See the clock? I'll just make sure I keep track of time.
Yeah, I was going to say, we can't quite see.
We're going to go ahead and get started again, you know, sort of in the later afternoon, so, appreciate you joining us. I know it's always a challenge after lunch, but we're really excited to be joined today by GATX. With me, I have Tom Ellman, who's the EVP and CFO of the company. This is not a name that we cover, so I think it's a really good opportunity to kind of get into the business a little bit and the dynamics that you guys are seeing in the market, because we clearly are very exposed to some of the other markets that you guys serve. I think this will be a great opportunity to kind of chat and get to know you guys a little bit better. First off, Tom, thanks very much for joining us. Really appreciate it.
Thanks for having me.
All right, great. So let's kind of go through an orderly process through the business here. So let's start with your largest business, which is railcar leasing in North America. Maybe you could provide us an update on kind of the current environment, what you guys are seeing at. We'll talk a little bit about lease rates, kind of frame up where we are oÏn the cycle, if that's OK?
Sure. Just for background, for anybody who's less familiar, GATX has been around since 1898, 110,000 cars in North America. Railcar leasing in North America is how we started, still the core of the business. Railcar leasing is a pretty cyclical business. That cyclicality is primarily driven by the supply side of the business. If you look at the major demand statistic in railcar leasing, it's railcar loadings. How much loading of commodity has gone into railcars? If you look at the commodities that really make up the backbone of GATX's fleet, that railcar loading number has been pretty consistent all the way back to the early 2000s. You basically see a little bit of a dip during the Great Recession. You see a little bit of a dip during COVID. Otherwise, basically a straight line.
So if that's the case, why is the business so cyclical? It's really the cyclicality is driven by the supply side. So what happens is in good times, the industry produces too many cars. The replacement level of demand in railcar leasing is somewhere between 35,000-45,000 cars. Production capacity is somewhere between 60,000-80,000. So the industry has the ability to overproduce and has on multiple occasions. The reason I say we're in a pretty good part of the cycle right now is we've worked off all the excess supply from the last upcycle and have been in a period where supply and demand are basically in balance for pretty close to two years. So we've seen, to your point on lease rates, we've seen lease rates improve sequentially for about two years. GATX puts out a stat called the Lease Price Index.
What that statistic is, is it's comparing the new lease rate with the expiring lease rate. So customer renews a car, same customer, what's the new lease rate compared to the expiring one? And what's the average across our fleet? Right now, that number is at about a + 30%, which is about as good as it's ever been. Historically, about as bad as it's ever been is kind of - 20%. So you can see that the lease rates can move pretty substantially. And the reason I focus on lease rate is there is another potential source of variability in the business, and that would be the utilization of the cars. So how many cars are on lease? What kind of lease rate did they get? The reason I focused on that lease rate is GATX's fleet is almost always 98% or 99% utilized.
I could talk about why that is, but I'm afraid that Chris won't ask me any more questions because he has a really simple one and I give a really long answer. So I'll stop there and let you go to your next one.
No, that's great. It's good color. I want to pause for a second on that +30 relative to the -20. So I guess let's talk about that, maybe dissect it in terms of how much of that is relatively easy comps relative to overall strength. Is there a way to kind of break those two pieces apart when we think about the +30?
Yeah, so the short answer is no. But on a higher level, what I'll tell you is I mentioned that cyclicality. So one of the things that is absolutely true is we've been in this stronger market, I said, around two years. In the early part of that, you had a lot of very easy comps because virtually the entire fleet was coming off of a low lease rate environment. Average lease terms are somewhere between three and five years. So we're now starting to get a slightly higher percentage that's coming off of tougher comps. So the answer to your question is absolutely both. But it is not a case of simply the comps are easy because at this point, you're starting to get at least some of the fleet that's renewing off a little tougher environment.
OK, that makes a lot of sense. I appreciate that clarification. I guess when you think about the path for industry lease rates going forward, I guess is your expectation is we hold pretty steady at these levels here? I mean, I guess how does rail velocity and sort of service, which we're hopeful is on a more consistent, sustainable, improving path, how does that play in?
Yeah, so I'll take that in two parts. One, at the beginning of this year on our fourth quarter earnings call, we talked about our lease rate expectations for the year and noted that we had several quarters in a row of improving absolute lease rates. And our expectation coming into 2024 was that they might increase a little bit beyond that, but most of the strength for the year, and we guided, for anybody that doesn't know, somewhere between $7.30 and $7.70 EPS for the year. And that range counted on those strong lease rates maintaining and renewing a larger and larger percent of the fleet at those higher levels, not so much with an absolute increase beyond. And in the first quarter, we saw very marginal lease rate increases, somewhere between flat and up a couple percent.
To your point on what's going on with the railroad metrics, what we have great visibility to is the macro Class I metrics of velocity and dwell time. So measured on that basis, railroad performance absolutely is improving. The cars are moving faster and they're dwelling less often. What we don't have insight into and wish we did from a qualitative standpoint is what's going on with the first mile, last mile. Anecdotally, when we talk to shippers, they haven't really seen a material increase in railroad performance. If they did, that would long term, particularly with sustained, be a very good thing. Because in that scenario, anybody that's debating between truck and rail, if the railroad performance is better and expected to continue, it really encourages a shift. There is a possibility that in the very short term that it could be a negative.
If cars move faster, that might mean you need less cars. But generally speaking, as long as that happens in a measured way, which is the way railroad performance generally increases, you actually wouldn't expect that really any headwind, not so much because of modal shift, but because what is also happening on the other side is older cars are getting scrapped out, newer cars are getting built. What you would likely see, particularly in the strong scrapping environment, strong scrap prices, you'd probably see a little bit more scrapping and supply and demand would stay in balance. So that kind of wraps in a lot of the concepts you asked about.
Yeah, that's very helpful. I guess you touched on this a little bit, but maybe we can expand on it in terms of rails and share capture. We do see more sustained improvement in service. We think there is a big opportunity for the rails to take some share here as that service is sustained and they can kind of convince shippers that this is a more viable option for them going forward. That, I would imagine, would generate a little bit more interest and potentially more demand for cars. Is there a way to translate any of this? I mean, do you guys use any sort of rough metrics or rules of thumb about how to think about that?
Yeah, so at the highest level, there's an old rule of thumb that says one mile per hour change in lease rate, I'm sorry, change in average velocity translates to about 50,000 cars. It's nice, but it's not actually all that helpful because of that first mile, last mile that I talked about. So if that first mile, last mile improved to the same degree, that metric would kind of work. But given that shippers are not seeing that, you really don't have it. If you did, what you would see, I mentioned that railcar loading line that never changed, if you did see that, you would see that increase.
Yeah, OK. That makes sense. I guess in terms of customer inbounds, what are you hearing from customers? Is there a degree of increased demand that you're feeling at this point?
So there's an oversimplification. GATX has 160 different car types. So anything I talk about, when you're really managing it, you're managing it on a car type by car type basis. But overall, if a customer wants to get a new car right now, they generally speaking would have to wait about a year to get it. So if they truly need the cars they have, the cars that are coming up for expiration, it's really not all that viable to wait for the new car. And because of that, that's what's leading to the strong lease rate environment I talked about, that +30 on the Lease Price Index. However, most customers also are not in a position where they're really looking to expand, really looking to add to their fleet.
So they definitely want to hang on to the cars they have, which is leading to this strong lease rate environment on the existing car. But where they have an opportunity to incrementally increase the size of their fleet, that's the kind of business builders are hungry for. And so they will compete heavily to get that. So you have this strange dichotomy where you have a really strong existing car lease rate environment and kind of an OK or tepid new car lease rate environment.
OK. And I guess maybe moving around a little bit, I think you guys have talked about your maintenance expense expectations in North America. I think flat to slightly higher is the guidance as it stands right now. So feel good about that? And what could some of the puts and takes be around that?
Yeah, so GATX does about 80% of our maintenance on our fleet in our owned facilities. That is quite advantageous for a number of reasons. One, we can do it less expensively in our owned facilities than going out to a third party network. But probably even more importantly is the commercial aspect of that. If a shipper is giving up a car for maintenance, what they care about is how quickly do they get it back and how certain are they about when they're going to get it back. And if they give it up to a GATX facility, they're probably going to get it back in about half the time than if they have to go to a third party network. And they're also going to get it back when we say they'll get it back. Those things, I talked about how utilization hasn't really changed for us.
It's been 98%, 99% for years. That's the key reason. We absolutely have to price to market. But provided we price to market, that service advantage means shippers are interested in taking a GATX car first and giving back a GATX car last, which is what allows that utilization to stay high. On the maintenance cost, I mentioned the cost advantage, our own shops versus third party shops. One of the things we've been doing is driving more and more work into our owned facilities. So maintenance costs in general, components, inflation, labor costs, those things are going up.
The reason that our maintenance is only expected to go up a little bit or possibly stay flat is offsetting that is bringing more work into our owned facilities, increasing the amount of production that we can do in our owned facilities so we have to send less to that third-party network.
OK. So let's talk a little bit about the secondary market for cars here. Just get a sense of how that feels in North America right now. Are there attractive opportunities? Is it the buy side? Is it the sell side? How do you think about it?
Yeah, so actually, strangely, it's both on both the buy and the sell side. So starting with the sell side, when GATX makes a new investment, what we assume is we're going to own that car from the beginning of its life to the end of its life. Everything's on a discounted cash flow basis. So we've been in the business since 1898. We have a really good sense of what does it cost to maintain that car over its life, what is it likely to earn once it comes off its initial lease. We obviously know the new car cost and we know the initial lease rate. So you do a discounted cash flow model, and if it pans out, you buy the car. If it doesn't, you don't. Even though that's how we analyze it, we don't actually hold every car its entire life.
We hold a lot of them its entire life, but not every car. The reasons we'll sell is primarily for fleet optimization purposes. One of the things that we really pride ourselves on is having the most diverse fleet in the industry. I mentioned 160 different car types. We're not overexposed to anyone commodity, anyone customer, anyone car type. So we will regularly sell into the secondary market to maintain that optimal fleet mix. And even though we're selling primarily for fleet optimization purposes, we're always doing that. So you can always kind of count on the fact that we're going to do some asset sales. Over the last 10 years or so, we've averaged about $65 million a year of gains on sales of assets. That number has varied. Low has been about $40 million. The high has been about $110 million, which was last year.
We noted coming into the year that we expected this year to be down about $10 million from last year. So lower than last year, but high on any kind of historical basis. And the reason for that is when we put out portfolios of assets, the response has been very, very strong. First quarter, we definitely are on track to meet that $100 million guidance that we provided at the beginning of the year.
I guess maybe let's say expanding beyond that, as you think out over the next year or two, are there any factors that you'd point to as saying, hey, this gives us more or less confidence in what sort of our ability to generate those gains might look like?
Yeah, no, we feel very good about it over the long term. One of the things to note is the assets that we sell are almost always with a lease attached. And there's a large portion, mostly of smaller players, of the lessors that don't have their own origination capability or don't have much of an origination capability. So their business model calls for purchasing assets in the secondary market. So as long as there's a lease attached with some term, that's what makes you feel good about being able to put that car out there. I mentioned that sort of that range that we've had, that $40 million number was the first year of COVID. So kind of wherever we are in the cycle, you can count on that demand being strongly there.
The main reason for the variability is actually where we are in terms of what we'd like to sell from an optimization purpose. Most of that year-to-year variance is based on what we decide to sell, not because the secondary market is particularly stronger or weaker.
As we wrap up on kind of North America before we move geographies, I guess as you think about the fleet size over time, is there any numbers or metrics we should be thinking about in terms of what you want to do over a multi-year basis with a North American fleet?
Yeah, so really importantly, we intentionally do not target that. So we take a very long-term focus. And what we're interested in is buying opportunistically. We buy new cars three different ways. We buy new cars through long-term supply agreements. So what we will do is work with a builder and commit to buying a certain number of cars over a certain number of years. Right now, we have a supply agreement in place that calls for us to purchase 3,000 cars a year over the next five years. We scrap about 2,500 cars a year. So that is largely a replacement program. The other two ways we buy are much more opportunistic. We will also buy new cars in the spot market, which is that discounted cash flow analysis I talked about.
Then we will be a buyer of existing cars in the secondary market, same kinds of things we're doing on the sell side. Obviously, if we're both a buyer and a seller, we're not, generally speaking, buying and selling the same car type. As a general rule, GATX will tend to focus as a seller on more vanilla straightforward car types and tend to focus more as a buyer on more niche service-intensive car types because of that maintenance capability that I mentioned.
OK, that's very helpful. So let's kind of move around the world a little bit and talk a bit about Europe and how that leasing business is doing kind of given the soft macro backdrop, although I guess the last couple of data points maybe have been marginally better there. So let's talk a little bit about Europe.
Yeah, so the European market has been amazingly resilient. Even during COVID, lease rates in Europe went up. They didn't go up very much, but they went up. And there's two big drivers to that. One is that I mentioned in North America this boom-bust cycle and the production capacity being well in excess of replacement demand. In Europe, those two numbers are virtually identical. It's around 15,000 cars that they can produce in Europe and around 15,000 cars that get scrapped every year. So you don't have that boom-bust. The other thing is the real aggressive efforts to take traffic off the roads and put it on the rails. That really is a source of constant demand in that area. So much like in North America, I talked about the supply-driven cycle.
The supply-driven situation combined with that modal demand shift really has made Europe quite resilient and a good lease rate story even when some of the macro indicators aren't as strong as you might like.
Any material difference in the profitability between Europe and North America?
In general, Europe for that reason would be a little bit more profitable than North America because you don't have that same volatility.
Got it. Let's talk about India now and sort of how you think about that. That seems like an interesting growth market. So can you talk a little bit about what the prospects are there for you? And I mean, I think it's, yeah, so let's talk about that, about what you think the growth opportunity is there.
Yep, so GATX was the first private leasing company in India. In 2012, we worked with the Indian government and the Indian Railways to develop what they call the scheme for private leasing of rail cars. Prior to that time, any cars that weren't owned had to be provided by the Indian Railways. The Indian Railways, looking to, much like in Europe, but even more so, provide an avenue to take traffic off the roads and put it on the rails, was looking for someone else to provide capital and expertise. On the expertise side, they really liked the fact that GATX engineers and GATX inspectors would be looking at the new cars that local manufacturers were producing. In every part of the world, manufacturing happens very close to where the cars are used. Nobody has figured out a way to economically move rail cars over the ocean.
The Indian Railways and the government knew that if inspectors were ensuring GATX got a good car, the Indian Railways would also get a good car. They would also benefit from that. We worked closely, initiated that. It took us about five years to get to 1,000 cars. Since that time, growth has really accelerated. We should pass the 10,000 car mark this year. Demand, both from more and more car types becoming eligible for private leasing and the benefits of the Indian government investing billions in what they call a Dedicated Freight Corridor. In India, most rail lines give preference to passenger traffic. That is an inhibitor to freight growth. There are now certain lines that are only for freight cars. As more and more of those come online, that will be an even greater source of demand.
So we expect to continue to be able to add 1,000-2,000 cars a year in India for the foreseeable future because of those growth prospects.
I guess similar to the European question, how do margins there and maybe returns on that growth compare to the rest of the business?
Yeah, so as you would expect, this new market, the way you generate return is from the initial lease and then what do you get in the residual period? What do you get beyond that initial lease? Obviously, in India, there's not the experience. So certainly, on an absolute basis, the highest returns are in India, followed by Europe, followed by North America. What we will see as time goes on is what are those on a risk-adjusted basis. But certainly, the absolute returns are highest there.
OK, that makes sense. Let's switch gears to the engine leasing business there. So this is an interesting business, in my opinion. I guess number one, how do you think about how it fits into the portfolio of mostly rail assets?
Yeah, so it might not seem like it, but there's actually quite a bit of similarity between a rail car and an aircraft engine. A little bit of a history lesson. GATX used to be an aircraft lessor as well. We got out of that part of the business because that was not a great fit for us. That is, can you buy in bulk? We couldn't. Do you have the lowest cost capital? We don't. But the engines are different, just like rail cars, long-delivered, essential use assets with a heavy service component where you're really competing on that service and your asset knowledge. So when GATX decided to get out of the aircraft leasing business, we hung on to the engine business. At the time, that business was exclusively a joint venture between GATX and Rolls-Royce, 50/50 JV. Ad there's about 400 engines in that joint venture.
About half those engines are leased to airlines. So when an airline needs to take an engine off-wing for maintenance, they put a spare on-wing, they do whatever maintenance they need to do, they continue to have use of the aircraft. The other half of those engines are unleased back to Rolls-Royce, the parent. And Rolls-Royce uses those in their TotalCare maintenance program. So Rolls' deal with a variety of airlines is they get paid a fixed amount, Rolls takes care of the maintenance and provides a spare engine while that maintenance is being done. And they source a lot of those from the joint venture. So the reason the business fit is because of those characteristics. The business has been very successful because in addition to the growth that has occurred prior to COVID, air passenger miles doubled roughly every 15 years.
The percent of engines that airlines leased went from about 10% when we started the JV in 1998 to about 50% today. Growing market, growing percent of leasing, really good for us. COVID really provided the greatest test of resiliency you could get. Air passenger miles went to zero. The JV prior to COVID earned about—our share of it was about $90 million. Even during COVID, it only got down to about $45 million. When you have no traffic and you're still making money, that's a great test of resiliency. It's really because if an airline loses use of an aircraft, that's unpleasant. If they lose the spare engines, it can really shut down the whole airline because you just can't service it. Our hypothesis was that that would be the absolute last thing an airline gave up.
COVID provided a test and showed indeed the case. Going forward, we certainly we've recovered from COVID. Air passenger miles are back. We expect to resume the growth trajectory that we've seen in that business. I mentioned that when we started, it was all 100% joint venture with Rolls-Royce. Today, we have that, but we also own 29 engines for our own account, about $700 million of NBV. Those engines basically work commercially exactly the same as the engines in the JV. The partnership manages those engines for GATX. So commercially, it's indistinguishable from the JV engines. We said on our last call, we expected to do about $250 million-$300 million a year in investment in those engines for our own account. The reason we started doing that is, again, during COVID, that was a time where Rolls had other uses for their capital.
GATX is a countercyclical investor. Just like in rail cars, in aircraft engines, that was a good time to invest. A lot of other people weren't doing that. Upon doing that, it was really good for both partners. It's good for GATX, great source of investment, good for Rolls because they could focus on other things and expect to continue to do that. All 29 of those engines are Rolls-Royce engines.
Where does growth capital go? Does it go to the JV or does it go to your own portfolio for the first dollar?
So the good news is we haven't had to make that decision.
OK.
Both. We'll invest through the JV and on our own.
OK. The return profiles of those two sort of side by side? I don't know if it's comparable, but obviously, they're kind of getting pulled together to some degree anyway. How do you think is there a difference in the return profile?
They're very similar and strong.
Got it. OK, that's helpful. Let's talk a little bit about CapEx here. I just want to get a sense on some of the numbers here. How do you think about kind of overall CapEx for this year and then maybe what are the opportunities beyond this year?
Yep, so first, our capital allocation framework is, first and foremost, we want to invest in economically creative assets. Kind of right with that, very important that we have consistent access to cost-advantaged capital. So we work very closely with the rating agencies. GATX is rated BBB+ at Fitch, BBB at S&P, and Baa2 at Moody's. So a notch higher at Fitch than the other two. Moody's has GATX on positive outlook for upgrade. So hopefully, that'll happen and we'll be to the other way. But those kind of I said one and two, but it's almost one and one A. And then we will return excess capital to shareholders. GATX has paid a dividend uninterrupted since 1919. That's got to be one of the longest streaks of anybody. And then we will periodically also do share buyback.
To your question on capital, last year, we did about $1.6 billion of investment. We announced we expected to do a similar amount this year. The last four years, we've been over $1 billion. But I started at GATX in 1993. Most of my career, we've been kind of in a $600 million-$800 million range. So that high level of investment opportunity has really, we haven't done material buyback in several years now. And it's because of the investment dollars that we've been able to put to work. I mentioned that we expected 2024 to be similar to 2023. And that's really true across each of our businesses. And we talked about the growth opportunities in India. We talked about what's going on with aircraft engines, sort of steady investment opportunities in Europe.
And then the combination of our committed supply agreement and the attractive environment for buying existing cars with assets attached, I mean, existing rail cars with leases attached, we expect that to be similar as well.
OK. I forgot to mention, if there are questions, feel free to raise your hand. We got a couple of minutes left. I do want to hit on one thing. If you do have questions, we'll get you in. Let's talk about the earnings numbers, right? Let's just dial this down to what you guys expect here. I think $7.30-$7.70 is the range that you guys have affirmed for the year. I guess, what gets you off track there? Obviously, there's a lot of stability in the business just given the nature of the term structures and the portfolio that you're managing. What could be the driver that gets you sort of above or below that range?
Yeah, so by far, the biggest source of up or down is going to be what happens in the secondary market. That's both in rail North America and with the aircraft engine portfolio. We have a great line of sight into what kind of opportunities are out there. It's very difficult to precisely call the timing. It can go either direction. More sales can close than you thought or less. When we sell those rail cars, we're selling 10, 20, 30 individual packages of leases. It's not one big asset sale to one player, but you don't know exactly when those will close. Engines, by their nature, are very lumpy. A new XWB costs over $30 million. So when you're selling an aircraft engine, that's also pretty lumpy. So that would be number one. Number two would really be the timing of maintenance expense.
Same kind of thing, pretty easy to call what's going to ultimately happen. But do you get more cars in a certain year or do you get less cars in a certain year? That's a little tougher. What you might expect is differences in lease rates. That ultimately is incredibly important, but particularly as you move through the year is a lot lower on the list because North America, over 100,000 railcars, about 20,000 come up for expiration in a given year, so chance to reprice. And then obviously, as you move through the year, you have less months remaining. So we watch that really closely from an operating statistic, which is why we do things like put out the LPI. Ultimately, that'll be very important. But for 2024 itself, a little less so.
OK, got it. So there's not, as it stands right now, nothing you guys have highlighted that you would see concern around the range one way or another.
Correct.
OK, got it. That's helpful. Well, I think that actually takes us up to the end of our time. So I appreciate you joining us here. It's very helpful and great to have you. Thanks very much.
Thank you. Appreciate it.
All right. Thanks, everyone.