were sold were Rolls-Royce engines? Or did they include certain percentage that was third-party customers?

I know half the third-party customers, I think, you mentioned on the last call, were on payment deferrals because of the current environment. So I was just wondering if there were any third-party customers. And if so, if any of those were the ones that are on payment deferrals?

No, they were all Rolls-Royce engines, and they were all leased to Rolls-Royce from the JV.

Yes. Matt, just, Matt, I just want to be sure you understand. Everything in the large transaction that Brian talked about was on lease to Rolls-Royce. There was -- there were other remarketing and residual realization activity that were different sources.

Got it. And did you guys give the number of engines that were actually involved in this transaction? Just trying to gauge the future impact on earnings.

No, we did not provide that.

Is there any -- can you help us try to gauge what the impact on earnings could be going forward from the sale?

Yes. That's -- that's not something we have right now. Let us look into it and see what we can provide.

Got it. And then another question on Rail North America. First of all, do you guys have a similar number of renewals coming up in 2021 to 2020? And if so, and if lease rates keep making modest sequential improvements like they did in 3Q, could we actually see a revenue per active car in North America hold steady next year or even improve slightly?

Yes, Matt, you know that we generally provide that information at our first quarter earnings call -- for the fourth quarter earnings call for the following year, but in general, directionally, given what we've been doing with lease terms recently, it's fair to say we'll probably have more expirations, more renewal opportunities next year than we did this year. And likewise, directionally, the expiring rate challenge probably will be a little bit easier next year than this year.

That's helpful, Tom. And then just one last one, I'd love to get your thoughts on the sustainability of the sequential industry fleet utilization improvement. I think we've seen 3 consecutive improvements over the last 3 months. So how much of that do you think is attributable to intermodal versus grain or other things? And what I -- what it means to you guys?

Yes. So coming into the year, we anticipated a slow and gradual recovery as far as lease rates go. And although car loadings were up 11% versus Q2, they were still down 12% versus Q3 2019. However, on the supply side, we are seeing some builders retrench, and we are seeing some scrapping activity increase.

In fact, the net North American fleet declined slightly per the most recent [under data] for the second straight quarter. And the industry metrics on idle cars and storage declined by 75,000 cars. So those are all positive signs. However, there's still too many idle cars in the industry. So even though we saw a flat to marginally improving lease rates in the quarter, we still have a long way to go to get back to those long-term averages.

So absent an unanticipated demand catalyst, like we saw with crude oil in the last upmarket, we anticipate it will probably take several quarters before the supply correction mechanisms can meaningfully increase lease rates.

So several quarters of rail traffic improvements, rail traffic going in the right direction, decelerating and then growing potentially next year?

Yes. So again, it's hopefully the beginning of a trend, but it's early innings, so we'll have to see how that develops.

And we will take our next question from Justin Long of Stephens.

Brian, some of the comments you provided earlier around maintenance were helpful. I wanted to see if we could get a little bit more color on what you're expecting going forward for North American maintenance expense. I think you said that some of the improvement should be sustainable into the fourth quarter and going forward. Does that mean that maintenance expense can remain kind of flattish sequentially next quarter and into next year? Or is there a little bit more color you can provide around that order of magnitude?

Yes. Our expectation is flat to down. That trend to continue at least for the short term, but I don't want to project too far out because so much of it is dependent on commercial success. But looking to next quarter, I would think about this trend continued.

Okay. That's helpful. And then next year, just with some of the tank car recertification work that could be coming up -- could be coming up in 2021. Do you think something kind of flattish for maintenance expenses is possible relative to 2020? Or is there a ballpark you can give us on that?

Well, we'll give it to you in January, but I will say we've pulled forward a lot of that compliance work on the tank certification into last year and this year. So we're -- we've evened out that workflow more. So I don't expect a big increase, and I would hope this trend continues.

Okay. That's helpful. And following up on the gain in RRPF. Tom, you mentioned there was some other kind of remarketing and residual gain in the quarter. Could you provide what that number was? And then I don't know if you have the pretax number for the larger game, but that would be helpful as well as we kind of put together those different pieces?

Yes. So I do have that. So maybe what I'll give you, Justin, is both the year-to-date and third quarter numbers, which hopefully will be helpful. So year-to-date for the JV, income from operations that were recognized on that line item were about $30 million and gains were about $63 million, which add them together, you get the $93 million that you see there. For the third quarter, gains in operations were $10 million -- operations earnings was $10 million and the gains was $37 million. Those are all pretax, so for a total of $47 million. So of that $37 million in gains, about $32 million related to the item that Brian talked about being the $24 million after tax. So relatively modest across everything else. And just to come back to Matt's question, probably no reason we can't tell you that it was 18 engines that were sold.

Okay. And on those 18 engines, anything that we should consider in terms of like a disproportionate impact on revenue? Or if we look at those engines as a percentage of the total engine portfolio, is that a good way to kind of ball park the impact going forward?

Yes. That probably won't work too well because, as Brian noted, they were older engines that were -- it was good to rebalance the portfolio with those. So I would not say they're a representative cross section.

Okay. All right. Very helpful. And last question I had was on the acquisition pipeline. Obviously, we've seen some extreme volatility in both directions in the economy and rail volumes, et cetera, but as the market has started to bounce back here in the third quarter, have you seen more acquisition targets come to market? And maybe you could speak to the valuation multiples on deals if you're seeing a pickup?

No. We really haven't seen anything significant. So there's not much to talk about there. I still think there's opportunities for consolidation. I don't [want to sound] -- be a broken record, but there's just too many investments made by new players and aggressive players that are economically under water. So I still think it will come, but there's really been nothing significant. I mean you haven't seen anything change in probably since our Element transaction. So really nothing to report there.

We will take our next question from Bascome Majors of Susquehanna.

Yes. Going back to the Rolls JV, can you give us an update on perhaps the percent of revenue that's being deferred right now? Just trying to think of the delta between revenue recognized and cash flow. And if that has shifted any in the last 2, 3, 4 months?

Yes. So customer deferral requests have really continued to slow, and the numbers are similar to what we've talked about previously. To date, about half of the JV's airline customers have requested deferrals. The JV is selectively granting those deferrals on a case-by-case basis. The requests, again, as a reminder, typically 3 to 6 months’ rent deferral. Some have been as long as 12 months. And then as far as a percent of revenue, the request represent a little over 10% of annual revenue.

And does this outcome with perhaps older engines that were closer to end of life, but a fairly significant gain for just 18 of them. Does this give you some confidence that you can share with us that impairing issues or perhaps overblown by some investors in that portfolio as you look to year-end testing?

Yes. So what I can tell you is there's been no material impairment so far this year. The JV's impairment analysis consists of cash flow analysis for each engine type and independent appraisals. As you know, the accounting rules require impairment to be taken whenever the testing indicates impairment.

Yes. A little color around that transaction. So that started very early in the year. I think the values were probably more reflective of the pre COVID environment. So there's no question, airline and engine values have come down. We still think we're strong long term, but I don't know if that transaction is immediately repeatable at those sites.

Thank you for the candid color there. Just 2 more on North American Rail. You talked a little bit about having more renewal opportunities next year, and hopefully, with the more moderate expiring rate to comp against. Can you just give us a little finer detail on the cadence of the expiring rate on your portfolio as it sits today with understanding on our end that, that may change depending on transactions, et cetera?

Yes. To a degree, Basc, and you answered your own question. The challenge for us on providing that too far in advance is it does change, which is why we went with -- we'll provide the directional guidance that we would expect it to be higher than this year, but it's really hard to say that as transactions are constantly being done.

Meaning your expiring rate next year will be higher than the rate this year.

Yes. I'm sorry, I thought you were talking about the number of expirations.

Understood. Well, I was actually asking about both. So any color you could give would be helpful.

Yes. And so it opposite directions. We expect more expirations, more renewal opportunities next year than this year because of the short lease term. But as we get further and further away from the upmarket, we expect the comparator rate, the rate it's expiring off of, to be lower next year than this year.

Last one, back to M&A, focused on North America, but maybe not exclusive to it if you have some different comments for another region, but understanding that -- I guess, it sounds like from your earlier response to the M&A question that the motivated sellers really aren't showing up yet, and it's understandable, some of the reasons. Can you help us understand maybe your process? Like when you're kicking the tires on one of these deals, like financially, what is your objective? Is it to -- how much can this improve our ROE? Is it an earnings accretion, estimate? Just if we could get maybe a peek into your process and what a "good deal" will do for your financials? I think that would help us understand kind of where your guys' heads are.

I mean that could be a disappointing answer. Really accounting earnings is the last thing we look at. We have a different risk-adjusted return of greater return required for every geography, every business, not every car type, but every business. So we'll run through our projections and we'll totally focus on shareholder value added. We express that. At GATX, we look at it as what we call ROI, so it's a return based on original investment, but it's correlated with [SVD], shareholder value added. And accounting earnings fall out. Honestly, we're not going to let that stop us from doing an economically accretive transaction. And the reason I say that, it's important to note that any kind of young portfolio of railcar leases generally is not very accretive from an accounting perspective early in the life.

So you just can't focus on that, but it materializes over time. So yes, it's -- there's nothing really to report in North America. I think there are -- you need a motivated seller, right, I can't make somebody sell. We can certainly make our interest known. And I think the market knows about our interest in acquiring portfolios at attractive prices, but I think people have to come to grips with the value of their fleets, especially the way some of them are composed in this industry.

In Europe, it's a little different story. I think most of the growth, I don't say most of the growth has been organic. I have been a little disappointed over the last few years that we haven't seen portfolio acquisition opportunities in Europe. They just haven't materialized in a couple of years. There's a couple of things, though. Right now, the European teams are working to develop a more liquid secondary market in Europe, similar to one in North America, it just hasn't been there to date. We did a small [flake] car acquisition earlier in the year, but that market is going to take some time to develop.

The second possibility in the Europe is SNCF, the French National Railroad may sell all of our partner, (inaudible) that's been discussed in the market for years. But just in the last few days, there's been stories in the press saying that the SNCF or has made a decision to sell all or part of (inaudible). So we should know in the near future if that's a possibility or not. So looking at North America and Europe, that's the only thing being talked about right now.

And we will take our next question from Justin Bergner of G. Research.

Tom, to start off, I guess, before Rolls announced their major set of transactions to raise equity and debt capital. One might have expected that GATX could have been a source of liquidity for them somehow through changing ownership or other sort of financial transaction to the JV.

Is that a possibility for GATX going forward to increase its exposure to the RRPF JV and help provide Rolls' liquidity in the process? Or is that sort of now no longer relevant given Rolls' announced transactions?

Well, I definitely think the liquidity pressure is off on them, right. They upsized that bond deal, I think, just in the $2 billion from $1 billion. They did the rights offering. I think it's all part of that I think it's a EUR 5 billion plan to raise liquidity. So it seems like they've gotten through that. At least the market has responded positively to it. So given its long history of impressive growth and profitability and with a robust outlook for growth post COVID and engines. Now I look forward to continuing to invest in this business. But for right now, I'm focused on being the best partner I can to Rolls-Royce. So I wouldn't rule it out. But honestly, right now, given engine values, I think, that's unlikely that Rolls' would want to exit that investment because I think long term, it's got a very attractive outlook. So I'm just focused on being the best partner I can.

Okay. Understood. That makes sense. Moving on to the topic of the maintenance expense. I think you said that the overall maintenance was done 90% through your own shops this quarter. Just big picture what changed over the last few years that has made the economics inherently more favorable to bring in-house non tank car maintenance when beforehand it was outsourced. Was this always an opportunity just took a while to how to do it? Or was there something that changed in sort of the economics to make this a better opportunity in-house?

Yes, Justin, it wasn't so much a change in economics because doing the incremental car in your own shop is always a better thing than spending all that money externally for an outside party, but we've put some investment into our shops so that they can handle more capacity. Done things like putting track down for storage and movement of the cars, increasing our cleaning and finishing capabilities. So we've put some money into the shops to be able to take -- to increase their capacity to take on cars.

Okay. And that 90% number, I mean, that's obviously extraordinarily high, if I've heard it correctly. Is that sort of a sustainable benchmark? Or was that enhanced by just the weak operating utilization environment for the major rails?

Yes. So again, our target has for a while been to go from the 75% or so of tank and specialty freight maintenance up to that 90% number. So that certainly remains our target. And as we mentioned, I wouldn't characterize the environment that we've been in as particularly easy to get cars into the shop because we've talked about the fact that when the renewal success percentage is lower, you see a little more churn in the fleet, more cars have to visit the facilities. And then there also have been some challenges related to COVID where occasionally, we had to close the shop for a day or 2 to do some deep cleaning or other activities like that. So it really is -- this has been a challenging environment to get cars through.

Yes. And I should pile on there and say, that's something we haven't talked about. I think it's a rough estimate between direct cleaning, between paying employees that are out on quarantine and a variety of other things we've done around COVID. I mean we estimate $3 million to $5 million of expense this year. So it really does put into perspective how impressive the maintenance performance is year to date to do all this in the face of COVID.

Got it. Yes, good job there. Just to wrap up, 2 quick ones. Did you repurchase shares this quarter? If so, how much? And then secondly, can you set long-term goals on sort of the desired size of your India fleet? And if so, are you willing to share with investors at this point?

So we did not purchase any shares this quarter. We still have $150 million on our existing authorization, and I'll turn it over to Brian for India.

Yes. On India, had very little growth in second, third quarter. In fact, I think we had 0 deliveries in the second quarter, like 125 in the third, probably delivered half of what we expected in 2020 coming into the year, but to the extent that COVID doesn't create a major shutdown, it could end up just being a blip in their growth for the next -- for the last and ongoing 12 to 18 months, but longer term, still got the same growth prospects. It was a great market coming into COVID and hopefully, it should be a great market coming out. So looking out 4 to 5 years, there's no reason that we could, for instance, double the size of the fleet.

We will now take the question from Barry Haimes of Sage Asset Management.

I had 2 questions. One is you mentioned that the rate on expiring leases next year should be lower than what we saw this year. Since that's kind of a known number, can you give us a little bit more help on magnitude of that? And I had one other one.

Yes. So unfortunately, we really can't because again, the mix of cars coming off is constantly changing as we do things like have remarketing activity. Some of the cars that go into the shop, you have the scrap versus repair decision. There's just a lot of things that make it hard to quantify that. So we really need to just stick with the directional guidance.

Okay. And then the other question, just following up on the fact that lease rates were flat to up sequentially. So maybe the market's stabilizing some. But if we were to look at by car type of your major car types in terms of demand versus what's in storage, what are the car types that were maybe this a little less in storage, and you might see rates firm a little sooner versus which of the car types where you kind of have to slog through a bunch of supply and storage for us before you get that improving?

Yes. So relative to last quarter, the car type that saw the greatest increase was were probably grain cars, expectation of a good harvest. And so that really was the car type that I would call out for -- versus last quarter. When you want to talk about versus sort of long-term norms, the car types that have performed the best throughout have been general service tank car types that are not serving the energy market.

We will take our last question from Steve O'Hara of Sidoti & Company.

Yes, can you hear me?

We can hear you.

Okay. Good. Yes. So just on the sale of the engines, did you say what the dollar value of the sale was?

We did not.

And do you have that by any chance?

I don't have it in front of me.

Okay. And I think there -- so you said there was 18 engines sold, and there's approximately 478 in the portfolio, is that correct?

Correct.

Prior to that. Okay. Okay.

Yes.

All right. And then just on the -- maybe just -- if you look at remarketing gains and things like that, that have happened and where maybe railcar values are now versus several years ago.

Has there been any -- I mean is there -- I think, going forward, do you see your ability to generate those gains may be somewhat suppressed versus prior time periods?

I mean I think if I look back, at some points, you were doing kind of north of almost $100 million in 2015. That was obviously a very strong year, but 2018, I think it was $73 million. I mean is that -- are we more likely in this neighborhood now in the maybe $50 million to $60 million range, do you think? Or just given what the market is doing? How do you think about that?

Yes. So as you know, from following us for a while, the gains on asset sales really can move around quite a bit quarter-to-quarter and year-to-year. And it's really hard to say what they'll be for any given time period. As you know, the way that we look at cars that we identify for that is really a portfolio management activity, where we're looking for car types that we might be a little long on, either because of where we think the market's going or the exploration profile or the -- just the car type or the credit. So that's really how we identify cars that we're going to put into a sales package and calling exactly how that gain will move over time is pretty imprecise.

It's so opportunistic. I mean the best example would be we sold -- I don't know how many, but a lot of small cube covered hoppers back a few years ago before the current crisis. And the ability to do that disappeared pretty quickly. So it's so depending on the commercial environment. But we -- it's very opportunistic on our part. Let's get the managed customer exposure, commodity exposure, equipment exposure. So a lot of that can change in a hurry and how perceptive the market is for that.

Okay. That's helpful. And then just on the -- the -- what was it, sorry, within -- on the Rail North America side, I think, you said lease rates are up sequentially. And -- but did you kind of talk about where they were versus kind of the lease rate maybe kind of coming into the year kind of in average terms or year-over-year?

Yes. So versus a year ago, most tank car types are probably down around 25% or so. Freight cars have been challenged longer. So they're down less because they've been in a challenging environment longer. They're probably down around most car types, around 10%.

Okay. All right. And then if there was any -- within the sale of the engines that Rolls, the JV, was there any dividend paid on the -- any cash proceeds? Or was that reinvested? Or how was that dealt with?

Yes. The cash from them was used to pay down the associated debt.

There's no kind of left over cash after that.

Correct.

And we do have a follow-up question from Justin Long of Stephens.

Tom, just had a couple of things I wanted to clear up on the model going forward. For SG&A, do you feel like the third quarter is a good run rate for fourth quarter and beyond? And then also wanted to get any thoughts you had around the tax rate going forward?

Yes. So as far as SG&A, we came into the year expecting it to be around $180 million if you account for the ASC sale. Through 3 quarters, we're at $126 million. So if you look at it on a run rate basis, we would finish around $10 million to $15 million below that original expectation. And I think that's probably a reasonable way to think about it. Of course, there's things that can make that move around a bit, but that's probably a reasonable way to look at it. As far as the effective tax rate, we would expect that to end the year somewhere around 29%, 30%.

And we have one other follow-up question from Justin Bergner of G. Research.

I just want to make sure I heard you that industry participants, from your understanding, are moving to sort of scrapping more cars? Because it looks like GATX didn't scrap any of its cars in the third quarter or year-to-date based on a line item here. Maybe just any additional comments there?

Yes. So first of all, as far as GATX, on the last page of the press release, we provide the information on the cars we scrapped in the quarter, which were 623 cars this quarter. As far as the industry goes, what we would anticipate is with a little higher scrap price.

And scrap prices are up over the last couple of months where they're around 245 now, which is pretty close to the long-term average. They had been more in the 200 range.

As those go up when people, like us, make our scrap versus repair decisions, it makes it relatively more likely you're going to scrap when you compare those proceeds to what you can get from repairing the car. So our expectation would be, in general, if scrap prices stay where they are or increase, you would see a little more scrapping activity in the industry.

Okay. That makes sense. It was just that -- it's just that you didn't generate any gains from that activity, but you still scrapped a good number of cars.

Correct. We scrapped a little over 600.

Speakers, at this time, we have no further questions.

Great. I'd like to thank everyone for their participation on the call this morning. Please contact me with any follow-up questions. Thank you.

Thank you. Ladies and gentlemen, this concludes today's presentation. You may now disconnect.