GATX Corp (GATX) 2018 Q3 法說會逐字稿

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

At this time, I would like to turn the conference over to Jennifer McManus. Please go ahead.

Good morning, everyone, and thank you for joining GATX's 2018 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 discussed in GATX's Form 10-K for 2017. GATX assumes no obligation to update or revise any forward-looking statement to reflect subsequent events or circumstances.

Earlier today, GATX reported 2018 third quarter net income of $47 million or $1.22 per diluted share. This compares to 2017 third quarter net income of $49 million or $1.25 per diluted share. Year-to-date 2018, we reported net income of $162.1 million or $4.21 per diluted share. This compares to $159.9 million or $4.04 per diluted share for the same period in 2017.

2018 year-to-date results include a net negative impact of $5.8 million or $0.15 per diluted share attributed to costs associated with the closure of a railcar maintenance facility in Germany. 2017 year-to-date results include a net gain of approximately $1.1 million or $0.03 per diluted share associated with the planned exit of the majority of Portfolio Management's marine investments. These items are detailed on Page 12 of our earnings release.

Now I'll briefly address each segment. Our third quarter results are reflective of a continued improving operating environment in Rail North America. Certain rail industry metrics are favorable for lessors relative to 2017, including railroad carloads and velocity. Rail North America's fleet utilization increased to 99.2% and our renewal success rate was 82.9% during the quarter. The renewal rate change of GATX's Lease Price Index was negative 11.5% with an average renewal term of 33 months. Revenue pressure continues. However, we did see quarter-to-quarter sequential improvement in absolute lease rates across most car types in our fleet.

As indicated in the earnings release, this improving environment has resulted in lower railcar maintenance expense in 2018 than we originally anticipated, especially for mandatory tank qualifications, as customers are holding on to and utilizing their cars. While this is a benefit in 2018, we expect this will result in higher maintenance expense for 2019 as this work has only been deferred, not eliminated.

We continue to successfully place cars from our committed supply orders with a diverse customer base. We have already placed over 7,700 railcars from our 2014 Trinity supply agreement. And scheduled deliveries with customers have been placed through July '19, meaning our earliest available scheduled delivery is in August 2019. Additionally, we have placed 450 railcars from our 2018 ARI supply agreement, and our earliest available scheduled delivery under this agreement is in the first quarter of 2020.

The secondary market remains active. Rail North America's remarketing income was approximately $7.2 million during the corner (sic) [quarter], bringing total remarketing income for the year to $61.7 million, which is slightly above our full year expectations. While we are always active in the market, we expect any activity in the fourth quarter to be modest in size.

Within Rail International, the European railcar leasing market continues to experience gradual improvement across the chemical, petroleum and freight markets. GATX Rail Europe is seeing steady demand across the fleet with utilization increasing to 98.4%. Rail International's investment volume was approximately $40 million during the third quarter, with over 1/3 of this due to investments in India. We are still on track to double our fleet count in India by the end of the year.

Portfolio Management's results were primarily driven by the strong performance of the Rolls-Royce & Partners Finance affiliates. RRPF's results were driven by both higher operating performance and remarketing activity in 2018. Year-to-date, segment profit in Portfolio Management is down from 2017, primarily due to the $8.4 million residual sharing fee earned in the second quarter of 2017.

American Steamship Company continues to perform well, with 10 vessels currently deployed. Higher water levels and increased demand for iron ore has more than offset the initial delays ASC experienced early in the season.

Finally, GATX repurchased nearly 150,000 shares for approximately $12 million during the quarter. Year-to-date, we have repurchased over 500,000 shares for $37 million. At the end of the quarter, approximately $213 million remains available under the aggregate $250 million repurchase authorization.

Those are the prepared remarks, so I'll hand it over to the operator and we can open it up for Q&A.

(Operator Instructions) The first question comes from Allison Poliniak with Wells Fargo.

On the 2019 maintenance expense, is there any way to help quantify what you expect to get pushed into '19 for us?

Allison, it's Brian. No, not at this point. We're trying to get as many cars in as -- before the end of the year as we can. So we'll have an update for you, obviously, at the end of the fourth quarter. But it really is much lower than we expected coming into the year. And the reduction, it's both the tank qualification work which has not yet materialized. As Jennifer said, they're really reluctant to send in their cars. You're looking at a market where car loadings are increasing velocity and the railroads are slowing and trucking is untenable right now. So customers are really reluctant to give up their cars to an increasingly busy maintenance network. So we anticipate getting a lot of this backlog in, in the fourth quarter. But where we sit today, well, we're not going to get all of them that are due. We can just tell that's not going to happen. They're not going to get in, in time. So what happens is our commercial teams, and eventually, legal teams, who work with the customers to get them in, and at some point, if these cars are not received by GATX, they will be embargoed and then unable to load. And that's when you see them flood into our maintenance facilities. So we're trying to get in as many as we can before the end of '18, but we can tell you right now it's not going to happen as far as what is due. So that's why we can already see increased maintenance expense coming in '19.

Understood. And then you talked about sequentially improving lease rates. Obviously, your term is still pretty low. Could you give us some perspective in terms of where we are relative to, say, maybe the average lease rate across your portfolio? I know it's probably a little bit more difficult.

Yes. I always get admonished internally about trying to get too generic with that because they have such a diversified fleet and it really does depend on car type. But ignoring that for a minute, if you look sequentially quarter-over-quarter for absolute market lease rates, it probably declined a few percentage points for freight, and that's really due to small cubes having a sizable decrease in the quarter. But it increased pretty seriously for tank. I think over 20%, 25% or more absolute lease rate increase on average for tank cars just in the quarter. So we are seeing significant strength in the market.

(Operator Instructions) The next question will come from Justin Long with Stephens.

So wanted to start with a question on the guidance. Thinking about this quarter in the commentary, it seems like in North America, you've seen remarketing income and maintenance expense both outperform your expectations. So I was a little surprised that the full year guidance range didn't change. Can you just help us think through that and the areas of the business where you might be seeing weaker than expected trends that are offsetting some of that upside?

Yes. This is Tom, and I'll address that. So as you mentioned, we continue to expect full year earnings to be in the $4.90 to $5.10 range. Some of the key drivers of the quarter-over-quarter decline in Q4 will be the lower remarketing income in Rail North America that we talked about; the maintenance expense catching up from deferred maintenance that Brian just mentioned; at ASC, we have some incremental cost from putting an additional vessel into service; and historically, the fourth quarter is a more challenging operating environment, and we expect that to continue to be the case. It's also important to note that we will continue to see a drag from lease rates in Rail North America. Even though the absolute lease rates are increasing, as Brian said, they're still below, for the most part, expiring rates, as evidenced by the negative LPI. So we continue to see a drag on incremental renewals and we see the full quarter impact of the negative drag that we had on third quarter renewals going forward. So those are some of the reasons.

Yes. The only thing I want to add there. I know it's -- I know what you meant by outperformance in maintenance, but really, from our perspective, it's underperformance in maintenance in 2018. We'd really like to get those cars in this year because they're due. So we're not celebrating the fact we couldn't get them in. It's actually not a good thing, it's just deferring maintenance into a subsequent year.

Yes. That makes sense. That's all really helpful. And maybe to ask the earlier question a different way, like, do you know the -- or do you have a number on the total dollar amount of maintenance expense that has been deferred year-to-date? I know you're -- it's tough to predict when this will actually be deferred or materialized in 2019, but what was the total dollar amount of -- that was deferred?

Yes. Don't know yet because cars are coming in now in the fourth quarter more than they were earlier in the year. But the problem is, if that they start coming in now or to the mid-November, it's too late to turn those cars for them in the year because, not just our maintenance network, the industry's maintenance network is getting increasingly busy. So generally, a couple of hundred cars that are due, tank cars that are due for tank quals, in a given year spill over into the next year. And you work with the customers to get those in as soon as possible, otherwise, they won't be able to load them. We see that going up multiple times in 2019, but I can't tell you how much just yet. And to give you an indication, that tank qual, by the time it goes through the maintenance network and attracts all the other work that may be due on that car, you may have to do on that car for other reasons, it could be well over $10,000 per maintenance event. So it's a sizable number, I just don't know the exact amount yet.

Okay. And then maybe secondly, I wanted to ask about the rollout of the PSR operating model. Obviously, it's becoming more broad-based across the U.S. with UP's announcement in September. Do you have any thoughts around how that announcement and this operational change could impact your business, and just the broader railcar leasing market as well?

Yes. So other railroads have implemented PSR in the past and they have, generally speaking, shed some cars. So it's reasonable to assume that UP will look to do -- will look to reduce the size of their feet. Having said that, we don't know the exact manner in which the UP will implement PSR, so it's difficult to make accurate predictions regarding any changes to their railcar needs. As with all customers, we spend a lot of effort optimizing our fleet exposure with them to achieve an attractive risk-return balance. We have a strong relationship with the UP and we're comfortable with our expiration profile. A great example is boxcars, which is a car type that historically been exposed to PSR implementation. The vast majority of our boxcars have long lease terms and don't start expiring for another 5 years. And even when they do expire, most of them are near the end of their useful life, even if returned. As it relates to UP, we think our biggest risk is other lessors getting cars returned and GATX having to compete against those idle cars in the market. And I would say that, that general theme, that last item, is the thing that we're most concerned about generally with PSR, is knowing exactly how railroads will implement it and how we might have to complete -- compete with additional railcars in the market.

The next question will come from Matt Elkott with Cowen.

Tom, you mentioned that despite the sequential improvements we've seen over the last few quarters and spot lease rates, renewals continue to be a drag, if we can -- if this spotless -- spot lease rate improvement continues into 2019 and through 2019, can we expect that drag to turn into a tailwind in 2019?

Yes. So basically, you're asking about when the LPI will turn positive and that event is an extremely difficult thing to answer because it's a constantly moving target with additional renewals coming online. That will be something that we will update you on at the January earnings call. We -- it's difficult to get too far ahead of that. But at this point, it's difficult not to see continuing rate pressure in that regard in 2019.

Got it. But you still believe in the sustainability of the -- in lease rates on a spot basis for the foreseeable future.

Yes. I mean, particularly as it relates to tank cars, we've now seen several quarters in a row of positive incremental lease rates, and we would expect that to continue.

Got it. And speaking of tank cars, I'd love to hear your thoughts on the current tightness in the DOT-117 supply. And if you can -- if you're able to tell us how many of those cars you guys have in your fleet and if you plan to add more. And whether you would add newly built DOT-117 Js, or you would consider adding retrofitted ones.

Yes. So I'll start at the end of that and then work my way towards the beginning. And if I miss any of it, let me know. So we are adding DOT-117 cars to the fleet, newly built cars. And Jennifer will have that number.

We have about 2,100 right now in our fleet.

As it relates to retrofit, we do not intend -- and we've said this before, we do not intend to retrofit any legacy 30,000-gallon cars. What we have done is, and we'll continue to do, is retrofit of the CPC-1232 jacketed car, which, as you know, is the very inexpensive, just a couple thousand dollars, to modify that car. As far as what's going on in the market, the order information just came out this morning, and there were, on a net basis, a net 10,000 tank car orders. We expect that a significant portion of those are tank cars for the Canadian crude market, and would expect that to continue. And as far as how we will invest and how we'll look at that market, just like we've said in the past regarding this, we will invest in a measured way, being careful not to get overexposed to a car type which ultimately will probably be oversupplied as pipelines will ultimately come in and because it's the most efficient way to move that commodity. The length of time it will take is difficult to predict, but on the scale of railcar leasing investments, it's short enough that we will not get overexposed in that area.

Got it. That's very helpful, Tom. I just have a one quick follow-up. I understand about you guys resisting the temptation to go in, in a big way on the DOT-117s. But are there conversion opportunities to ethanol? Because the ethanol fleet has to be replaced with DOT-117s, I believe, in 2023. So even if crude by rail, Canadian crude by rail, is somewhat of a stop-gap measure until more pipe comes online, are there conversion -- can you modify at least a certain variety of the DOT-117s to ethanol? And if that's the case, will that be the type of DOT-117 you would be investing in?

Yes. And I don't want to get too much into the weeds on this. But we have and we'll continue to do some investment in DOT-117s for ethanol. The particular car that serves Canadian crude and the car that serves ethanol, it's not configured the same way, so that particular move doesn't -- conversion doesn't really work. But certainly, opportunities to place non-jacketed cars in ethanol is attractive to us.

Yes. And Matt, I'd also add that cars that are -- the 117s that are going into Canada -- Canadian crude, in our fleet anyway, are 25.5s. They're lined and they do carry other commodities. So it's different than the legacy 30s, what everybody invested in years ago.

The next question will come from Matt Brooklier with Buckingham Research.

So just wanted to dig a little bit deeper in terms of your -- the commentary on lease rate improvement on the tank car side of things. I'm assuming a big portion of that is flammable service cars. But could you maybe talk about the nonflammable service tanks in your fleet, or from a market perspective, if we're also seeing improvement there? Or is it, right now, it's all kind of coming from the flammable services side of things?

Yes. Let me preface that by saying, when I said tank increased over 25% on average in the quarter, I was excluding legacy 30s, as an example. So that's not an investable car type, obviously, it's on its way out. So broad improvement across the tank car market. And probably the best performer overall, Tom, would be the high-pressure car in the quarter.

Okay. And then could you -- you talked to some weakness on the small cube, covered hopper side of things, specifically what's happening in the sand market. Do you mind giving us an update in terms of what GATX's exposure is currently?

Yes. About 2% of our fleet is in sand service. We've talked about this before, with the move from the northern white sand to the Texas brown sand. This has been coming, and it's impacting the industry in a significant way already and will continue to grow, which is why we kept our exposure fairly modest. And what we do have is termed out pretty well.

Okay. And just a last question here on the tax rate that was up in the quarter, I think a little bit ahead of our expectations, and I think, your original guidance. Maybe you could just talk to some of the components of what drove the tax rate up in the quarter. And how should we be thinking about fourth quarter?

Yes. So last year, our effective tax rate was 38.4%. We expect it to be at 25.1%, which is a little bit different percentage than the Tax Act. And those differences are primarily influenced by our mix of domestic to international business.

Okay. And then rough guidance on where we think the tax rate falls for fourth quarter?

So for the full year, with that 25.1%, that's probably about the best guidance we can give for the fourth quarter.

The next question will come from Mike Baudendistel with Stifel.

I just wanted to ask you, is there anything to read into your average lease renewal term only being 33 months? I gather some of that is just lease rates aren't back to where you'd like to see it. But is there anything in terms of just mix of the types of rail cars that were renewed during the quarter being a little bit on the weaker side?

Yes. So one of the things we've talked about in the past is it's difficult read too much into any single quarter. Overall, as we mentioned earlier, most car types are still below their long-term averages, so we're still generally trying to go short. What everybody saw I'm sure is the decline from second quarter to third quarter. But if you go back 3 quarters before that, they were all kind of in the low to mid-30s, the 41 last quarter was the high one, and that was really driven by a couple of unique transactions, that are, for commercial reasons, were at long terms. The term has pretty much -- with the exception of those couple of unique transactions, the term has been pretty consistent over time. There are a couple of car types where rates are good enough that we're starting to try to push term. But at this point, it's not broad enough across the fleet that you've seen an increase in that number.

Got it. That makes sense. And then just also wanted to ask you, on the long-term agreement you signed with ARI, just any additional detail you can give. Sort of why now? Why use ARI? And is there anything in the agreement that -- it seems like it looks similar to the Trinity agreement in a lot of ways. But is there anything that's significantly different there in terms of does the price of the cars change with market conditions? Any of those things?

Yes. So we definitely thought it was the right point in the cycle to pursue supply agreements. As you know, we did 2 of them. We did the one in May with Trinity and then the one in August with ARI. Part of our strategy of supplier diversification. And they're both high-quality builders with a great catalog and great capability. As far as the specifics of the agreement, can't disclose a lot of details there. What we can tell you is that they're both cost-plus agreements. The Trinity agreement is 4,800 cars over 4 years, the ARI agreement has 450 cars in 2019 and then 1,800 cars a year from 2020 through 2023. As Jennifer's already mentioned, those 450 cars in 2019 have already been placed.

(Operator Instructions) We'll take our next question from Justin Bergner with Gabelli & Company.

First question just relates to ARII. In light of the transaction announced yesterday, I was curious if you had any comments on that. And just given that the price was pretty high, how does that sort of affect your capital allocation going forward in terms of repurchases versus adding to the lease fleet?

That's a multipart question. I'll take the valuation part of that. Clearly that's a manufacturer, Justin, it's relatively small from a lessor perspective, so it's definitely not the same thing. I think it went out at 2x book, so it looked like a good price for them. As far as the other half of that, Tom, you want to?

Yes. So earlier in the year, we provided guidance that our stock buyback was targeted to be in the $100 million range. As I think everybody knows, we were blacked out for Q2 and a portion of Q3 due to negotiations with our 2 supply agreements. We are back in the market and we will try to reach that $100 million goal by year-end.

Okay. Just a quick follow-up clarification, another question. The tax rate guide, how does that compare to your earlier tax rate guide? And just any rough estimate as to what the long-term tax rate is as you look out into '19 and beyond.

Justin, I would say, for 2018, our guide was the effective rate was 25% for the year, and that hasn't changed.

Okay. And it's safe to sort of assume a similar rate looking out to next year?

Well, as the U.S. -- assuming the U.S. business continues to recover, that's at a higher tax rate. So you could say -- we don't quite know yet, but directionally, it should go up over time.

Yes. And we would provide that guidance in our Q4 call for the 2019 tax rate.

Okay. Great. With respect to Portfolio Management, it looked like the income from affiliates was down year-on-year as well as quarter-on-quarter. But you mentioned in the press release that you're very pleased with the Rolls-Royce joint venture performance. So just any comments on how to true up the decline versus the positive commentary.

So Justin, first, I would say year-over-year, share of affiliates income's up. So it's $38 million to $47 million for the year, just so we're clear on that. Quarter over quarter...

That's year-to-date?

That's year-to-date, right.

Okay.

Okay. Q3 last year versus Q3 this year, we'd had higher income from operations. But in the quarter this year, we had lower remarketing income. So generally, Rolls is performing extremely well. It was just, as we said in the past with Rail North -- it's very similar to Rail North America with remarketing income, that it's lumpy. And so that's what we're seeing in Q3 this year.

The next question will come from Willard Milby with Seaport Global.

Looking at the North American fleet, and I don't know if I'm reading too much into this, but the non-boxcar fleet growing over the last 2 quarters. Is this -- could this mark the bottom, I guess, of you all being a net disposer of railcars? Do you think the market conditions are right for you to kind of grow that fleet? And tagging onto that, as you look at the open market for pools of railcars to potentially acquire, what do valuations look like now versus maybe a year ago? And are those valuations attractive for fleet additions?

Yes. I can take that. I mean, that's why we placed the 2 orders this year, we do intend to grow the fleet going forward. As far as the secondary market, we haven't seen any slowdown there at all, pricing is still very strong, and that's why our activity's been so high over the last 2 years. In a situation where the market's been oversupplied and lease rates were still relatively weak compared to historical levels. But prices in the secondary market are -- have hung in there and are still very healthy. And as far as what would cause it to slow down, I mean traditionally, access to cheap capital has been the historical governor on that. So if credit tightens, the rates increase, you might see a pullback. But so far, it's quite healthy. And so I would think that you would continue to see -- as long as it's healthy, you'll continue to see us generate remarketing income.

Okay. So the -- I guess the recent orders with Trinity and American Rail, those will be net additions rather than strictly, I guess, replacements. Kind of, I'm just curious about the multiyear trend that you have been a disposer of railcars for the most part, if that's starting to flip.

Yes. So between the 2 orders, when they get going, it will be about 3,000 cars a year. We typically scrap somewhere around 2,500. So it's a small net addition. But where we really look to add is additional spot new car business and additional fleet acquisition business above and beyond that. And as Brian mentioned, certainly have been interested in pursuing that on the fleet acquisition side over the last couple of years, it's just been challenging at the valuations that we've seen the market. Historically, when we hit a down market, those have been great buying opportunities, but there's been so much capital seeking yield this time around that it's been harder to do that.

All right. And if I could go back to maintenance, maybe harping too much on this. But sounds like you're trying to get as many cars as you can through the maintenance network here in Q4. Do you have a sense of what the step up in cost might be from Q3 if things go your way and you get the cars that you're planning?

Kind of depends on how many cars we get. We are seeing more cars come in, and we'll definitely trend up in the fourth quarter. Unfortunately, it sounds like I'm avoiding the answer, but the answer is we really don't know how many we can get through in the fourth quarter and how many will spill over into next year, other than it's going to be a lot more than recent years.

The next question will come from Stephen O'Hara with Sidoti & Company.

I was just curious about -- just on the -- I mean, it looks like the -- sequentially, lease revenue in North America, I think it was the first uptick you've seen since, I think, 2015. And I'm just wondering, I mean, it seems like lease rates are pretty positive. Is it more of that? Or is it more kind of the -- maybe the addition of cars in the fleet? And -- I mean, is there a -- how do you think about fourth quarter on a sequential basis? I mean, assuming kind of a stable market from here, have we kind of hit the inflection point of maybe a declining lease revenue on that -- in for the North America?

Yes. So the actual numbers for lease revenue in North America were down for the quarter. But the thrust of your question, the lift was provided primarily by having more cars. The downward pressure was the lease rates, because despite the sequential improvement that Brian talked about, it's still lower than the expiring lease rates. So rates are still a net negative on the lease revenue.

Okay. And then just maybe on the maintenance piece again. When you say you're trying to get cars in, you're getting more cars in, in the fourth quarter, are these cars coming in because customers aren't as concerned with demand at the current time? Or is it just a slower time period? Or is it you kind of have to call them in because you've got to -- there's a contract where you guys have to get it done by a certain period of time, and you can't do that if you don't start doing that?

I mean, it's stenciled right in the side of the car. So if it's tank qualification, it's due this year. And generally, they don't like what happens when it spills into the next year. Because as I said earlier, the commercial team works -- obviously, is working right now to get them in. And eventually, that turns over to the legal department and eventually over to the AAR if it's late and into next year. So they want to avoid that as well, so that's why we're going to see a higher load in the fourth quarter. It's just that a lot of the cars coming in now, they're not going to get turned and sent out this year. So what I can say with certainty is it's going to be much higher in the first quarter because of what's spilling over. I just have not been able to -- we don't know yet how much higher it's going to be in the fourth quarter and how many are going to spill over. So they are required to get them in before the end of the year.

Okay. And then maybe just lastly on the last -- maybe not last call, but prior to that, you guys have been fairly vocal about level of backlog and things like that. And I know you guys have added to that somewhat. But are you comfortable with the level of backlog today, given -- I mean, obviously, lower would be better, but are you more comfortable today than you were maybe 6 months ago? I mean, it looks like it ticked up in 2Q. There's not a read for 3Q yet.

Yes. So one of the things we look at in addition to the numbers of cars in the backlog, it's the amount of time. And for most car types now, you're looking at something approaching a year, somewhere between 9 months and 12 months, to get a new car. And that is a large reason for the incremental increase in tank car lease rates that Brian talked about, is that, that alternative is a little harder to come by. To your direct question on the size of the backlog, it's still well above any other trough that we've seen. So we still think there is room for the backlog to come down in a full recovery scenario. But we are being helped by that length of time to get a new car increasing a little bit.

The next question will come from Bascome Majors with Susquehanna.

I know we're not into 2019 yet, but do you guys have a sense of how you might be incentivizing your sales force differently in 2019 versus '18?

Yes. So the sales incentive plan is something we look at every year and is tweaked a bit for where we are in the market. We're in the process of going through that right now. Certainly, as we hit inflection points, we become more interested in taking lease rates up and lengthening lease term. As I mentioned, for most of the car types in the fleet, we're not at that length in lease term point yet. We'll see when we get there.

But it's a great question, one we haven't had before. I mean, obviously, last few years, it's all -- been all about utilization and early renewals and things like that, trying to keep that feet utilized. So once -- what Tom says, once we start to incentivize rate and term, that's a really good sign. Not quite there yet, but hopefully get there next year.

All right. So getting there, but not quite there yet. I appreciate the color there, guys. And just one more real high level from me here. I mean, it feels like a lot's changed since your last earnings call, at least in the marketplace. I mean, your primary supplier has laid out a strategy that could have them investing billions of dollars in growing their own lease fleet. I mean, just this week, we heard your other key suppliers changing hands to PE ownership. Your largest competitor may or may not be up for sale in some capacity. With all of these changes, it feels like if certain things happen, this could be more significant than incremental to the industry. How do you guys think strategically about this market and sort of the competitive landscape perhaps being different 2 to 3 years down the road versus today? And what opportunities or risks does that create for your business model and strategy?

Yes. I mean, it's a good question, but honestly, it's more of the same. Trinity has been an aggressive lessor for a long time, prior to the split. So that's nothing new. And yes, it's changing -- ARI is changing hands, but you must remember that there was another lease fleet car to ARL that was heavily supplied by ARI. Icahn had an ownership interest in them both. So this is nothing new for us, competing against lessors who also supply us for manufacturing side. So there's really not a dramatic change there. What's more concerning to us is if somebody really changes in their behavior and gets extremely aggressive. And like I said, Trinity's been aggressive, an aggressive lessor, for a long time.

So you haven't seen the captive manufacturing finance models become more or less aggressive in the last few quarters versus the last couple of years?

No. I mean, as Brian pointed out, they've been aggressive, and it has been that way for a while. So can't really point to a more recent change in what we face out there as we try to place cars.

The next question will come from DeForest Hinman with Walthausen & Co.

A couple of questions. On the remarketing activity, any areas of strength that you would call out where demand has been strong?

No. Just across the board. I mean, most of our remarketing is in freight cars because there's a limited number of tank car buyers. But I'd say it's pretty healthy across the board for good customer credits, especially with longer lease terms.

Okay. And earlier, you mentioned generally scrapping around 2,500 cars a year. We're not really at that pace this year. Steel prices have been up. Should we expect a lot of scrapping activity in the fourth quarter? Or is this just going to be a lighter year on the scrapping side?

So the way scrapping decisions are made is it's not a targeted amount. We're not looking to scrap a certain amount of cars. What happens is a car comes into the shop and we look at what the projected cost to repair the car is, and what it's likely to earn over its remaining life versus the cash flows we get from scrapping from the car. So naturally, you tend to get more scrapping activity in a higher scrap environment. The 2,500 is more of a multiyear average. It's not something that we target year-to-year. As far as how many we've scrapped...

Yes. So...

We've actually scrapped less through the third quarter than we did last year, but at about a 20% average higher price. So the gains are higher this year by a few million, even though we scrapped a lower number of cars.

Okay. And this more of a high-level question. You've talked about third-party capital for a long period of time, and now we're starting to see rates moving up. Generally, that's kind of a good thing for us. But has there been any discussions internally or even at the board level about -- thoughts about when does that money go elsewhere and look at a different asset class? Have we had any thoughts or discussions on that?

Yes. Well, there's discussions all the time because we think that's been one of the things that driven oversupply, I should say, in the market, has been this third-party money without really a direct relationship with the shipper customer. As far as it's exiting, I mean, when you talk to them, one of the reasons they've entered the market is they talk about stable asset class and all that, but really, it's been because the yield -- as it's been described to me, the yield in the railcar market's been higher than the ones they've been able to achieve in their traditional financial -- especially in the banking -- banking investments of near 0 yield. So that's was driven them into the market. As far as them getting out, you would -- higher interest rates, underperformance of the fleet, because I do think a lot of these investments that were made were at too high a price and in not necessarily the best car types. So generally, what happens is as interest rates increase and fleets underperform, people start to exit. So really haven't seen a lot of exiting just yet, other than perhaps Element getting out of the majority of their lease fleet. But I don't know the reason that was, but I think they stated they just wanted to concentrate on their core business. So I think there's a number of reasons people get in and out, but as far as what we've heard, it's been more of the yield in our business is higher than the alternative that was traditionally available. So we'll see what drives them out over time. Traditionally, as I said, it's been when their feet underperforms or interest rates go up.

The next question will come from Justin Bergner with Gabelli & Company.

With respect to the tank car qualifications, are the periods that, that car is in the shop for service long enough, given the tightness in the market, to actually have a positive effect sort of on the lease rates that you're seeing sequentially in third quarter and into the fourth quarter?

Yes, it really isn't. So just -- first of all, what happens when the car is in the shop, it experiences rental abatement, so it's a draw on total revenue for us. But as far as the number of cars in the shop getting TQ at a given time, it's not enough to drive market lease rates.

Okay. Great. And then just a question on your overall comments about the market and the tightening nature of the market. I think you mentioned demand, railcar velocity and a tight truck market. Maybe if you could just sort of expound a little on each of those. And which of those factors is the most significant to the tightening in the leasing market that we're seeing now?

Yes. So I would say it's actually the truck and the railroad performance has been more impactful to date than the uptick in demand. Uptick in demand on the bed of already tightness due to railroad supply and the trucking shortage is certainly a help and helped accelerate things in the third quarter here. But more meaningful and impactful, I think, are those first 2. And in particular, those are what are driving the difficulty in getting the cars in for maintenance that Brian has talked about a few times.

Yes. And on the demand side, it's the Canadian crude that's really driven a lot of demand for tank cars. And as we've said repeatedly over the years, there's an ending out there for them, and I think it's when those pipelines come online just in the next couple of years. Like you have Enbridge Line 3, you got the Trans Mountain pipeline, you got Keystone XL out there. So when that happens, that's why we've been very careful investing in crude.

The next question will come from Willard Milby with Seaport Global.

Maybe this gets taken offline. I don't know if you might have this data in front of you. But working on some old data on industry-wide tank car recertification levels. And I think 2018, the value I had was about 49,000 cars, industry-wide, stepping down to about 40,000 in 2019 and half that in 2020. Does that sound accurate? Is 2018 the peak for tank car re-certs? And is 2019 just kind of a -- maybe a minor step down from those levels?

As far as what is due in a given year, 2019 and 2018 are pretty similar. It then trends up for us over -- from there. So it all depends on when a certain tank car owner purchased -- what they're purchasing pattern was 10 or 15 years ago. So for us, 2019 and 2018 are pretty similar and then it starts to take off after that. Scheduled.

Thank you for the question. There are no further questions at this time. I'd like to turn the conference back over to Jennifer McManus for closing remarks.

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 call. You may now disconnect your lines. Have a great day.