Jul 19, 2019
Operator
Good morning and welcome to the Kansas City Southern Second Quarter 2019 Earnings Conference Call. All participants will be in listen-only mode.
[Operator Instructions]. After today's presentation, there will be an opportunity to ask questions.
[Operator Instructions]. Please note this event is being recorded.
This presentation includes statements concerning potential future events involving the company which could materially differ from events that actually occur. The differences could be caused by a number of factors, including those factors identified in the Risk Factors section of the Company's Form 10-K for the year ended December 31, 2018, filed with the SEC.
The Company is not obligated to update any forward-looking statements in this presentation to reflect future events or developments. All reconciliations to GAAP can be found on the KCS website, www.kcsouthern.com.
It is now my pleasure to introduce your host, Pat Ottensmeyer, President and Chief Executive Officer for Kansas City Southern.
Pat Ottensmeyer
Okay. Thank you and good morning, everyone.
Welcome to our second quarter earnings call. Just to make a comment on Slide 4, which is the list of the line up for today’s presentation.
I’m not going to read through this. I think it is very well known to the audience here.
So I will move on to Slide 5 with a quick overview of the quarter and a few headliners. As you saw from the press release, revenues increased 5% to a record level for any quarter of $714 million.
Second quarter operating ratio, little bit of explaining necessary here. We reported a 70.9% operating ratio, which includes the impact of impairment charge, additional impairment charge we took in the second quarter.
Adjusting for that impairment charge, we got to 63.7%, which is a 30 basis point improvement from last year. That included negative impact of 130 basis points of pure accounting, I will say, related to the loss of the IEPS tax credit and the increase of fuel surcharge revenue.
Mike Upchurch will go through this in much greater detail later and explain the 130 basis points and probably more importantly explain what this looks like going forward. This is really pure accounting as we're just inflating the revenue side as well as the cost side of our income statement, which has a negative effect on the operating ratio, but has very little effect on operating income or other things that matter as you go down the income statement.
So I think it is correct to look at the 30 basis point improvement from the adjusted level, excluding the impairment charge as well as the 130 basis point excise tax impact that we absorbed or overcame in the quarter due to the pure accounting consequences of both higher fuel costs and higher fuel surcharge revenue. And I think Mike has a slide later on that goes through that in really good detail.
Second quarter diluted EPS, $1.28. Adjusted again for the impairment charge was at a record $1.64, increase of 6% versus prior-year.
Couple of key themes that we will probably cover 2x or 3x over the course of the call and Q&A. Operating performance significantly improve, which drove improved customer service.
Operating metrics and cost profile, again looking at the improvement in core operational savings, which is the 30 basis points year-over-year plus overcoming the impact of the fuel, the IEPS, the tax situation. And as you'll see later on we’re making great progress in PSR implementation and other operational excellence initiatives.
And we are going to give you updated guidance on the full-year impact and kind of the 2019 impact of those savings and those initiatives, and the punchline is, we're more than doubling the positive impact of those savings. I am very, very pleased with the commitment and enthusiasm of the team.
It's truly a cross functional. It's a very high level of commitment and engagement across the organization.
We’ve weekly PSR update meetings, led by Sameh, with a large number of people from many different organizations. And I said on those periodically, I sat in on the meeting we had on Wednesday of this week and is just very, very pleased with the level of engagement that we're seeing.
Another common theme and that you will hear more, I mean, obviously the volume outlet, outlook or volume performance was not as strong as we would have like. Volume is basically flat.
But there's a story behind that, that I think Mike Naatz will reveal in his comments that we look at our portfolio of business and breakdown by business unit going forward. We still see 60% of our portfolio as favorable in terms of the growth that we are seeing, 10% -- I’m sorry, 20% neutral.
So that could flip either way as circumstances develop and market conditions change over the course of the year and only 20% negative. And as you can see in the difference between our Carload volumes and our revenues, there is a mix story here.
So we certainly don't see gloom and doom ahead of us and opportunities for growth in some of the oversized growth areas that we’ve talked about for the last several quarters. And the PSR work we’re doing, the focus on operational excellence has put our network in really terrific position to deliver strong operating leverage as and when volume growth and revenue growth improve and Mike again will give you more detail on our estimates, our current estimates for PSR savings going forward.
Flipping to Slide 6, just want to update our guidance. This is similar guidance.
Categories that we gave at the end of the first quarter and we’re changing a couple of these. Volume growth we now see flat to slightly down for the full-year in 2019.
But again I'll restate what I said a minute ago, when Michael get into this in more detail. When we look at our total portfolio, we think 60% of our business is showing favorable positive trends, 20% neutral and 20% negative, which includes Intermodal which of course is driving the big explanation between the flat volume levels, but continuing to hang in there with our revenue guidance that we gave earlier which was 5% to 7% for the full-year.
Same guidance we gave at the end of the first quarter, so we still feel that that revenue growth is going to be possible for us this year. Operating ratio, we are moving to the low end of our previous guidance of 60% to 61% by 2021.
And I will again bring up the impact of the fuel tax issue, the loss of the IEPS tax credit. So not only are we going to the low end of the range, but we are absorbing the impact of the loss with the IEPS credit and the inflation of the balance sheet, both on the cost and the revenue side that, that will have to operating ratio going forward.
And as I mentioned Mike will get into a more detail about the magnitude of that during the quarter and then going forward. EPS, we're staying with our previous guidance of low to mid teens on a compound annual growth rate for the next three years.
So no change in EPS. And then, CapEx, we are bringing our capital expenditure forecast for 2019 down.
We are confident that we will be able to trim our CapEx forecast from previous guidance by $40 million to $60 million. Our previous guidance was $640 million to $660 million for 2019.
We now expect that number to come in below $600 million. And again, a lot of that is a function of improved operational efficiency.
And we're really just beginning -- in fact had a kickoff meeting here a couple weeks ago with the team who is looking at completely redesigning our transportation service plan, as Sameh will refer to this as whiteboarding when he gets into his comments, which we think will have further benefits in the second half of this year and beyond in terms of the number of train starts and the yard congestion and overall improvement in our operating performance. So, with that, I will turn the presentation over to Jeff Songer.
Jeff Songer
All right. Thank you, Pat.
Good morning. Starting with the review of key operating metrics for the quarter on Slide 8.
Dwell of 21.2 hours, improved 11% year-over-year, while gross velocity of 12.5 mile per hour improved 9% year-over-year. We continue to see strong year-over-year improvement in our key metrics, particularly in Mexico were dwell across the majority of our major terminals has shown significant improvement.
Flooding had some impact velocity interchange in the North end of the U.S network as we encourage several days of outages across some segments. The greatest impact was to interchange of bulk commodities in Kansas City.
Regarding asset utilization, I would like to highlight that the number of operational cars online has decreased 17% versus peak 2018 levels. Our focus has been to reduce high cost foreign equipment by improving cycle times and working with customers to right size fleets.
Border performance for the quarter remain solid and supported a 10% year-over-year growth in cross-border volumes, led by significant growth in Mexico Energy Reform shipments. A milestone was reached in our border operations in July as international crews have now been in place for one year.
This initiative continues to support cross-border volume and we're seeing a 20% reduction in time spent crossing the bridge for those trains utilizing international crews. Expanding upon PSR metrics on Slide 9, we continue to see positive trends in all metrics.
Through the first half of the year, gross velocity has improved 10% versus prior-year and we're now tracking this metric to exit the year at 14 miles per hour. Reductions in line of road equipment failures, improved execution of TSP and modifications to our service design and reduced train starts have all helped to a system fluidity.
Train length improved by 2% as our TSP and train starts consolidation work continues. In the second half of the year, we will undergo additional service redesign, which Sameh will discuss in more detail.
Fuel efficiency has improved by 3% for the year. Multiple initiatives, including velocity improvements, locomotive utilization, train start reductions and fuel technology are all contributing to this improvement.
In the second half of the year, we will install fuel technology on an additional 100 road locomotives, which will bring us to roughly 50% of our road fleets equipped. Also, of note, we’ve received all 50 of the new locomotives planned for 2019 and do not currently plan any locomotive purchases for 2020.
As Pat indicated, we feel very good about the progress and pace of our PSR initiatives and as you will see in Mike's financial review, we are more than doubling the expected PSR cost benefits for the full-year 2019. I will now turn the presentation over to Sameh to provide more detail on our overall PSR efforts.
Sameh Fahmy
Good morning. As Jeff mentioned, we have been making significant improvements in network velocity in about 10% that are year-over-year and well about 8% better.
We are seeing the best consolidated gross velocity this time of year across the network than we have seen in any of the last five years. We also see further improvements ahead as on many days we achieve 15, 16 miles per hour in U.S and 13, 14 miles per hour in Mexico.
These are much higher numbers than the average of 11 miles per hour reported last year. These improvements were achieved by very hard work every day by the whole operation team with intense focus on every low velocity train and peeling the onion on causes.
Example, a train that arrives in Nuevo Laredo and departs 24 hours later as it waited for a window to cross the border bridge. That has also been an intense effort to scrutinize work in yards.
As an example, Vanegas having nine trains doing work versus a planned six. As we drill down, we find that cars were added to a train in Sanchez and now needed to be switched out in Vanegas which is not our best yard for switching.
Velocity which is not our best yard for switching. Velocity and dwell gains translated into the removal of 12% of our locomotives and 7% of our cars and great improvement in fuel efficiency.
In addition, four waves of train consolidation resulted in a 10% reduction in planned line of road crew starts in Mexico. We continue to focus our efforts into converting our operational gains to the bottom line and as Pat mentioned, Mike Upchurch is going to go more into the financial details of this.
Now moving forward, the same way as a field visit to Houston in March brought fruits in partnership was -- was Union Pacific. A one week visit to San Luis Potosí in June was a true eye opener.
We observed a low level of compliance to plan with yard assignments adjusted daily, which is an indication of a suboptimal train plan with field personal trying to compensate for the deficiencies. But by doing so they create impure car blocks that need to be a reclassified in the next yard, and the one after that and so on downstream.
This is when it became clear that a new redesign, the whiteboarding as Pat mentioned earlier, needs to be done was great focus on yard and the border and will be completed in Q4. The objective will be to eliminate excessive classifications, free up capacity, increase velocity of reduced dwell and improve service.
As you know an important part of PSR is to create capacity for free. And we're taking a big chunk out of our original capital plan as Mike will describe in this presentation.
Example, money that was set aside to upgrade yards is being taken out and we are going to the side, what to do with that when the plan, the new design plan for trains and for classification and for yards is completed. An immediate action ahead of the new TSP is to implement on August 12 a new blocking agreement for northbound traffic interchanging with the UP at Laredo that will emphasize creating blocks at origin points, further south in our network and will reduce a significant portion of the switching at Nuevo Laredo yard.
That will reduce the number of times we touch a car, which again is one of the key principles of PSR. Mechanical delays will also be a great focus.
We have made significant improvement in failures per day per year in auto locomotive from 4.6% down to 3.1%, but will still have lots of room to go to take it down to 2.0 and we still have a lot of room to go on cars. We get about 40 to 45 hours of train delays per day due to mechanical and that can be cut substantially and we’re working on it.
Last and this has come up on the -- in the past couple of days, we are also beginning to shift focus to the engineering side to increase the productivity of our gangs, ties per hour and rail per hour -- rail footage per hour, and that would allow us to shrink the curfews and affect less the train delays and at the same time improve our capital dollar. I will turn it now to Mike Naatz for the commercial slide.
Mike Naatz
Thank you, Sameh, and good morning everyone. I will start my comments on Page 13.
Overall, second quarter revenue was up 5% on flat volumes. As you can see our Q2 performance was mixed across business units.
We continue to see very strong performance in our cross-border traffic, led by the Mexico Energy Reform business and this drove growth in our cross-border revenue and volumes of 13% and 10%, respectively. We are mindful of the economic environmental trade situation and we continue to carefully monitor the current state of affairs.
And as you’ve heard from Pat, 80% of our portfolio have a favorable or neutral outlook. During the quarter, our revenue per unit grew by about 4%, driven by a shift in business mix from the lower revenue per unit intermodal volumes to the higher revenue per unit chemical and petroleum business.
Increased fuel revenues also affected our revenue per unit. During the quarter, we achieved pricing renewals consistent with the prior year and we view the current pricing environment is stable.
As expected, the Chemical and Petroleum segment delivered strong quarterly year-over-year results with our Mexico Energy Reform business providing revenue growth of 136% and volume growth of 125%. Our Ag/Min units showed a slight year-over-year volume decrease to a large part to an unexpected customer outage and timing of volumes contributing to tougher year-over-year comps.
However, we continue to see favorable transportation cycle times, resulting from the benefits of operational improvements at cross-border traffic. We expect to see this trend continue into the second half.
Energy volumes were down due to decreases in frac sand and Canadian crude or partially offset year-over-year growth in utility coal traffic. And this was despite the flooding impact that occurred in the second quarter.
Year-over-year revenue in our industrial consumer business unit was down 2% and 7% lower volumes. The volume declines were primarily attributable to softness in the paper market, which was driven by abundant truck capacity and a shift from cardboard to alternative packaging methods.
The Automotive segment saw revenue increase of 5% on flat volumes. Plans were impacted by planned outages early in the quarter and we expect year-over-year growth to improve in the second half of the year as plants reopened and new production begins to slowly ramp up.
Intermodal segment was down slightly, seeing a 1% reduction in revenue and a 3% decrease in volumes. We continue to see strong growth in our cross-border intermodal business with a 10% increase in volume and a 7% increase in revenue.
This was offset by decreases in domestic traffic due to truck availability and the Lazaro volumes and revenues were up 2% and 4%, respectively. Moving on to Slide 14.
You will see our revenue outlook for the second half of 2019. And as pat mentioned, we are expecting year-over-year volume to be flat to slightly down, but we do maintain our revenue guidance on the lower end of the 5% to 7% range.
We remain very positive on the chemical and petroleum business unit and the Mexico Energy Reform story. Additionally, we expect to see growth in the Automotive and Ag and Min segments, primarily due to continuing service improvements.
The Industrial and Consumer segment outlook is adjusted to neutral for the second half of 2019, given recent trends in the paper markets and available truck capacity. And truck availability paired with pricing impacts have increased fuel costs will drive a somewhat unfavorable second half outlook for our intermodal business unit.
And with that, I will turn things over to CFO, Mike Upchurch.
Mike Upchurch
Good morning, everyone. I will start my comments on Slide 16.
Second quarter revenues were up 5%, primarily due to a 4% increase in revenue per unit. Reported operating expenses increased $69 million, primarily due to a restructuring charge of $51 million consisting largely of an impairment of locomotives and freight cars no longer required in our business.
Adjusted operating expenses increased 4% leading to a 63.7 adjusted operating ratio for the quarter. It is important to note that the Mexican government's decision to terminate the fuel excise tax credit for the rail industry beginning April 30, 2019 had a negative impact on our adjusted operating ratio of approximately 130 basis points.
I will review more details on the impact to our financial statements on the next slide. Setting aside the negative impacts of the loss for the fuel excise tax credit, our operating performance was far better than our reported or adjusted results.
Our adjusted effective tax rate for the second quarter was 29.2% in line with the 29% to 30% guidance we provided. For 2019, we do expect our full-year adjusted ETR to be in the range of 27% to 28% due to the Mexican fuel excise tax been included in income taxes.
Since IEPS is now been eliminated for the remainder of 2019, we will no longer adjust the fuel excise tax credit in the operating expenses. In June, the Treasury Department proposed additional guidance concerning guilty tax and we now expect our effective tax rate to be 28% to 29% in 2020 and beyond.
And from a cash tax rate perspective, we're forecasting 2019 to be 21%, 2020 to be 23% and 2021 to be 25%. And while the Treasury Department has largely solved concerns that caused unintended consequences to KCS, we will still end up paying approximately $16 million in taxes in 2018 and 2019 that we believe were not intended under the principles of Tax Reform Act of 2017.
Finally reported EPS was a $1.28, which is inclusive of a $0.38 impact related to our restructuring charge. On an adjusted basis, EPS was $1.64 and was predominantly due to increased operating income, although we also saw some benefit in reduced share count.
Turning to Slide 17, I want to now review the estimated impacts of the loss of the IEPS fuel tax credit. In the first column just represents our adjusted 2Q results through operating income.
The second column represents the estimated 2Q impact of losing the fuel tax credit beginning April 30. And the amount of 2Q revenue build the customers in our fuel program that started May 11.
As you can see, we recorded an incremental $7.2 million in revenue during the quarter that would not have been recorded had the excise tax credit been retained. Offsetting that incremental revenue is the $13.6 million lost fuel excise tax credit resulting in $6.4 million less operating income during the second quarter or a 130 basis point negative impact to OR.
The final column is our attempt to provide you some guidance for future quarterly periods. The final column reflects the same loss, excise tax credit of $13.6 million, offset with an estimated $12.8 million recovery in our fuel program.
We do not recover a 100% of loss credit as some customers who have all-inclusive rates including fuel costs that won't be able to be adjusted until contracts expire. As illustrated by the operating income impact of less than a $1 million in the quarter, we do not expect a material impact to operating income or net income from the lost fuel excise tax credit.
However, moving the lost fuel excise tax credit from expenses into our fuel program will create an approximate 80 basis point deterioration in our OR since the increased fuel expense will have essentially an equal amount of fuel surcharge revenue. So if there is three takeaways here, the quarter was considerably better, if you set aside the loss for the credit.
There's an immaterial impact to operating income and net income, but we do have the dynamics of an 80 basis point increase in our operating ratio as a result of moving the credit into revenues. So with that, let's turn to more exciting things on Slide 18.
I want to provide you an update on our PSR expense savings. On our first quarter call, we reviewed an initial PSR saving estimate with you.
At that time, we had visibility into a reduction of approximately $16 million of 2019 expenses or $25 million annually. After an additional quarter of work, we now have visibility to 2019 expenses that are 2.5x our previous estimates or $40 million for 2019 and $55 million annually.
As you can see in the slide, we're seeing better progress in every single expense category. Some of our early initiatives that focused on reducing the size of the fleet, particularly older less reliable locomotives.
Not only are we seeing depreciation and maintenance benefits from a reduced fleet, we’re also experiencing fuel benefits by shedding older less fuel-efficient locomotives. And we’ve improved the reliability of the fleet by reducing service interruptions by 35% and reducing the duration of service interruptions by 15% or 45% that have led to a 15% reduction in our mechanical resources.
With respect to fuel, we are also beginning to see more savings materialize from the adoption of fuel saving technologies, such as GE's trip optimizer to run trains in the most fuel-efficient gear or notch, AESS to minimize idling time and smart HPT to right size the locomotive power being used for each train. Additionally, train consolidation is also helping fuel efficiency by running longer trains.
So in summary, we will continue to aggressively pursue more expense savings as we streamline our operations and develop a new transportation service plan. In the quarter, we did incur a $51 million restructuring charge which is predominantly related to disposal of more locomotives and freight cars that we won't need in our business.
As we continue to develop plans to streamline our operations, it is possible we could have further restructuring charges in the future since for we are only six months into our efforts. Turning to Slide 19.
Quarterly adjusted operating expenses increased 4% over 2Q 2018. KCS generated a fuel excise tax credit of $8 million in the second quarter of '18, but obviously no credit in operating income in 2019 due to the change in tax laws, thus creating a negative year-over-year comp.
Wage inflation of 3% in the U.S and 5% in Mexico led to a $4 million increase in compensation costs and foreign exchange contributed $2 million increase. Offsetting those increases were $2 million in savings from an improvement in fuel-efficiency, $1 million in labor savings due to crew start reductions and $4 million of improvement in equipment expense, the result of a more fluid network as we saw our foreign cycle times continue to improve 8% year-over-year.
On Slide 20, compensation and benefits expense increased $6 million or 5%. Average quarterly headcount excluding the in-sourcing of a mechanical facility and in-sourcing of IT contractors declined 50 basis points.
Wage inflation increased comp and benefits by $4 million and in-sourcing and FX each contributed to $1 million increases. But we did begin to see approximately $1 million in savings from crew start reductions and continue to believe our overall headcount will be down for the year as we continue to make progress with PSR, particularly once our new transportation service plan is finalized.
Turning to Slide 21. Fuel expense increased $2 million due to increasing fuel prices in Mexico.
As you can see in the bar chart, Mexican fuel prices have continued to increase from $2.86 a year-ago to $3.15 in 2Q, a 10% increase. Offsetting those increases, we achieved efficiency gains by moving about the same GTMs this quarter as we did in 2Q 2018, but using 3% fewer gallons, an indicator of technology improvements and running more fuel-efficient trains.
Finally on Slide 22, let me briefly discuss our capital allocation priorities. As our free cash flow continues to grow, we will have a nice problem.
A significant buildup in cash enough to continue to invest in our business and return more capital to shareholders. Our top priority is to continue to invest in our business to generate better revenue growth.
However, we are reducing our 2019 capital expenditure guidance from $640 million to $660 million to less than $600 million. This reduction is due to two factors.
First, our PSR effort to freed up capacity on our network as fluidity improves. And second, we're adjusting our capital program to the realities of less than expected volume growth.
We’ve reduced investments in various growth and capacity projects until the volume environment is more predictable and we have finalized our transportation service plans. That said, we will continue to invest in our cross-border capacity to stay ahead of the double-digit growth we've seen in the last 18 months.
As for shareholder returns, we continue to buy shares under the $800 million share repurchase program authorized by our Board in August of 2017. During the quarter, we repurchased 773,000 shares for a total of $92 million and are pacing ahead of a pro rata schedule to repurchase our shares under the authorized program.
In light of this acceleration, Management plans to review our capital allocation strategy with our Board of Directors before the end of the year. And with that, I will turn it back to Pat.
Pat Ottensmeyer
Okay. Thanks, Mike.
I will just touch on a couple of things and restate a few things for emphasis before we open it up for Q&A. We feel really good about the progress we're making.
Obviously, we'd like to see a more robust volume environment that as Mike explained, 60% to 80% of our portfolio we still see as favorable. And with the exception of intermodal, we still see good growth opportunities in the quarters ahead and the -- what we -- what I refer to as the oversized growth areas that we talk about a lot.
Automotive, refined products moving into Mexico and plastics coming out of the Gulf Coast. So those businesses still look very solid.
They’re producing both volume and revenue growth and in spite of overall flat volumes we're still seeing good revenue growth and sticking with our volume revenue guidance from earlier this year. Mike also mentioned our cross-border product continues to grow double digits in both volume and revenue and we feel there's lots of opportunity for that to continue.
We are producing and delivering core operational improvements, which are driving improved service levels, better asset utilization and increased profitability and cash flow. We are handling the same volume levels overall that we have of last year with 12% fewer locomotives, 7% fewer railcars and 10% fewer crew starts.
And we think there's more to come in the months and quarters ahead as Sameh and Jeff both mentioned, we're just beginning to completely restructure and whiteboard our transportation service plan, which we are confident will lead to further asset savings and cost savings and service improvements in the months and quarters ahead. We are taking our PSR related operational savings up by more than 2.5x in terms of impact to 2019 and beyond.
And we will continue to update investors as we get further into this and particularly in the third and fourth quarter when we see the results and the benefits of our TSP whiteboarding exercise. And we still feel very good generally about our overall portfolio and I will just restate 60% to 80% neutral or favorable, a few outliers that are having a bigger impact on volumes than they are on revenues, reducing our CapEx accordingly, improving free cash flow.
I think our free cash flow conversion rate is increasing as well. So, all in all, in spite of some troubling signals in the economy and still some unresolved trade issues, we feel really good about the progress we are making and still feel good about the outlook.
So with that, I will open the call for questions.
Operator
[Operator Instructions] The first question comes from Chris Wetherbee of Citi. Please go ahead.
Chris Wetherbee
Yes, hi. Thanks.
Good morning, guys.
Pat Ottensmeyer
Good morning.
Chris Wetherbee
Maybe, Mike, can you start on Slide 17. I just want to make sure I understand sort of what the moving parts with the fuel tax credit this week and kind of get that out of the way.
So in the second quarter I think you guys have been expecting an 80 basis point headwind that ended up coming in a bit bigger than that. Sort of the way that we’re thinking about this now, the way that we should be looking at it going forward.
So maybe if you could sort of parse what happened, the difference in the quarter between the 80 and 130? And then as we go forward in the context of the rest of the year and obviously the 2021 guidance, we should be assuming that include this 80 basis point ongoing headwind?
Mike Upchurch
Yes, exactly, Chris. The guidance assumes we are going to absorb this negative 80 basis point impact and we are going to be on the low end of the range.
So you should view that as improvement driven by both of those items. And the 80 basis points is what we believe will be the impact going forward in each quarter.
The reason it was a little higher in the first quarter or second quarter was really for two reasons. One, we have the timing issue of losing the credit on April 30 and not getting it on customer bills until mid May.
And then, secondly, since we are not going to adjust this anymore, the -- in quarter benefit related to the IEPS credit that we did get is sitting down in income taxes. But that's exactly right.
I think you got that right.
Chris Wetherbee
Okay.
Mike Upchurch
Chris, there is a -- the 80 basis points over time there is a part of our as Mike mentioned, a part of our customer base that pays and -- an all-in rate so between contract renewals we don't have specific fuel surcharge. So it'll take a while for us to catch up, if you will, and perhaps adjust those contracts to reflect that all-in rate versus rate plus fuel.
So that 80 basis points could come down as those contracts are renewed. And it's also important, I think we’ve mentioned this in other venues, we are challenging this decision.
The loss of the IEPS tax credit applies to railroads only. So our truck competitors and other modes of competition did not have the same regulation and rule come down in their case.
So we think it's discriminatory for us and we're challenging it and we certainly elevated this in our dialogue with government officials in Mexico, but certainly don't want to make any prediction about the outcome of that. But we're trying to defend ourselves here from a market competitiveness standpoint as well as our customers interests, and that will continue to do that.
Chris Wetherbee
Okay. And ultimately there were very little impact in operating income.
It looks like less than a $1 million a quarter going forward.
Pat Ottensmeyer
That’s the point. Because you’re increasing your revenue items and an expense item by roughly the same amount, it has a bigger impact on operating ratio.
But when you really go down the income statement and think about the dollars and paces involved it's minimal.
Chris Wetherbee
Okay. And if I could ask a follow-up just on the capital structure.
Mike you had some comments there. It couldn’t help, but notice that CapEx guidance come down a little bit.
Obviously, profitability with the OR improving, is going up. The leverage is sort of trickling towards the lower end of maybe where the peers are moving, so how should we think about sort of the capital structure as we go forward?
It seems like a high-class problem to have that free cash flow generation is likely to step up as we move forward and we think about things like accelerated stock repurchases or anything else that might be thinking about from a capital structure perspective.
Mike Upchurch
Yes. Well, you got that exactly right.
We’ve a nice problem and we will continue to have nice problem. More and more cash flow generation as a result of the items that you indicated, obviously increasing cash from ops and lower CapEx, we are going to create a better dynamic for us going forward.
Obviously, this is a decision we will have to reach with our Board and I did indicate we would be meeting with the Board specifically on our capital allocation policy before the end of the year. So stay tuned.
We don’t have any news today, but we will soon.
Chris Wetherbee
Okay, perfect. Thanks very much for the time.
I appreciate it.
Pat Ottensmeyer
Okay. Thank you, Chris.
Operator
The next question comes from Amit Mehrotra of Deutsche Bank. Please go ahead.
Amit Mehrotra
Thanks, operator. Good morning, everybody.
Pat Ottensmeyer
Good morning.
Amit Mehrotra
Mike, I’m not going to ask you about the excise tax credits, so you’re welcome.
Mike Upchurch
Thanks, Amit.
Amit Mehrotra
But I did want to ask about the revenue growth guidance being left intact despite the lower volume outlook. Obviously, it implies a uptick in yields in the back half compared to the first half, which typically isn't the case if you look at prior years.
So if you could just help us think about the underlying assumptions there, discuss a little bit more comfortable with the assumed deceleration?
Mike Upchurch
Yes, sure. We had indicated in or Pat has indicated kind of volume being flat to slightly down.
If you look at our results through the first half of the year, we were down roughly a point. A lot of that related to first quarter, teacher strikes, but obviously we are not -- we certainly hope won't happen in the back half of the year, so flat to slightly down.
And then I think you need to think about mix. Obviously, the intermodal business is a little bit challenging in today's environment.
So with a lower revenue per unit you can read into that. We’ve got a little bit softer outlook on intermodal in the back half of the year, but offset by refined product which has a very good length of haul and what we think will be good grain volumes going forward.
Amit Mehrotra
Right. So it's not -- it's really the mix of volume as oppose to any assumed acceleration in pricing?
Is that the way to think about it?
Mike Upchurch
Correct.
Amit Mehrotra
Okay. And then let me just -- for my follow-up, I just wanted to ask about the different buckets of the cost structure.
Are we going to see a more sizable reduction in the workforce. It's by far the biggest piece of the cost structure.
I would just imagine the structural improvements you are making could translate to great opportunity for labor productivity whether that means proactive reductions in workforce or maybe a limited need to replace those who leave to attrition on your attritions a little bit, may be high single digits every year. So can you just talk about that and when, if at all more step function increases labor productivity are going to be part of this plan going forward?
Pat Ottensmeyer
I'll provide some initial comments here and then maybe ask Sameh and Jeff to provide little more detail, but -- I will state the obvious the best of all possible outcomes is that we improved labor productivity and then we see growth opportunities come into create demand for those services, those assets, those people. But we don't necessarily have complete control over the rate of growth that we’re seeing.
Yes, I think obviously the things we're doing in terms of fewer train starts, redesigning the network, I said this before in previous quarters, we’re getting out of our own way. We are removing congestion, we're running a more fluid and efficient railroad and that's going to result in fewer.
At the same volume levels, that's going to result in the need for fewer locomotives, railcars and crews and then have a further ripple effect in terms of mechanical assets engineering possibly as well. So in a flat volume environment the things we're doing should result in further productivity improvements and possibly lower headcount requirements going forward.
We also have and we’ve talked about this for a long time. We have work rules in Mexico that are not lined up with the way things work in the U.S and we think there are some opportunities for gains there as well.
The mantra of our PSR efforts as some of you know is service begets growth. So we're hoping that by doing all of these things and improving the way we run our railroad producing a more consistent reliable and resilient network that we are able to take better advantage of some of the growth opportunities that we see and we can put all those assets to use.
But if that growth doesn't occur quickly, then we will see further headcount and labor productivity savings. Now, Jeff or Sameh, have anything to add to that.
Sameh Fahmy
I will just add and then I will let Jeff expand on it. Jeff is spending a lot more time on this on the labor relation front.
But in general, by reducing train delays, which we have been putting a lot of emphasis on and that's how we have been achieving a lot of -- the velocity improvement, it clearly have reduced or improves as an example. By reducing train starts, well again, you have a need for less crews because you have fewer trains.
By reducing switching in yards and enforcing compliance at least with existing plan to begin with and hopefully with the new plan, the new service plan, train design and yard design, we should reduce switching even more. Well, by definition, while going to need also less crews in yards.
So you look at all these and then add to it the mechanical side where we have taken the worst locomotive result of the fleet that had 8 failures per year and 10 failures per year when decent good locomotives have about 2 failures per year. There is a significant reduction in the workload in shops.
So you add all that together, you achieve a lot of staff reduction. Now on the mechanical side, we have gone and done a lot of these staff reductions.
Some of that has been offset by some of the in-sourcing activities because labor in shops makes a lot more sense to be a reporting to the company, not reporting to a vendor and as a result we have had significant savings in money on that. So, yes, the comp can go up, but the money I think one shop alone was $4 million less per year in expense.
We also went ahead and close one shop in Monterey. So in-source San Luis Potosi, but we close Monterey.
So, in the mechanical side we have taken the actions. On the transportation side because of the work rules that Pat mentioned we cannot just arbitrarily go and take action.
We still have four men crews in many places, in many situations and we have to work on that was our colleagues from the union and I think Jeff you can expand on that.
Jeff Songer
Yes, I will just add a couple more talking points. The tale of two countries a bit.
In the U.S., we’ve actually added some crews as part of our efforts. And Sameh talked about Houston, that's -- again that’s just -- it is more fluid today, but it remains a difficult spot.
It's our highest re-crew area of our network. And so, we’ve said it's better to have a little excess resource there and maintain and improve the fluidity versus trying to manage too tight.
Cross-border, we’ve talked about we’ve added several crews in the Laredo and with international crews. On the flipside, in Mexico we do have opportunities.
We’ve talked in the past about still having three and four person train crews. Those are collective agreements.
We're actively working on. However, that's just progressing little slower than we would like.
Keep in mind though that the crews while we don't have furlough ability similar to U.S., if the crews are running trains as we talked about crew starts, they’re collecting just minor benefit from a dollar perspective. I know Mike mentioned, we've had about a $1 million to kind of efficiencies with what the train start reduction.
So not 100% focused on the headcount for Mexico because of those cost factors, but still something we are going to continue to work on through the rest of the year. We believe there's opportunity and hopefully we get some work completed with the Mexico Union.
If not, that will just proceed a little slower as you mentioned to attrition.
Amit Mehrotra
Okay. I appreciate the comprehensive answer.
Thanks everybody.
Operator
The next question comes from Justin Long of Stephens. Please go ahead.
Justin Long
Thanks and good morning.
Pat Ottensmeyer
Good morning.
Justin Long
So just to clarify on the 2021 OR target, does that just include the $55 million of PSR OpEx savings since you spoke about or some number that’s higher than that based on other initiatives you have in the pipeline? And then, Mike, also on the PSR savings, could you talk about what you recognized in first and second quarter as we think about what's incremental in the back half to get you to that $40 million number?
Mike Upchurch
Well, let me start with, I guess, your second question on PSR savings. You know, the amounts that we disclosed here on page 18 are recorded throughout the course of the year.
So we now expect $40 million to be the savings as a result of PSR actions, but that would go up each quarter as you’re getting full benefit for those initial reductions that we identified. So I don't know if that entirely answered your question, the $40 million is kind of a buildup throughout the rest of the year and the exit rate will be higher, which will lead us to $65 million of annual expense savings in 2020.
With respect to your other question, we clearly have a view today of having better expense savings as a result of our first six months of PSR activities, that's why we moved to the lower end of the guidance that we had set for 2021. And remember, not only did we move to the lower end of the guidance, but we're also going to absorb the 80 basis point negative impacts from the loss of the fuel tax credit.
So, obviously, where we’re at in PSR savings today did factor into us improving our outlook out into 2021.
Pat Ottensmeyer
Just to add to that there are lots of moving pieces to operating ratio, including revenue and volume growth, incremental margins, pricing etcetera. And then there are other things that we’ve talked about just recently here and in the presentation about the PSP whiteboarding, we fully expect that that's going to produce another -- possible another wave of improvement here and then some of the labor issues that Jeff and Sameh talked about.
We -- I will say, we don’t have all of that because we’re at the beginning stages of some of that into our guidance. So, we will update guidance as we get better clarity and better information on some of these initiatives, but I don't -- I will state it for you, you didn’t see it this way, Justin, so bear with me here a little bit.
We certainly haven't declared victory on this at this point. We still feel like we're fairly early on and have room to improve further.
Justin Long
Okay, That’s helpful. And then secondly I wanted to ask about intermodal.
Clearly, this is an area where we’ve seen a bit more pressure recently, but when do you think intermodal volumes can start to inflect positively and do you need the truckload market to improve in order for that to happen or can you see intermodal growth from company specific areas, even in a looser truckload capacity environment?
Mike Naatz
I think that we can certainly pursue growth opportunities and we are doing that. For example the cross-border intermodal traffic continues to grow despite overall pressure on the U.S domestic marketplace.
I do think though with respect to the U.S domestic marketplace we’re going to need truck capacity to tighten up in order to improve that situation.
Justin Long
Okay. Thank you.
Operator
The next question comes from Allison Landry of Credit Suisse. Please go ahead.
Samantha Yellen
Hi. This is Sam Yellen on for Allison Landry.
Thanks for taking my question. As you just said, we’ve been seeing some softening in the U.S trucking market.
I was wondering if you could talk about the Mexican truckload market a little. Have you been seeing some more softening in Mexico?
And has that been a drag on Intermodal volumes as well? Thank you.
Mike Naatz
So actually we indicated that we saw improvements in our Lazaro intermodal business. It doesn't seem to be experiencing quite the same challenges that we're experiencing on the U.S domestic side of the marketplace.
We have a little bit of a disadvantage because of the discriminatory pricing actions resulting from the fuel situation. So we will have to address that over time as we look at our contracts in our volume requirements.
Pat Ottensmeyer
I will just add to that and what Mike said earlier about the U.S domestic versus other segments of our market. In our appendix and background information to our presentation we do call out the cross-border in Lazaro intermodal volumes and revenue, so you can see we are seeing growth in those markets.
And the area that we’re seeing the most weakness, which is a big part of our intermodal business, but certainly not all of it is the U.S domestic.
Allison Landry
Thanks. And then also can you give us an update on your initiatives in the Houston area, specifically are you seeing incremental progress in this area and what you view as the biggest risk to reconfiguring traffic flows?
Sameh Fahmy
I can comment on that. This is Sameh.
We have definitely made some operational adjustments to the Houston area. We changed from a method, if you like, of staging trains ahead of that area to try to avoid congestion and all that, a new method where we actually push our trains frankly through it.
And also not worry too much about some of the -- it's difficult to explain on a call, but some of usage of sidings to park trains, which end up using the sidings for parking instead of train meets. The other thing we did is that we also stopped using detours, which were costing us a lot of money but also losing a bit of control on the transit time, time of a train.
A combination of these things have really provided a lot more consistency in the transit time in that area. We see 12 miles, 13 miles per hour for that area when we use to see about 7 or 8 miles per hour.
And it's happening now. Whether there is a peak of volume or reduction in volume, we're getting consistency and that's really important.
And the other nice thing happening now after we did an exercise was our colleagues from UP for 2.5 days around-the-clock watching exactly what’s happening in that area is that not only we change our own method at KCS of avoiding that staging. I’m trying to predict and forecast on all that, but also the transit time itself in the area has improved because of much better dispatching and we see it in the amount of recruits that we endure.
We have a lot of trains now that goes through the whole area without any recruits. So you can get through that 120 mile very important area between Beaumont and Kendleton in six hours which is very good.
It used to be some times 15 and 18 hours. So I think we got this one pretty much under control and now we’re focusing as I said on my slide on the border and we are focusing on yards.
These are the next two big areas where we believe we can make a breakthrough and improve our velocity and our dwell and all the other things.
Allison Landry
Thank you.
Operator
The next question comes from Tom Wadewitz of UBS. Please go ahead.
Tom Wadewitz
Yes, good morning. I wanted to ask you, I guess, a bit about some of the yard work and whiteboarding.
What’s the -- I guess, you didn’t talk about the Monterrey yard, I think that’s something in the past that that has been mentioned as a point of congestion. Maybe just where you add in terms of how the yards are functioning?
Are they kind of cleaned up from the initial work which you’ve done? And primarily what you’re going to look for going forward as a reduction in train starts.
How do we think about the, I guess, what you're aiming for in whiteboarding and what the result maybe on the Mexico operations?
Sameh Fahmy
Let me start and then I will -- Jeff, maybe you can continue here. I think there are two things when we think about yards.
There is the inventory of the yard and I believe that's what you're referring to when you’re talking about Monterrey and that is an inventory that was a 1,000 cars higher back in December than it is today. And a lot of effort has been taken on that, trying to look at cars that that sit for a long time and trying to understand why they sit and working also with customers because a lot of these -- a lot of tracks in the yard are often used for what we call constructive placement which means the car is ready to go the customer, but the customer is not ready to take it.
So we have been working a lot with our key customers in the Monterrey area in a very positive and partnership way and it has brought great fruit. That's one side when we talk about the yard.
The other one is more of the operation and the arrival of trains and the departure of trains and the switching in the yard. And when I talk about San Luis Potosi and a week we spend there, walking in the yard and sitting in the tower and talking to people, okay?
Talking to people tells you much more than anything you can read in annual report. They tell you exactly what the issues are and what the solutions are.
When you look at the yard you find that, well, there was a train held up on the north side, there was a train held on the south side, because some trains came in close to each other not according to plan, but they came close to each other because of train delays. Then you also find that there is a train working in the yard meaning setting out a block and picking up a block, that was not even supposed to work in yard.
That's what I talk about when I say compliance with plan, same that go out of plan because another yard on its own initiative decided to add some cars to a train, but now you have to work on that same train like it was added in Sanchez, an example I gave. Now in San Luis Potosi, you have to take some of these cars out, okay?
And that was not in the plan. So you kind of blocked the movement in the yard, the configurations of these yards sometimes does not make things easy for the work in the yard and blocks through trains as an example.
And that's where that compliance becomes important. Even with the existing plan and making sure also that train arrive when they’re suppose to arrive and not bunched up.
And then in the new plan, the new design, the new whiteboarding, then you work on trying to avoid the work itself in the first place, like get smart about where you do the blocking and of the cars for their destinations and try to do that blocking as deep down in the network as you can at origin. And this way you don’t switch the car and then switch it again and then switch it again in three different yards but you try to reduce that amount of work.
And when you reduce that amount of work, that’s how create capacity without spending any money, because you don’t have to reconfigure the yards now, okay? And you may even change the mission of some yards.
We looked at San Luis Potosi, there is a nice inter-portal yard for intermodal, which is not far from it. And it's a great field, you can build very long tracks without any problem.
While the old San Luis Potosi yard is constrained in the middle of a city, it's a very old yard. So trying when you get a long train in a yard like that, you have to do what we call doubling it, like split it in two tracks and on the departure you have to combine it again.
You go in a new yard like inter-portal and you can build a long track. I think I gave you a very large explanation, but I think we are passionate about this exercise.
Tom Wadewitz
Yes, that -- I mean, that’s great. I guess for the follow-up, it sounds like there -- maybe two different -- a lot of opportunities where one of them is compliance and then one of them is the plan.
What is it that -- what are you doing to really drive the cultural change and the compliance? And then just in terms of the way you implement after the whiteboarding, is this a design across all the terminals in Mexico?
Do you kind of flip the switch in fourth quarter or do you go terminal by -- yard by yard and do things one yard at a time? Thanks.
Jeff Songer
Yes, this is Jeff. I will just expand on that.
You got it right on. So I’m looking at this kind of two phase.
First phase we’ve seen some good work is simply compliance. And compliance to current TSP, as Sameh mentioned in Monterrey yard, that been on the one big differences that we’ve had and just execution, increased focus, increased focus on a daily basis through Brian's team and the use of additional data and dashboards and tactical kind of business process folks really sitting in and dissecting the day-to-day operation of the yard, that's all added benefit as well as working with customers, rightsizing fleets.
Sameh mentioned CT car, so I think that's really the bulk of the work that’s going on thus far is more that day-to-day tactical accountability and execution. As we look forward, again we’ve talked about a lot of the PSR opportunities with TSP is in Mexico because of our -- I will say overly complex intermodal manifest product that’s just how we operate down there.
So looking at the yards and the handlings, reducing handlings, simplifying those products for improved fluidity, but also better product offering for our customers is really kind of the next second half of the year, I really think that's really what we're going to get to more long-term and structural changes to our TSP.
Tom Wadewitz
Thank you.
Operator
The next question comes from Brian Ossenbeck of J.P. Morgan.
Please go ahead.
Brian Ossenbeck
Hey, good morning. Thanks for taking the questions.
Pat Ottensmeyer
Okay.
Brian Ossenbeck
Pat, I want to ask you more about the service. That was your mantra and you're expecting to see some growth here maybe in the second half of the year and in places like grain where service starts to improve.
We can see that in your metrics, but whereas the ones that we still don’t see externally is some sort of up service besides being dwell. So can you tell us about how you’re viewing that internally which you’re talking to with the customers, do you have a plan to roll out a plan compliance or something along those lines?
Pat Ottensmeyer
Yes. We have a number of different metrics that we look at as far as customer service and customer satisfaction, I feel like from the moment that we actually started and announced our PSR initiative and started to work with customers, we saw almost from day one improvement in a lot of our key customer metrics.
And the feedback we’re getting from customers and I will go back in a minute to talk about why kind of picking up on some of the couple, the previous questions about how we are going about this in a really cross functional way, even as we redesign the TSP, which we just talked about the engagement of the sales and marketing team, the communication with customers, all of those things, we’re trying to do this in a really thoughtful and cross functional way, so that we can get the full benefit of the changes we're making and making sure that our customers are very well aware and informed before we make certain changes so that they can make adjustments not only make adjustments to their supply chain strategies and execution, but they become aware of improvements, capacity, availability. Best example is grain.
One of the reasons we think that we’re going to see some improvement in our grain and Ag/Min business is improving cycle times has allowed us to do the same amount of volume with certain customers with fewer assets. Our first choice when we look at the consequence of that is not to immediately go cut up cars and go to the scrap yard is to say can we fill that capacity with additional volume and go out and try to sell that capacity, take advantage of our improved performance and shorten cycle times.
And as you can expect, it's not like customers immediately have sales on the shelf ready to go, so it takes a while for that cycle to be completed. In other words, customers are going to want to see that these improvements in cycle times and operational metrics is sustainable for some period of time before they go out and put their neck on the line and sell that capacity to their customers.
But that's always the first choice. And I think we talk about this with a number of you, one of the barometers that I look at, it's kind of a high-level metric, but it's not terribly sophisticated that the number of unit grain shuttles that we're able to run across our network and for me that gives me an indication of the health and fluidity of our entire network because a lot of those shuttles run from Eastern Illinois or Kansas City, all the way down into the heart of Mexico.
So there are five or six different traps. We run into trouble around Shreveport, they get trapped.
We run in trouble around Houston or at the bridge or at Monterey yard even though these trains aren't being handled in Monterey. So there are a number of different traps.
But when we look at the total number of grain trains that we're able to run in any given month or whatever period you choose, that’s just sort of an indication of the total overall health of our entire network and in the last couple of months we’ve had record levels of performance even to some extent that have been affected by the flooding at the points of origin in Illinois and Missouri. So for example, we ran 42 grain shuttles in the month of May and that freed up a lot of capacity.
So our first question we ask ourselves when we do that is can we take that capacity? Can we take those cars, crews and locomotive and sell them and fill it up?
So now I’m probably wandered away from your core question, but that that’s kind of an indication of how we’re thinking about taking advantage of the benefits that we're producing here and getting back to that service and growth orientation.
Brian Ossenbeck
All right. That’s great.
Thanks, Pat. I appreciate the comprehensive response there.
Pat Ottensmeyer
We got way too comprehensive answers here so …
Brian Ossenbeck
That was a good one. I appreciate it.
The shorter question follow-up for Mike and Mike, we’ve seen volumes come down from 3% to 4% to flat to possibly down, but the revenue guidance stayed the same. I can understand the mixed shift and some of that accelerating in the back half of the year, but as I look at the change in IEPS and more revenue, fuel surcharge revenue going from the top line, it looks might -- that might be 2 percentage points.
I don’t know if you could help put some context around there. How much of revenue guide staying the same is really from the mix shift which we’ve seen through the first half of the year and how much from the change and how fuel surcharge is being pushed through the geography of the financial statement?
Thank you.
Mike Naatz
So I think revenue is going to be driven by changes in core pricing. Changes in fuel related expenses and then, of course, changes in volume.
With respect to the volume, one other things to keep in mind is that the intermodal volume is disproportionate compared to the revenue. So for example, if I were looking at my volume numbers, you might see that 45% of our volumes are related to intermodal business but only 15% of our revenues are related to intermodal business.
So there's a very disproportionate impact on volume and revenue. And I think that's what you're seeing the revenue continues to be higher even though we decreased our overall volume forecast.
Does that help?
Brian Ossenbeck
Yes, and just maybe to clarify, the fuel surcharge in the slide that Mike Upchurch had about how that impacts the top line. I don’t know if that was another kind of factor to consider here?
Mike Upchurch
Yes, that’s what we try to do with the last column on slide 17 is to give you an estimate of what the future quarters might look like. Obviously, we're not talking about what the quarterly total revenues are, so I’m going to stay away from percentages, but I’m sure you can do those calculations with your model, we are kind of looking at that $12.8 million.
Brian Ossenbeck
Okay. Thanks.
That helps. I appreciate it.
Operator
The next question comes from Ken Hoexter of Bank of America Merrill Lynch. Please go ahead.
Ken Hoexter
Hey, great. Good morning and solid job on the improving PSR results.
Great to see. Just on -- Mike, maybe on the $51 million charge, maybe you can kind of delve into that, talk to us what that’s for, given that you had a two quarters in a row now.
You mentioned they may continue -- this is, I guess, not really typical for rails adopting PSR to take such charges, but I know historically you’ve been a little different given your leases and some of the things you’ve been trapped in. So maybe talk about what that generates for you and what that’s contributing from?
Mike Upchurch
Yes, Ken, the charge in the second quarter was predominantly related to locomotives and freight cars that we no longer need in our business, so we will be disposing those either selling or scrapping those. I don't think the fact that we’ve taken two charges is unusual.
And I think there could be another one coming, sometime in the third or fourth quarter. We will continue to make decisions about equipment.
We will continue to be making decisions about number of resources we need to move trains as we go through the train consolidation and the service plan redesign.
Ken Hoexter
I guess just trying to clarify that others have kind of just parked locomotives, is this getting rid of leases that you’ve had or is -- I just want to understand what the charge as opposed to just parking appointment [ph]?
Mike Upchurch
It's writing off the net book value of the equipment that we're no longer going to need.
Pat Ottensmeyer
And I think the sequence that we've gone through, Ken, in the case locomotives, again we take an action, we reduce train starts, we do something that we think is going to free up capacity and reduce congestion. We parked the assets for a period of time and then once we sort of become convinced that change is going to stick, we sell them, scrap them, get rid of them, they’re gone.
And we've sold quite a few of the locomotives and the impairment charge I think relates to those assets that we have maybe gone through that cycle of parking for a while and then conclude that they’re no longer necessary and actually get rid of them.
Ken Hoexter
Okay. I don’t know I would presume that helps your DNA through there.
Just a follow-up on the train length, I guess, Sameh, you talked about maybe lengthening train, target of 6,000 feet, is there something you need to invest more on sidings in Mexico or the U.S side? Is that something that that's your goal because that's what your network is designed for, maybe talk about the potential that’s increased that train length size?
Sameh Fahmy
From a -- Ken, from a siding lengths, we are okay, but we still have a lot of very short trains in Mexico. In the U.S the train lengths is very decent.
Grain trains, coal trains, manifest trains are very nice and quite long and the tonnage is heavy, which is the way you want it to be. In Mexico, we have some very, very short trains and we see them every day, particularly at the endpoints of the network like Lazaro and Toluca and some of these locations because the way our structure of trains is, we drop-off a lot of blocks until we get to the endpoint and then if the train becomes short.
And then when it departs in the opposite direction, it starts short and then it picks up a lot of traffic. So in the new design we are going to look at this to see how can we increase the length even at the endpoints of the network.
So it's more of a design issue, Ken, than a sidings issue.
Mike Naatz
Ken, I would add just a little color. Our infrastructure is in good shape for that.
I think as we started this process 6,000 was the target. And as we get into more of the TSP definition and redesign and whiteboarding as Sameh mentioned here in the second half of the year, I think your targets for next year will be above that.
So I think that was the starting point for us to say, here is opportunities we see in these commodities, I will say it again with Mexico intermodal manifest products. We know we have opportunity.
As we get through the whiteboard exercise I'll be very surprised if we don't have more opportunities than that, but the infrastructure we have is fine to handle that.
Ken Hoexter
Wonderful. I appreciate the time.
I feel like I’m giving you guys short answers. Thank you.
Operator
The next question comes from Scott Group of Wolfe research. Please go ahead.
Scott Group
Hey, thanks. Good morning, guys.
Almost good afternoon. The guide -- the long-term guide sort of implies around a 150 basis points on margin improvement a year.
Do you think we get that in the second half, Mike or is the fuel headwinds sort of too much to overcome? And if we don't get that for 150 for '19, do we have visibility at this point to over 150 for next year?
Mike Upchurch
Well, I think if you're asking can the back half of 2019 be better than the first half? Yes, we think so.
With respect to our longer term guidance, you can kind of do the math getting into low 60s. Yes, the 80 basis point headwind on the IEPS and then you can do the math on what that requires on an annual basis, but if you think about what we've done through normal operation this year in the first six months, we are about a 160 basis points better than a year-ago.
Scott Group
Let me ask this differently. And is it fair that we are changing the guide just six months in that we wouldn't be sort of improving the OR guidance if it was going to be about a 2021 step function.
We got to have some confidence about more of a step function coming sooner than later, is that fair, Mike?
Mike Upchurch
Well, we wouldn’t put the guidance out if we didn’t have some confidence in that. I maybe take you back to our January discussion around how we set this.
We were barely a month into the process and so shouldn’t be very surprising that we are seeing bigger and bigger savings opportunities as we move through this. But we do still have things like PTC headwinds that we're trying to overcome in 2019 that won't be there in 2020 and 2021.
So all along, our guidance had been the first year improvement may not be as good as the second and third year improvement.
Scott Group
Okay, makes sense to me. Just real quick.
I see, Sameh, you've got a slide, we’ve taken 7% of the cars out, we’ve line of sight to another 3%. The locos you've taken 12% out, what do you have line of sight to there for incremental locos?
Sameh Fahmy
The -- it will depend a lot on the revenue. I mean we -- like Pat mentioned a number of times, we are hoping that the improved service, which is palpable and our customer are telling us that they see the improvement in service, that will generate more business.
As a matter of fact, I think on the commercial call, Mike, this week or last week we won back the customer. The customer came back to us because of improved service.
So if we improve service that will have an impact on how many -- how we will size our assets. Like Pat said, we would love to keep the remaining assets that we have and use some for absorbing more revenue.
Pat Ottensmeyer
As you know, it's easier to get line of sight on cars because that tends to relate to specific customers or commodities or aspects of our business. Locomotives reduced across the entire franchising and entire portfolio, so that's why we're being a little more cautious about guiding to any further locomotive reductions.
Sameh Fahmy
And on the car side, we still have a lot of system intermodal cars that we have in storage. I mean, we would love to get more business and use them for that.
Scott Group
Okay. Thanks for the time, guys.
Pat Ottensmeyer
Okay.
Operator
The next question comes from Bascome Majors of Susquehanna. Please go ahead.
Bascome Majors
Yes, thanks. You talked a little bit about attrition earlier.
I don't think you gave an attrition rate for the business right now. Can you talk about the kind of inherent attrition rate in the U.S and Mexico and where it's tracking today?
Mike Upchurch
It's kind of 6%, 7% normal attrition.
Bascome Majors
Is that the same on both sides of the border?
Mike Upchurch
That I don't know. I just know overall.
Bascome Majors
All right. And on CapEx, Mike you talked about 18% of revenues feeling like the right level when we get beyond this year.
Historically, you've done a lot of chunky coinvestment projects to grow the business with some of your customers. Would that include those kinds of projects?
And regardless of that answer, do you have a line of sight into anything like that coming up next year or the year after? Thank you.
Mike Naatz
Yes, sitting here today, we don't have big projects like the Sasol if that's what you're referring to. If those opportunities come our way, we will take a look at them and see whether it makes sense to invest.
But no, there's nothing of that scale built into the 2020, 2021 CapEx to revenue guidance that we've provided.
Bascome Majors
Thank you.
Operator
The next question comes from Brandon Oglenski of Barclays. Please go ahead.
David Zazula
Good morning. This is David Zazula on for Brandon.
Thanks for taking my question. Maybe for Mike, you guys had an interchange partner go on media yesterday ordering passage of the USMCA.
Just wondering if you could comment a little bit on how you think the potential added foreign investment might impact flows on the railroad and how that could affect your operational and capital planning?
Pat Ottensmeyer
I will take that, and -- yes, we are very much in support of USMCA. It's something that's very important to our company, our customers, our communities.
And I've been quite vocal in both Washington and Mexico City supporting the passage of that trade agreement. In spite of the uncertainty of USMCA and for the last couple of years, look back over the last several quarters, we've seen just exceptional growth in our cross-border volumes.
We do feel that the cloud of uncertainty has curtailed investments even though some of the data that’s published about direct foreign investment in Mexico continues to increase. We think that there are investments that maybe companies have been considering that they've held off or delayed because of concern about the passage of the trade agreement.
I've gotten out of the business of predicting what Congress will do a long time ago, feel like there is no reason that this trade agreements should not be approved. It's certainly better than NAFTA and it's way better than no trade agreement and we will continue to be involved in trying to influence the outcome in the direction of the debate and hope to have some resolution on that in the next, probably not until after the recess in the U.S.
Congress, but hopefully by the end of the year.
David Zazula
Great. Thanks.
Operator
This concludes our question-and-answer session. I would like to turn the conference back over to Pat Ottensmeyer for closing remarks.
Pat Ottensmeyer
Okay. Well we will -- I will wrap up very quickly.
Thank you for your time and attention and when we do our next shareholder and investor survey, please give us high marks for comprehensiveness of answers. Thank you very much and we will be back with you in 90 days.
Operator
The conference has now concluded. Thank you for attending today’s presentation.
You may now disconnect.