CSX Corp (CSX) 2020 Q2 法說會逐字稿

Good afternoon, ladies and gentlemen, and welcome to the CSX Corporation Q2 2020 Earnings Call. As a reminder, today's call is being recorded. (Operator Instructions)

For opening remarks and introduction, I would like to turn the call over to Mr. Bill Slater, Chief Investor Relations Officer for CSX Corporation.

Thank you, and good afternoon, everyone. Joining me on today's call are Jim Foote, President and Chief Executive Officer; Mark Wallace, Executive Vice President of Sales and Marketing; Kevin Boone, Chief Financial Officer; and Jamie Boychuk, Executive Vice President of Operations.

On Slide 2 is our forward-looking disclosure, followed by our non-GAAP disclosure on Slide 3.

With that, it is my pleasure to introduce President and Chief Executive Officer, Jim Foote.

Thanks, Bill, and thank you to everyone for joining our call this afternoon. Wow, where do I start talking about this quarter. This was the most disruptive quarter I have experienced in my career, with both the fastest decline in volumes followed by one of the most rapid increase in volumes in the company's history. Reacting to those extreme swings while dealing with the pandemic has been and continues to be challenging. CSX employees have been absolutely amazing in their response, and I want to thank them for the hard work and dedication to keep the railroad running well. And I would be remiss if I didn't extend special recognition to our employees and their families who have fallen ill from the virus. Our thoughts are with them.

In difficult times, companies -- strong companies adapt, and that's exactly what we have done. Despite all that has been accomplished to date in CSX's transformation, we are still finding opportunities to be better, faster and more reliable. The actions taken over the past months will not only make CSX more efficient, but more importantly, they will allow us to be better delivering service to our customers in the future.

Turning to the presentation, on Slide 5 is an overview of our financial results, which Kevin will discuss later. Due to the economic impact of the pandemic, second quarter EPS declined 40% to $0.65 while the operating ratio increased to 63.3%.

On Slide 6 you could see that all business lines were negatively affected by the pandemic. Second quarter revenue declined by 26% year-over-year on 20% lower volumes. Sequentially, the volume decline was the largest in CSX's history, and almost twice as severe as any quarter during the 2009 recession.

Merchandise revenue and volumes declined 22%, with the largest headwind coming from the automotive plant shutdowns. Automotive volumes declined 71% in the quarter, including 6 consecutive weeks where volumes were down more than 90% each week. Intermodal revenue declined 18% on 11% lower volumes, as both the international and domestic businesses were impacted by lower consumer and industrial demand.

Coal revenue decreased 48% on 44% lower volumes. Both the domestic and export markets were negatively impacted by weak demand from the combination of a reduced power consumption, low natural gas prices and export benchmark prices. Other revenue declined 19% due to lower affiliate revenue and declines in demurrage charges and intermodal storage revenues.

Turning to Slide 7. Safety remains my top priority. Even though we had an all-time low number of train accidents in the quarter, the frequency rates for both train accidents and personal injuries increased due to the lower level of volumes. While CSX continues to lead the industry year-to-date in safety, we must do better. We will never be satisfied with our performance if any of our employees get injured while at work, and we are undertaking a comprehensive safety engagement initiative this quarter, focused on strengthening critical rules compliance and systemically identifying and eliminating unsafe acts.

Moving to Slide 8. Let's review our operating performance for the quarter. Despite the challenging operating environment, the railroad continued to run at a high level. The operating team successfully implemented significant plan changes while maintaining fluidity and driving efficiencies across the railroad. In addition to these service design changes, we also drove increased yard productivity, setting a new record for cars handled per hour worked. Additionally, CSX continues to lead the U.S. Class I railroads in fuel efficiency, setting a new quarterly operating record of 0.96 gallons of fuel per 1,000 gross ton miles.

Fuel efficiency, along with ESG more broadly, are priorities for our team. We are focused on leveraging the inherent benefits railroads have as the most sustainable mode of land-based transportation, and are working hard to make sure CSX is building on these benefits to continue leading the industry in safety and sustainability.

The topics of ESG and driving positive social impact are critically important to CSX, our customers and the communities we serve, and we look forward to providing additional detail on key initiatives in our sustainability report to be released later this month.

Slide 9 clearly highlights the unique operating challenges faced during the quarter, as volumes dropped more than 25% and then bounced back nearly 20% since Memorial Day. The team acted decisively to adjust the network for the changing volumes, reducing total train starts roughly in line with volume declines and realizing line of road efficiencies even greater than the volume declines. These results reflect the strength of our operating model and the ingenuity of our team. In addition to consolidating trains across the network, they found new opportunities to integrate auto and unit train business into the merchandise and intermodal networks as well as other unique service changes to be -- to more efficiently leverage yard and local operations.

When volumes began to recover late in the quarter, we started recalling employees and added train starts to meet demand, but we are confident that many of the changes we made during this period are durable and make CSX a stronger company.

Slide 10 shows our hourly trip plan performance in the quarter. Carload trip plan performance of 80.5% was consistent with first quarter results, and intermodal performance remained high at 94%. Volume has been volatile, but we are working closely with our customers to ensure they are appropriately resourced and prepared to handle incremental volume when it comes.

I'll now turn it over to Kevin for more detail on the quarter.

Thank you, Jim, and good afternoon, everyone. The second quarter of 2020 represented the most significant revenue drop in the company's history. Faced with an $800 million revenue decline, our company clearly responded. From a cost perspective, our focus through this period was to accelerate initiatives that will deliver sustainable structural cost reductions that position us to leverage growth as the economy recovers. As a leadership team, we were deliberate in our strategy, focused on avoiding the pitfalls on making short-term cost decisions to the detriment of growing the business profitably, as the economy rebounds from the pandemic.

As you can see on Slide 12, total expense was down 19% in the face of a 20% decline in volume. This includes the impact of approximately $60 million of unique headwinds we faced in the quarter, which included lower real estate gains, severance costs related to a management restructuring, specific COVID-19-related costs, higher depreciation and property tax as well as equipment impairments in the quarter. Excluding these items, costs would have been down 22% year-over-year. I frequently get asked the question, especially over the last 3 months, how much of your cost base is fixed versus variable. And naturally, everyone wants to hear how variable it is when volumes are declining, and how fixed it is when volumes are on the rise. The reality is, very little of our expenses falls out with lower volume. Action and a lot of hard work is required. It's a balance to ensure we variabilize the cost structure while maintaining service levels.

In the second quarter, we successfully identified creative solutions to help drive efficiency on a long-term basis. Some of these efforts will show up near term, while others will benefit us in the quarters ahead. As volumes improve off the bottom, I am confident you will see very strong operating leverage, driven in part by a lot of the hard work and effort the team has put in over the last several months.

Walking down the expense line items. Labor and fringe was 22%, lower reflecting the benefit of reduced crew starts, lower operating support costs and reduced labor at our terminals as well as dramatically lower overtime spending. Crew starts were down 24%, exceeding volume declines in the quarter and saving over $70 million in total T&E labor expense. Additionally, the active locomotive count was down 25% year-over-year in the quarter. The smaller fleet, combined with fewer cars online and freight car repair efficiencies, helped drive a 20% reduction in the mechanical workforce, while at the same time reducing mechanical overtime expense by 56%.

We continue to remain extremely focused on overtime. Through workforce efficiency and management execution, we reduced overtime across all operating departments by 50% versus the prior year, driving $20 million of overtime savings in the second quarter alone. We also achieved significant, largely sustainable reductions in our engineering contract labor expense in our intermodal terminal workforce. In addition to these efficiency improvements, we had $39 million of lower incentive compensation expense in the quarter, primarily reflecting lower projected payouts on existing plans. Finally, the quarter included $10 million in severance costs resulting from a management restructuring to align our resources and improve efficiency. From these actions, we expect $25 million in ongoing annualized savings.

As I've described before, MS&O expense tends to be less volume-variable than labor. But we still reduced this line item by 9% in the second quarter, or 14% when you adjust for the impact of $26 million in lower real estate and line sale gains versus last year. Continued improvement to the train plan, combined with increased network fluidity, enabled a 37% reduction in crew travel and repositioning expenses versus the prior year.

On the mechanical side, the active locomotive count was down 25% year-over-year, and we closed the quarter with over 1,000 locomotives in storage. In addition, fewer cars online and freight car repair efficiencies helped drive a 20% reduction in car material expense in the quarter. The safety of our employees remains CSX's #1 priority. In the quarter, we spent approximately $10 million in COVID-related supplies to ensure our employees were safe and following CDC guidelines while they kept the railroad running. We leveraged our purchasing power to procure large quantities of safety supplies to stage around the network. Going forward, we expect a run rate of $1 million to $2 million per quarter for the balance of the pandemic.

I already mentioned the decline in the real estate gains in the second quarter. For modeling purposes, our current assumption is for minimal sales activity across the balance of the year. We continued to see sales opportunities longer term while the team is also focused on accelerating strategies that will increase recurring income streams that are tied to our real estate portfolio. Fuel expense improved $143 million or 61% year-over-year, driven by a 50% decrease in the per gallon price, lower volume and significant efficiency improvements. I probably sound like a broken record, but we achieved another all-time record of quarterly fuel efficiency at 0.96 gallons per GTM.

Looking at other expenses. Depreciation increased $7 million or 2% in the quarter, mainly driven by our 2019 equipment study. Equipment rents expense decreased $14 million or 15%. Lower volumes as well as improved cycle times in our intermodal market was the most significant driver.

Turning below the line. Interest expense increased primarily due to higher debt balances, partially offset by a lower weighted average coupon. Income tax expense decreased $123 million or 45%, primarily resulting from lower pretax income. Absent unique items, we continue to expect an effective tax rate of approximately 24.5% for future quarters.

Closing out the P&L, CSX's operating income declined 37% year-over-year with a 63.3% operating ratio.

Turning to the cash side of the equation on Slide 13. On a year-to-date basis, capital investment is roughly flat. We continue to invest in our core track, bridge and signal infrastructure, prioritizing investments that provide safe and reliable train operations. We are not cutting capital, especially investments in the safety of our core infrastructure and projects that produce attractive returns.

Through the second quarter, free cash flow before dividends was $1.4 billion, down 15% versus prior year. Free cash flow has continued to be a key focus for this team. Even in this challenging environment, we saw free cash flow conversion on net income exceed 100%. Our results this quarter really highlight the significant improvement in our operating model. Our ability to generate positive through-cycle cash flow allows us to continue to invest in our network and pursue incremental investments that will further strengthen our business. Importantly, our cash and short-term investment balance at the end of the second quarter was $2.6 billion. This industry-leading cash and liquidity position provides us ample flexibility going forward, and we remain committed to distributing excess cash to shareholders.

With that, let me turn it back to Jim for his closing remarks.

Thanks a lot, Kevin. Concluding on Slide 15. Boy, are we happy to see the volumes recover from the May trough as the economy strengthens. However, while these trends are encouraging, the ultimate path of the recovery remains too wide to accurately predict at this point. As for CapEx, we are still expecting the full year to be at the low end of our initial $1.6 billion to $1.7 billion range. We will never reduce or defer any spend that impacts safety, and we will continue investing to maintain the integrity of our network. We are, however, evaluating our capital spending programs to maximize efficiency and prioritize high-return projects. We're applying the same disciplined approach to our capital budget that we do to our operating expenses, and making sure every dollar is spent productively while actively seeking out and eliminating wasteful spending. It's this mentality that has generated CSX's industry-leading cash flow profile and strong liquidity position. Not only does our transformed cash flow profile provide additional operating flexibility during these periods of uncertainty, but it allows us to continue investing for the health and growth of the business throughout the cycle. CSX is a better company today due to the actions taken over the past few months. No matter what path the recovery takes from here, we will remain laser-focused on serving our customers and making this company stronger than it has ever been. I'm proud to work with such an exceptional team of railroaders, and I'm more confident than ever that the best is yet to come for CSX.

Thank you, and I'll kick it back to Bill.

Thank you, Jim. In the interest of time, I would ask everyone to please limit themselves to only 1 question. With that, we will now open the line for questions.

(Operator Instructions) Our first question comes from Tom Wadewitz from UBS.

Yes. Wanted to see if you could offer some more thoughts on the structural changes to cost and how we might think of those. I guess the simple framework I think of is just kind of the schedule, and what would your train schedule look like if the volumes came back to pre-COVID levels. I don't know if that's the right framework or if there are other things. But just wanted to see if you could give more of a framework of the structural cost takeout and how that looks when volume comes back.

Sure, Tom. I think, probably, Jamie is probably the best that can give us a little bit of the color behind everything he's been working on so hard for the last 3 months.

And no, absolutely. Thank you for the question, Tom. As Jim mentioned and both Kevin, we've seen a bit of a rebound, obviously, on our volume levels here over the past 6 weeks. We have -- being up 25% from our low point during the COVID period here. And if you take a good look at the numbers on Slide 9, you'll see that this business has come back with us only using 13% more locomotives and 14% more road starts. So we have -- we use this time during COVID almost like our practice hour to be prepared on how to run trains differently and how to readjust our network. We're using more distributed power. As Jim had mentioned in his opening, we are doing more of a train mix. Our auto network, which practically disappeared throughout the COVID period, has come back strong, and we are moving it in a different way than we ever have. We're mixing the auto network with our manifest and, in some areas, with our intermodal network. And reducing those train starts, road starts that we see out there is going to be a good lasting effect as we move forward.

Are we going to have to continue to add some assets? Absolutely. If and when volume continues to improve and go above the COVID volume -- pre-COVID volume, definitely we'll be putting some assets back in and bring some more people off furlough. But I've always said that every asset we bring back in, will earn its keep. And we're running the network very well right now. We're starting to get our mojo back, I guess, with respect to our mainline speeds, where we hit 21-mile an hour here today, which is a number that we're happy with, and we want to get that fluidity moving. And also in our yards, we've shut down a number of mechanical facilities that won't come back. We have made sure that we've reduced the proper shifts in some of our diesel shops as well as moved work, some of our queue-type work, which is some of our heavier semi-yearly work, to certain facilities, it's working really well. We'll continue to do that, which will continue to help us drive those savings.

And then particularly, on some of the areas of engineering. So if we take a look at not just this being the costs on the transportation side, on the engineering side, we have been very successful in making sure that we get good work blocks out there. Our engineering team, even though we're putting in still the same amount of ties and rail as we did last year, we're doing it with $17,000 less per mile. We're doing it over $3 less per tie installed. So these are items that we're going to continue to move forward, even as the volume continues to get back to pre-COVID levels. And we feel quite confident that we're going to be able to absorb some of the costs or some of the volumes as it continues to come back.

Your next question comes from Ken Hoexter from Bank of America Merrill Lynch.

Jim, can you maybe talk about, given the speed of the volume rebound that you're seeing and carloads now down only mid-single digits, thoughts on employees, maybe a little bit more details on what you did with the restructuring there? Did you eliminate layers? Were there shift in operations and maybe the pace of return of some of those employees? And then, I guess, in a bigger picture, your, just thoughts on returning to possible operating ratio improvement as we get into the back half of the year.

That's a nice 3 question in 1 question. I like your style. Yes. Again, in terms of the structural changes that we made during the last couple of months here, I tasked the team with taking a look at everything that we do, and seeing if there is a better way to do it. Wasn't just an exercise in mandatory headcount reductions, but it was an opportunity to see if during this more difficult period of time, we could assess almost everything we did in the organization and figure out how we could eliminate layers and get things done more effectively. And as a result of that, we basically aligned duties and responsibilities in the areas where they probably should have been a long time ago, and the organization is stronger and better-performing as a result of that. Jamie can offer -- jump in a little bit here more on calling back people, more people in the field as we more forward.

Our #1 priority here is to make sure we deliver an excellent product to our customers. It would be -- when we saw that -- the dramatic yo-yo effect of our volumes over that 6-week period there, where everything just dropped off and then bounced back, that's not an easy way to run a railroad. And we've got all kinds of work rules, et cetera, associated with how long it takes people to come back to work. And so we're just now, as Jamie said, getting the railroad back up running with great fluidity and velocity. And now we will focus on getting back to what this team is now known for, being exceptional operators.

And as we move forward in terms of the third part of your 1-part question, there's -- the operating ratio is -- it's an equation. And we need to get the revenue stream back to where we expected it to be. And we think based upon sequential increase week after week after week in the carload volumes, which you see just like we do, things are beginning -- things are grinding their way forward. And we hope that, that continues. And if that continues, then we'll get back to where we think we should be, which is showing everybody that we can run a railroad as efficiently and as lean as anyone else in North America.

Jamie, do you want to just -- anything else I missed?

Yes. Just -- look, when it comes to the headcount and make sure we have enough people, the most important part of that recipe really is working with Mark and his team, ensuring that any customers who we'd get a head up -- heads up on, who are starting to pick up on some of the traffic flow, that we've got people where we need them, and we start positioning people ahead of time as much as we can. We've brought in hundreds of T&E employees back. They're out there working for us on the front line, moving the freight and making sure that our customers get the service that they expect out of CSX. But of course, it's not easy. As Jim mentioned, in 6 weeks to go from your all-time low during this COVID period to a 25% volume increase and be able to pick it up, is everything we've said that we were going to be able to do at CSX, and we've been able to accomplish that. So as more business at some point comes along, again, we work really, really close with our marketing team. And we've got a good number of employees still on furlough that we want to find work for them. And we'll continue to bring them back as we see the business level start to come.

Your next question comes from Brandon Oglenski from Barclays.

I guess, Jim or Mark, can you give us some color on where you see the recovery coming right now? I think from our outside perspective, it seems like a lot of consumer-oriented freight markets are maybe rebounding a bit faster. Could you talk about that and maybe the dichotomy of what we're seeing on more core industrial trends? Or maybe that's not the right way to describe it right now.

Sure. So it's Mark. So thanks for the question, Brandon. So interestingly, over the course of the last few months as our employees have been working from home and not spending time in airports and hotels and out there meeting with their accounts, we made it a key priority for our sales and marketing teams to remain close to our customers in order to sort of understand what they're seeing in the marketplace and we could continue to service their needs. And I think the general consensus throughout the quarter from customers was -- has been that the outlook remains unclear. Having said that, I think, as Jim said earlier, we and our customers are -- certainly have been pleased and encouraged to see the volumes recover off the trough from May. And I think -- but I still think it's too early to say that we have an accurate view on the trajectory of the volumes or the shape of the recovery in the second half.

Now getting to your question, I think in merchandise, obviously encouraged to see auto plants reopen, which has helped other markets like steel and plastics and auto parts, for example. Interestingly, even our ag and food business during the quarter, which traditionally is a pretty stable business in a recession, was impacted given the declines in ethanol, as nobody was driving into the office every day. Our beverage business was impacted given some of the shutdowns in Mexico, and those supply chains were impacted. And our feed grain shipments were low, given also some food processors were running low capacity levels. All these markets now are starting to see some good volumes across the merchandise portfolio, and we'll obviously continue to watch them very, very closely.

I'll just also touch a little bit on intermodal. In April and May, they were obviously tough months for us, both for domestic and international, but we started to see volumes rebound nicely in June, especially on the domestic side as the economy began to reopen and we saw inventories being replenished at some of the retailers. And we also saw strong volume surges for our e-commerce business, as individuals stayed home but shopped online. So that was encouraging. We think that strength will continue, and we're encouraged with the international business. As previously announced, some of the blank sailings have been reversed, but there are customers, direct quote from them, they remain cautiously optimistic for the back half of the year. And just rounding out the portfolio with coal, the outlook remains challenged, both for domestic utility and for exports.

So as Jim said earlier, Brandon, virtually every one of our markets and customers were impacted by the pandemic. There remains a lot of uncertainty around the pace of the recovery and the continuing impact of COVID-19 on states and businesses. Couple that with what's going on with the upcoming election in November and other geopolitical issues, that all shapes our view that while we're encouraged with the increases that we're seeing today, we remain cautiously optimistic about the back half.

Your next question comes from Allison Landry from Crédit Suisse.

So just given the wild swing in volumes, are you guys seeing any challenges or choke points as the traffic comes back, whether in your own network or with any of these interchange partners? It just seems like the service metrics have deteriorated across the industry recently. So just curious what you're seeing from a fluidity standpoint and if you think some of the declines we've seen recently are just temporary blips.

Allison, yes, it's Jim. I think I can't speak for the other industries, but again our experience, as I somewhat alluded to in my earlier answer, when we had that kind of really quick drop and had to respond appropriately by reducing our workforce, normally, in a more traditional economic downturn scenario, don't expect your employees to have to come back to work or don't expect them to come back to work so quickly. And in the normal scenario, you're always worried about "Well, geez, these guys are off for 6, 8, 9, 12 months, are they going to be available for me to come back to work when the economy rebounds and I need to start recalling employees."

So that was our principal concern, working our way through this kind of maze and difficult period of time as to what is really going on here. And so that was our concern was, "Where are these guys going to be 12 months from now when I need to get them back, to come back to work?" Well, as we've now seen, we had this short period of time. And in that circumstance, meaning we called people back, they have quite a lengthy time to actually return to work when we call them and say, "Okay. Business is back. Surprise. Surprise. Can you come back to work tomorrow?" And the guy goes, "Well, you know, I got a couple of weeks here before I've really got to mark up and return." And then there's a process through the seniority system where displacements occur. So it's just a work-in-progress for us to ramp up and get the trains fully staffed so we can get our velocity back to where it needs to be, our on-time originations and arrivals where they need to be. And that's taken us a little while. And yes, it shows in the metrics, but we think that we are in a pretty good place right now.

And again, I can't speak for the other railroads. But I would imagine they have the same, practically identical labor agreements that we do. They kind of probably struggled with -- through this as well. And I just think now that everything is kind of smoothing out, we didn't see any kind of major congestion points throughout North America, like we sometimes do. So I think everybody is working very, very closely and effectively. There is a certain like-mindedness now in the operating structure of the North American railroads, where everybody kind of thinks the same way and acts the same way. And as we've said in the past, the scheduled railroading model, as it is implicated (sic) [implemented] across North America, is a good thing for the industry, because we all know how to move trains now more effectively and efficiently than we could in the past.

Okay. I mean, basically, it sounds like the takeaway is that you think this is mostly transitory. Is that fair to say?

Yes. I think we're good. I think, like I said, I think we're in good shape right now.

Your next question comes from Amit Mehrotra from Deutsche Bank.

I had a very quick follow-up first. Kevin, you've been helpful in the past, more recently, in helping us think about OR in terms of the 6 handle, et cetera. Wonder if you could do the same for the third quarter. Is another 6 handle kind of in the cards because of the volume declines? Or can we get back to the 5 handles, given the revenue growth is much more -- decline is much more benign?

And then, Jim, you also talked about -- I was hoping you could talk about the underlying growth initiatives that you guys have been working on pre-COVID, how they're tracking in terms of truck back to rail conversions. You obviously have a little bit more of a track record in terms of good trip plan compliance, and truckload rates are ticking up. So I was hoping you can give an update there as well.

Yes. Amit, this is Kevin. I'm not going to give out any handles today. I thought it was appropriate last quarter, but given where we are today and we're not guiding, I'll probably defer on that one. I will say, as the volume comes off the bottom, you're going to see a lot of operating leverage in the model, I'm very, very confident, across all the expense line items. If you look at it on an expense per GTM, say, we're going to see improvement across every area there. And look, as the volumes come, if they come back stronger than what we anticipate, that will continue to improve the OR. So we see -- we just did 58.7 in the first quarter. If anything, our cost structure, we've improved it this last quarter. So we pretty -- feel very confident once the volumes return, that we'll be back to where we were.

Amit, yes. And as -- in reference to your growth initiatives, Kevin just talked about as the volume of the business comes back, and I just talked about the equation here being -- the operating ratio being revenues and expenses, it's only appropriate that we let Mark answer the question of how he plans to grow the business here in the next couple of weeks.

Yes, Jim, thanks. So Amit, I think the journey continues on all our marketing efforts that we started gee, it's a little less than a year ago now. It seems like a long time, but we haven't been at it that long. But our -- clearly, first priority at CSX was to fix the service product. By and large, mission accomplished there, I think we've got the best service product going into this pandemic in the industry, and we're very pleased with that. And so it allowed us to really focus on growing this company and really looking at all the marketing initiatives that we just haven't done in the past. And the team has been hard at work looking at all those things, peeling back the onion with our customers' books of businesses, looking at opportunities, looking at business that maybe once was on CSX but was lost, that had gone away over the years, business that traditionally never moved by rail and moved by other modes of transportation, reaching out to those customers and explaining to them the value of CSX.

I think part of the big benefit too, as we've been going through this pandemic is as we talk to different shippers, we offer a very competitive truck-like solution and we're cheaper than the truck, by and large. So in a time when companies are looking to save money during this pandemic, we become, with our service, a very interesting service option for them. So we continue that journey. We're only 9 months or 10 months into this, and we're very, very pleased with a lot of the progress that has been made across the portfolio, everything from metals and forest products and aggregates and a whole bunch of commodities where we've seen a lot of truck conversions happening. And that, as I said, that journey continues. And the work continues. But so far, our American team has done an outstanding job identifying a lot of those opportunities, and we've got a great sales team that are actually going out and converting them.

Your next question comes from Scott Group from Wolfe Research.

So I don't know if this is for Kevin or Mark on the yield. So rev per car was down 7%, I think yields ex fuel were down about 5%. Can you give us a sense of the price versus mix component in the quarter? And then I know mix is a tough one to forecast, but directionally, does mix seem more or less negative in the third quarter relative to second? And any thoughts, Mark, on the coal RPU with the benchmark rates, if that goes down any further? Any sort of help on the yield side?

A lot questions-in-one here today. Just kidding, Scott, I'd be happy to answer those. So listen, let me attack the pricing side, it comes to the heart of your question, I think. And obviously, this was a tough economic environment this past quarter. But our pricing philosophy has not changed in terms of continuing to price to the value of the service product, which we think is really good. I was extremely happy in the quarter despite the tough environment, that we were able to secure several significant longer-term deals with great pricing in the quarter. So the team is doing -- despite the environment, the team is continuing to go out and get the price increases that we're looking for, and I'm very, very pleased with the work they're doing. On the merchandise side we saw positive pricing year-on-year, inflationary-plus pricing, really good pricing there. Intermodal, again, increases year-on-year. And then coal, while we can talk about that at another time, but clearly, a bit of a challenge, especially [long on] the export side and on the domestic side as well.

With respect to RPU, Scott, I think the big story on the RPU in the quarter was fuel surcharge, just given what's happened with fuel in the quarter. If you look at our intermodal, RPU was down 8%. If you exclude the impact of fuel surcharge, that was about a 5- to 6-point -- 5 to 6 percentage impact there, so we'd be down 2%, 3% on intermodal for RPU. And merchandise, excluding the impact of fuel surcharge, merchandise RPUs would actually be up.

So a lot of mix issues, as we've talked about, you and I, and the rest of -- and everybody else, always mix plays an ongoing impact in any given quarter, just with the different RPUs and the different screens in the various commodity groups that we move. So we're always going to see that. But clearly, the big story this -- the headline story this quarter was the impact of fuel surcharge. So I'm not going to give much guidance going forward. But hopefully, as things moderate here with fuel, hopefully that we'll see a little bit less volatility on that side of the business going forward.

Your next question comes from Chris Wetherbee from Citi.

Kevin, you ran through, I think, a couple of costs that were -- it sounded a bit transitory from a second quarter perspective. I think severance cost around management restructuring, which may actually go from, I guess, a negative in 2Q to potentially a positive as we move forward. And I think COVID steps down a little bit. Number one, do you mind just kind of running through those again so we understand what sort of some of the discrete moving parts are between 2Q and then 3Q, maybe back half? And then when your previous comment about sort of cost on a per GTM basis kind of improving sequentially if things -- volume comes off the bottom, would this be incremental to that? Just want to get a sense of maybe some of the magnitude there. I'm not sure it's a lot, but I just want to make sure we understood it.

Yes. I'll go through a little bit of it. So I called out the $60 million in total on a year-over-year basis. Part of that was the COVID supplies, which I called out around $10 million this quarter. Quite frankly, we did everything cost -- we prioritized speed over cost to get these supplies out to the field and protect the employees. So we experienced some pretty high cost on that, that will be coming down. We think it's about $2 million going forward on a run rate basis, which I said in my opening comments.

On the severance side, we said that was about $10 million, with about $25 million annualized savings related to that. We had some equipment impairments within the quarter, call it mid-single-digit on the equipment impairments. And then, obviously, real estate gains on a year-over-year comparison basis was down significantly, about $26 million. So -- and then I also called out, depreciation was up this year following a late 2019 equipment study that we went through, essentially making its depreciation up year-over-year, which, obviously, has a lot of different components within it, but doesn't necessarily reflect the lower capital base that we have today.

In terms of going forward, there's a lot of things that we've done, and I spoke to this a little bit, a lot of other initiatives that we're working on currently that will have -- that you'll probably start to see show up in the third and fourth quarters, and quite frankly into 2021, that take a bit of time. When I think about leases, buying those out, getting out of leases that we worked with Jamie very closely to identify, that we don't need anymore. Those take a bit of time to get out of and see the savings. The procurement team has done an amazing job working with our suppliers. We're fortunate enough to be in a position today where we have -- we're a strong balance sheet, suppliers want to work with us, that provides leverage. We're working on cost savings that have a longer tail that can, hopefully, have multiyear savings behind them, and so those things will start to show up, which I'm very excited about.

But there's -- we have probably over a list of 30 different areas, which we're going through that are nonlabor-related that we think are great opportunities: vehicle spend, utilities. We've done a great job on the fuel sourcing side. We think there's big opportunities there over the longer term. We've gotten really, really efficient. We're doing great there, but we can source our fuel a lot better, too. And Jamie and his team are really helping us there, think creatively on that side. And then I also mentioned the management [of] labor, that will obviously have some impact in the following quarters and in the next year.

So we're pretty excited about those. And now that we can get through the quarter. We have -- we're all going to meet again next week and go through the list, and hopefully, it continues to grow.

Your next question comes from Brian Ossenbeck from JPMorgan.

Maybe one for Jamie. Just want to ask about train size and weights as it applies to CSX, maybe the Eastern network in general. And we're hearing a lot of that from Western and Canadian peers. Maybe if you can provide some context as to how much they grew sequentially, as volumes fell, if that was one of the levers you could pull. And are there any upper limits to what the network might be able to handle either from an investment perspective or perhaps a physical with grade crossings, if you think that's really a difference, or perhaps a constraint, when you look at some of the peers operating with a lot bigger train size and weights during this pandemic?

Yes, Brian, we did see some good gains with both of our tonnage and length here in the past quarter hitting some records for ourselves, pushing -- push ourselves over 7% higher on our tonnage and around 4% on our length. We've only given a small piece of that back through the -- to the 25% of volume that we've seen. So we expect to see that continue forward. And that is a -- it is a big item that we talk about. When we continue to design this network, and we continue to look for those structural changes, is that we'd like to continue to see our train length grow. And depending on which corridor you're at, there's lots of opportunity on that end. So we're not going to give it all back. I can tell you that right now. And we'll continue to -- as we -- as I mentioned earlier, with the auto network and how we were putting that together with our intermodal network and our merchandise network, that gives us those opportunities to continue to grow our footage and our tonnage on our network, and we have got a great network out there that has long sidings, that has double track. That is not an issue at all with respect to our network. We do have some grades in areas that may be more of a challenge in some areas, but that's where distributed power comes in. And in order -- in some areas, if we use a third locomotive instead of adding a train start, and pushing a little more footage or tonnage, we'll continue to do that.

Some other things, I guess, to your second question somewhat. We have found some more opportunities out on the network with respect to some corridors that we can do some small investments on, which we are. I'd use an example of a trip that I was on a couple of weeks ago, and Kevin joined me on that trip around the railroad. We found an area where we were backhauling cars millions of miles, really, backhauling automobiles that were coming off of St. Louis would have to go all the way up to Toledo to come down to Cincinnati because we had a subdivision, the Indiana sub, which was not cleared for autos. This past month, our engineering team has done a fantastic job, or I guess the past 2 weeks, and clearing a way for undercutting bridges and allowing the auto network to go direct now from St. Louis via Indianapolis down to Cincinnati, again, saving millions of out-of-route miles. So we continue to find those kind of opportunities out there. The more and more we get out on the network, the more and more we work with the local people who are out there to ensure that we don't miss some of these opportunities, those are lasting structural changes that we will continue to see across our network.

Your next question comes from David Ross from Stifel.

Wanted to ask about the service levels. On Slide 10, both the trip plan performance and intermodal ticked down a little bit, a little bit more on the intermodal side. Was there also a difference between that quarter average, say on the 94% for intermodal, versus how it was the beginning of the quarter versus the end of the quarter? Is that 94% consistent? Or did it go from kind of a 96% to a 92% as you moved through the quarter and volume came back? And how does that relate to taking share back from truckload, especially as rates are improving on the truckload side?

I'll take on the first part, and then I'll pass it over to Mark to talk a little bit about the customer side. But on our trip plan side, exactly what happened is your example, [we] were towards the growth when we started to see this large pickup on our business. When we went to this 25% increase in volume in 6 weeks, we definitely saw a little bit of a -- deterioration, I guess, really -- in some of our trip plans across the board. We're in a much better situation now. But it is an enormous task to turn this big ship in 6 weeks to absorb that kind of volume.

Operating team, I would say, has done an amazing job. Marcelo Estrada, who runs our Intermodal business, has done a great job making sure that trains that did arrive late to some of our ramps were offloaded faster by utilizing some of the assets we had at that end. But to your point, some of those percentages have dropped and we're starting to see that rebound come back. But definitely, very, very large volume swing.

Yes. And I'd say, just on sort of truck conversions, I guess good news/bad news, yes, to Jamie's point, we saw some of the trip plan performance numbers fall below where I think we'd like to see them for Intermodal. I mean we were sort of 98%, 99%. And again, full visibility to our customers. We provided that last year in October. So they see exactly what we're -- how we're performing to these trip plans. In the depths of this crisis sort of in May, volumes were down pretty substantially. And so wasn't too much of an issue, and we were doing okay there. So -- but certainly, as volumes have picked up since the trough in mid-May, the service levels have improved quite substantially and continue to improve every day. So as these volumes come back, we're -- our trip plan performance is continuing to improve really nicely. And so I think if you talk to our channel partners, they would say that we were there for them even during the trough, and we'll continue to be there for them, especially as these domestic volumes are rebounding and coming back.

Your next question comes from Justin Long from Stephens.

Thanks, and good afternoon. So I was wondering if you could talk about your active locomotive count. I'm sure that's something that fluctuated throughout the quarter, just given the volatility that you mentioned in volumes. So can you just give us a sense on how that progressed over the course of 2Q and how you're thinking about the active locomotive count in the back half of the year? And then also along those lines, would love to get your latest thoughts on the locomotive overhauls that you plan to do this year and longer term?

Yes, absolutely. Look, our locomotive fleet year-over-year Q2 was down over 650 locomotives. We have, obviously, brought back some of those locomotives. So we were -- we did hit some all-time lows down in the 1,800s -- 1,800 locomotives when we had a business decline that was pushing down towards 28%. Coming back to where we are now, we're just into the -- above, into the 2,000s, and we will continue to monitor that. We monitor it really daily when we look at our locomotive fleet. We have, as Kevin mentioned, over 1,000 locomotives sitting in storage. And we do have some locomotives that are ready to go in different areas. We position them properly. Now when we -- we're very selective of what locomotives we put in storage. We made sure that we put down those that weren't -- that were higher fuel burn that didn't give us the best performance. And we tuck those really towards the back of the storage area as we continue to rebound and bring back some of the locomotives.

But that's where it's important that we continue to do a rebuild program. We believe in that rebuild program. We're finishing up what we committed to this year and next year as well, 30-plus locomotives that we're going to continue to rebuild into next year. I mean the fuel efficiency you get from those locomotives as well as the reliability. At CSX, we found early on that we had a tired fleet and we needed to do something about it. We had well over 4,000 locomotives 3 years ago, to where you see where our numbers are today. And at CSX, what -- in the past, they just continued to buy locomotives and just add them to the fleet and never retired anything. So we've taken the opportunity over the last 3 years to make sure we've got the most efficient, reliable locomotives out there, and that's what we're pushing across the network now. And particularly, on the distributed power and that technology wasn't really utilized at CSX when we first got here. And this team has done an unbelievable job wrapping their arms around that and understanding the importance of that fleet. And one last point with our locomotive fleet is, we're very lucky that the majority of our fleet is AC locomotives, which is more reliable. We've put down the majority of our DC locomotives. We did that a while ago, and we finished off changing out some of those to the AC fleet we had stored, which allows us to pull more horsepower per locomotive. And again, that reliability.

Your next question comes from Jon Chappell from Evercore ISI.

Thank you. Good afternoon, everyone. Kevin, you mentioned the elevated cash balance, and you've been able to add to that board without really adding significant debt. So you're still very net debt, and be in a good position there. Started buybacks in June, it looks like, which was probably earlier than you had anticipated the last time you spoke to us in April. Can you just give us some cadence around how we should think about capital return in the second half of the year? Is the third quarter going to maybe look like a measured pace like June? Or could we see a return to the type of buyback pace of prior?

Yes. I don't think we're sitting here thinking we're out of the woods yet. So I think the approach is going to continue to be very, very prudent, Jim and I talk about that a lot every week, but also opportunistic at the same time. I mean, clearly, on a long-term, medium-term basis, we don't need $2.6 billion of cash on our balance sheet. We need something well north -- south of $1 billion to really run our business. So we have a lot of excess liquidity right now. Certainly makes me feel good at night to go to bed and have that cash balance, just given all the uncertainty in the world today, but we'll use it effectively over the next few quarters. And talk closely with our Board as well on the strategy around that.

Your next question comes from Jordan Alliger from Goldman Sachs.

One thing I'm curious about on the wage inflation or said another way, your cost per employee, which I believe was down pretty sharply in the second quarter. I know there's moving parts with employees, furloughed employees coming back, but any sense you could give how we should think about that part of the equation going forward?

Yes. Look, there was a few moving parts to that in the second quarter. One, when you look at our headcount number, it included 300 on the reserve boards, on average through the quarter. And obviously, the costs related to those employees is lower than what you would see on average. And we no longer have those reserve boards when you look into the third quarter. So you'll see some normalization there. Also, I highlighted the incentive comp, and that will also have an impact in the quarter as well, and we'd expect that to normalize a bit here.

Sequentially, when you look at the headcount, probably something more flattish into the third quarter, on average, the same average that you saw in the second quarter, but I would expect the overall per employee costs to go up slightly here, particularly with the midyear union wage increase as well.

The next question comes from Bascome Majors from Susquehanna.

Mark, much earlier in the call, you spoke about your efforts to reconvert some freight that the carload business had lost over the years, and add some freight that has always moved on truck and hasn't moved on the railroad. As the trucking market seems to be moving to an inflationary pricing backdrop pretty quickly here, can you take a step back and let us think about where you are in the sales cycle in some of these efforts that may have been disrupted by COVID and its impact on your business and your customers' business? And just really, do you size this up as a needle-moving opportunity for CSX and investors over, say, a 2-year time frame?

Yes. Thanks, Bascome. Listen, I certainly hope so. I think, as I said, it's been -- this pause, I would say, has been sort of a refreshing time period for people from working at home, to sort of step back and reevaluate things and where we are and what the priorities are and where we're focusing and look at opportunities opportunistically going forward. And I think we are seeing the opportunity to look at a lot of those conversion opportunities, and we know there's a huge market out there for us to be able to go out and tap and address, that maybe just because for various reasons, over the years CSX has ignored or just didn't want to move it or it wasn't focused on those opportunities. And I think as we are looking to grow and looking for opportunities with a better service product, a lower cost basis, which opens up the opportunities for us to play in some markets that traditionally we couldn't before, and to price them. I think maybe this is sort of a defining moment here, and we'll see -- we're extremely excited by the opportunities going forward.

Now you've got to understand the environment we're in. We're in a very tough situation, given the current economic environment with COVID-19 and the economy where it is. But certainly regardless, we're still seeing great opportunities. And I think those opportunities will continue. And when the economy does recover and industrial production does resume back at full strength and the consumer economy is back at full strength and business is roaring again, I think you will see those efforts pay off big-time.

Your next question comes from Ravi Shanker from Morgan Stanley.

Maybe shifting gears and actually looking out a little bit. I don't know if you guys have paid much attention given everything going on. But the kind of INVEST in America infrastructure build that's kind of going through the House right now has a few kind of rail provisions in there, some of which kind of include things like mandating 2-person crews and launching Congressional investigations into kind of PSR and the impact of PSR or something. So just wanted to get kind of your thoughts on that, if you've spent any time with your folks in D.C. and maybe labor, just your latest conversations on going to 1-person crews and maybe kind of sizing the risk that if you do have some kind of political change in D.C., it could potentially kind of undo some of the gains you've seen on PSR over the years?

Well, seeing as how it's a Washington question, I think it's appropriate that I say I'll have to answer the question in a politically correct manner because I don't want to tell you what I really think of it. Everything comes out with a wish list that's put forward by a well-known people with agendas. And I don't know that politically, billions and billions of dollars of capital investment over years and years and years on how to run the railroad appropriately and safely is going to be undone by some -- with some political mandate. But we're certainly not focused on that. And -- but again, I've been around a long time. And there's -- I'm sure somewhere in some State House, somebody is drafting up a bill right now to put the firemen back on the diesel locomotives. So we're aware of the situation. We follow all these developments in Washington and elsewhere, but it's just a part of doing business. But it's not something that distracts us.

Your next question comes from Cherilyn Radbourne from TD Securities.

Just wanted to ask a question on overtime. You referenced some substantial reductions there this quarter as you have in prior quarters. So was CSX pre-PSR organized in such a way that it was more reliant on overtime? And what do you think the remaining opportunity is there? And would you expect that to have a knock-on benefit on safety?

Overtime, we've mentioned, I think, over the past couple of quarters, how important it was for us to get it in control, which we have and we've hit out some great numbers across the board. And it's also an area where we are able to use to flex as well. So even though -- I know Kevin threw out some numbers, which were some very good percentages that we were able to drop. There are some areas as the business started to come back as fast as it did, that we were able to flex ourselves and utilize a little bit of that pent-up overtime, if you want to say, by bringing it down. So is there more opportunity? Yes, absolutely, there is. On our engineering end, I know Ricky Johnson is working really hard with his engineering team to continue to bring down the costs, particularly with the amount of time that they're getting out on the railroad, to ensure that we're bringing those numbers down.

On the mechanical side, we've done an unbelievable job, which Kevin had mentioned, with respect to where our overtime is at. And yes, we'll continue to drive down those numbers. But ultimately, we'd like to get those overtime numbers down so then when there's the time where we need it for a small surge, we're able to utilize those employees on overtime. If we get into an area where we need to do some hiring, we're able to use that as a bit of a bridge for us to get to that point. Overtime and safety, I don't really -- there isn't really a lot of correlation for me on overtime and safety. Safety is, as Jim had started out the opening, is forefront for us. The #1 most important piece of railroading and how we start all of our discussions around the railroad, and will continue to be.

Your next question comes from Walter Spracklin from RBC Capital.

I wanted to focus on a little bit of the structural change that's been precipitated here by COVID-19, and really focus on 2 elements, and love to hear your commentary on both. First of all, is whether your capital intensity is something you're relooking at or you're reexamining. Mark, you talked about getting into new areas you haven't been in before. But are you looking as well to getting out of areas where you have been and that have been in significant and structural decline? Particularly on coal, could we see some line divestitures that would come about as a result of the significant decline in coal?

And the second part of this structural change is how well you and the railroads have done both in the global financial crisis and now this one, whether you reexamine your capital structure and whether additional leverage is something that you could contemplate going forward, given the consistency that you've been able to provide from a financial perspective through some of these very difficult times.

Well, let me start, and maybe some of the other guys want to jump in here with some brief comments. We're not in -- we don't have a railroad that's up for sale. We're very comfortable with our network. We've -- we're in the business right now of looking for business. We're open for business. We're chasing every piece of business that we can get after. Our strategy is to grow this company. Maybe that wasn't the case when the cost structure was what it was before. But based upon our levels of efficiency, what we think we can do in the marketplace, we're not interested in divesting. And so our network is rock-solid.

In terms of the financial reengineering, so to speak, based upon what we've been through, I think for a company like CSX that has dramatically changed its financial profile and dramatically changed its cash flow, I think what was important to the rating agencies and others was to watch us work our way through various different economic scenarios. And I don't think you could probably draw up a better stress test for a railroad to see how you're going to perform in difficult circumstances than what we are going through not only associated with the pandemic, but what we've gone through before that, with basically an industrial recession. So you take the industrial recession, you throw in there the pandemic on top of that, you start figuring out what's going to transpire in the fall with the elections. And all we do is keep kicking out cash day after day after day, hour after hour. I think, yes, that we have proven to the world that we have transformed this company into a very, very strong and stable financial company. That -- and right now, we have no -- obviously there's just no reason to do that. Kevin just told you we got the -- he's having a hard time finding a place to store the cash, that if we decided that we wanted to lever up further, we've already -- we've proven to the world that we are -- we have the capability and the wherewithal to do that. So I think it's a -- I think we're, again, proof is in the pudding. People asked us, what are you going to do, 2.5 years ago when we started on this adventure. This is what we told everybody what we're going to do. And now we -- and again, we've proven it.

Your next question comes from David Vernon from Bernstein.

Jim or Kevin, I just wanted to follow up on this issue of kind of cost in relation to volume. It looks like, from the commentary earlier, it sounded like we've just fallen below a water mark where it's too difficult to pull out cost or it happened too quickly.

I'm trying to get a sense for what level of RTMs do you think you need in the network to where profitability would be kind of stable on a year-over-year basis? Or is that not the right way to think about it?

Profitability stable from an OR perspective on a year-over-year basis?

Just in terms of like an EBIT level, right? So volume was down 21% or so, operating profit, 35% or so. And clearly, it sounds like there was some deleveraging from volume. What I'm trying to get at is, how much volume do you need to get back and does it matter if it comes back as coal or intermodal or merchandise, or is it just RTMS? Like is it 45 million RTMs? Like, what's the magic number where you feel like that deleveraging doesn't happen?

Yes. I mean I feel a lot better today where volumes are down mid-single digits versus the 20s that we were looking at. That's certainly helpful. So I think we're in a much better place right now. And as I mentioned, I think you're going to see some -- you'll see across-the-board improvement on a cost per GTM level as we move into third quarter, assuming these volumes continue to recover from what we saw in the second quarter. So I'm very confident there. And I'd remind everyone, it was just a quarter ago where we were -- and traditionally, what would be seasonally the worst quarter for us, posted a 58.7. And I would -- I think we'd all agree here around the table today that all the things that we've done in the past 3, 4 months has made us a better company, and probably provided more leverage as the volume comes back. So I'm more confident sitting here than I was when we reported in April that 58.7, that we can really leverage it. It's just waiting on the economy to improve and going after it. So we've got a lot of initiatives that we'll continue to work on, that will hopefully continue to improve that.

And just to answer your other question, an RTM is not an RTM is not an RTM. I mean that's just basic railroading. We had a lot more coal in the first quarter than we had in the second quarter, and we'll probably -- and then we'll have in the third quarter. So they're not all the same.

Yes. I guess is that going to create a little bit of an added headwind here if coal stays weaker, then? Or like, how should we be thinking about the mix impact of the -- where kind of volume fell out of the network?

It's only been going on at CSX for the last 10, 15 years. So I mean it shouldn't be a revelation that it's going to have an impact in the third quarter.

Your next question comes from Jason Seidl from Cowen.

Wanted to talk a little bit about peak season and how we should start thinking about it, and what types of conversations you've had with some of your partners out there. And also taking into account that July seems like it's going to be a -- a stronger month than June, which is seasonally a little bit different than historical averages. So would love to hear some commentary around that.

Sure, Jason. So typically, we have really 2 peak seasons at CSX. We have the sort of Labor Day, late -- early September, late August, peak season as kids are going back to school, and we're seeing goods coming in from Asia and around the world, getting ready for all the fall activities, Halloween and Thanksgiving and all that kind of stuff. And as people are going out and buying new goods for the back-to-school and all that kind of stuff. So I think, clearly we've seen some good domestic intermodal volumes here over the last couple of weeks, as the inventories have been replenished at some of the retailers. I think it's really wait-and-see. If we see that sort of that bump up here at, call it, around Labor Day. We'll see what happens with back-to-school. We'll see what the -- if the states reopen, how the reopening process is, how the COVID cases are. So it's really, in my view, and our discussions with our customers are, we're cautious. We're watching it, and it's -- I think you heard some of that from one of the channel partners who reported last week. It's a wait-and-see game. So we're there, and we will be there, and we'll have the crews and the locomotives and everything to be able to move that freight when -- if it does come, and we'll see.

So then the other sort of traditional peak that we see is really around Thanksgiving, more of the e-commerce peak, as people prepare for Christmas and order goods online. Again, I think we'll probably see a peak. I think the peak may -- again, depending on whether people are back at work, whether people are -- the status of -- are the shopping malls open, are people comfortable going to the shopping mall. If the COVID cases are still sort of concerning and people are cautious about venturing out, then I think you may see that e-commerce type of peak a little bit sooner, as people stay home and sort of order stuff on the web. So we'll see what happens. But again, as we've said many times, all this is really a wait-and-see period here for us as we move into the back half of the year.

I think that e-commerce peak has been happening since almost April, at least in the Seidl household, for sure.

We have seen some good volumes on the e-commerce business. I mean, clearly, people are not venturing out and they're at home and they're ordering toilet paper and other things on the web. So our e-commerce peak is -- our volumes have been pretty strong in that area.

Okay. Well, I'll keep our fingers crossed for the outlook for after Labor Day then. Gentlemen, I appreciate the time.

Your next question comes from Ben Hartford from Baird.

Jamie, just wanted to get your quick take or clarify some comments made earlier about the safety performance. I think Jim had said that it was largely a function of lower volumes, which makes sense in terms of the uptick in kind of the index, whether it's the personal injury rate or train accident rate. But you're obviously undertaking some safety engagement initiatives as well. So what -- should we look at that as almost exclusively a function of lower volumes, and so we're seeing incidents pop up on a per unit basis as a result of that? Or do you think that there was some slippage in safety performance on an absolute basis, whether it was from service or some other factors that's leading to some of these initiatives that we're seeing in the third quarter?

Really, I think what Jim was mentioning, more than anything, was -- these are ratios, right? So if I take a look at the incidents that occurred, they are decreased, okay? There's less injuries, there's less accidents, but it's based off of a per mile and/or per hour worked. So as we had 1,000-plus employees -- really almost 1,300, 1,400 employees at one point -- furloughed, those hours disappear. So it creates a bit of a headwind when you take a look at how we do our FRA ratios.

And on the injury side, we were at 0.86, which is still top line in the industry versus 0.81 in 2019. And on the accident side, we were at 2.19 this year versus a 2.47. So we did see an improvement on the accident side. It's the injury hours that are different. But I mean, look, with that said, heading into the future, I think every railroad has got a little bit of a headwind with respect to engineers who went on the ground to work as conductors because they have seniority flowback rights. So they're doing jobs that they weren't used to doing. Retraining was something that we weren't able to do in most circumstances, because of social distancing. So we were doing the best we could to provide distance training and/or working with the employees to ensure that they're -- obviously were qualified, they were just doing jobs that they hadn't done for a number of years. So I think that's a headwind that we'll probably see across the entire industry. But absolutely, this past quarter, the man-hours that were removed from the furloughs is what would have affected our ratio.

There are no further questions at this time. I will turn the call back over to the presenters.

All right. Well, great. Thank you so much for spending your -- some time with us this afternoon, and we look forward to seeing you all in the near future. And be safe.

This concludes today's teleconference. Thank you for your participation in today's call. You may now disconnect your lines.