聯合太平洋集團 (UNP) 2018 Q3 法說會逐字稿

Greetings, and welcome to the Union Pacific Third Quarter Earnings Call.

(Operator Instructions) As a reminder, this conference is being recorded, and the slides for today's presentation are available on Union Pacific's website.

It is now my pleasure to introduce your host, Mr. Lance Fritz, Chairman, President and CEO, for Union Pacific.

Thank you.

Mr. Fritz, you may now begin.

Thank you, and good morning, everybody, and welcome to Union Pacific's third quarter earnings conference call.

With me here today in Omaha are Kenny Rocker, Executive Vice President of Marketing and Sales; Tom Lischer, Executive Vice President of Operations; and Rob Knight, our Chief Financial Officer.

This morning, Union Pacific is reporting net income of $1.6 billion for the third quarter of 2018 or $2.15 a share.

This represents an increase of 33% in net income and 43% in earnings per share when compared to 2017.

This was an all-time quarterly record for Union Pacific, even without the benefit from corporate tax reform.

Total volume increased 6% in the quarter compared to 2017.

Premium and industrial carloadings both increased 9%, while Agricultural Products grew 2% and energy volumes were down 2%.

The quarterly operating ratio came in at 61.7%, which was flat with the third quarter of 2017.

Higher fuel prices had a 0.3 negative impact on the operating ratio.

Strong top line growth was offset by an increase in volume-related costs, higher spending due to some lingering network inefficiencies and other cost hurdles.

While we reported solid financial results, we did not make the service and productivity gains that we had expected during the quarter.

However, we believe that Unified Plan 2020, along with other G55 + 0 initiatives and recent changes to our leadership team position us well to start driving larger service and operational improvements going forward.

We launched Unified Plan 2020 on October 1 in our mid-American quarter with a goal of creating more streamlined operations on the eastern 1/3 of our network.

While early, I am pleased with the initial results, as we've seen improvement in several key performance indicators on our network.

We've also made a number of other changes to drive near-term productivity savings that you will hear about today from the rest of the team.

Our entire Union Pacific team is fully engaged in the implementation of Unified Plan 2020 and our pursuit of running a highly reliable, more efficient network.

There is much more to come as we continue to rollout Unified Plan 2020 across our network, and I'm excited about the opportunities it's going to create for both our customers and our shareholders.

I'm confident we have the right people and the right plans in place to improve our operations, to provide more reliable service for our customers and achieve industry-leading financial performance.

The team will give you more of the details on the third quarter and Unified Plan 2020, starting with Kenny.

Thank you, Lance, and good morning.

For the third quarter, our volume was up 6%, driven by strength in our Premium, Industrial and Agricultural business group, with a partial offset in Energy.

We generated positive net core pricing of 1.75% in the quarter with continued pricing pressure in our coal and international Intermodal markets.

The increase in volume and a 4% improvement in average revenue per car drove a 10% increase in freight revenue.

So now, let's take a closer look at the performance in each business group.

Ag products revenue was up 6% on a 2% increase in volume and a 4% increase in average revenue per car.

Grain carloads were up 2%, driven by strong exports, predominantly shipping to Mexico, coupled with increased domestic demand resulting from lower corn prices.

These gains were partially offset by persistent weakness in wheat due to reduced U.S. competitiveness and the world market export.

Grain products carloads were up 6%, driven by sustained demand for ethanol and other biofuels.

This renewable fuel strength, coupled with increased meat production, also drove an increase in shipments for animal protein.

Fertilizer carloads were up 5% due primarily to continued strength in export potash.

Energy revenue increased 1% for the third quarter as a 2% decrease in volume was offset by a 2% increase in average revenue per car.

Coal and coke volume was down 3%, primarily driven by a contract loss and retirements, coupled with lower natural gas prices, which fell 3% versus the third quarter 2017.

Sand carloads were down 23% due to the impact of regional sand and market decline in the Permian Basin.

Furthermore, favorable crude oil price spreads drove an increase in crude oil shipments, which was the primary driver for the 40% increase in petroleum, LPG and renewable carloads for the quarter.

Industrial revenue was up 13% on a 9% increase in volume and a 3% increase in average revenue per car during the quarter.

Construction carloads increased 10%, primarily driven by strong market demand for rock and cement and favorable year-over-year comps due to Hurricane Harvey that impacted the Texas Gulf in the third quarter of 2017.

Likewise, plastics carloads increased 14% due to the same favorable third quarter comps resulting from the hurricane and strength in polyethylene shipments with increased production.

Industrial chemicals volume increased 14% due to the continued industrial production growth.

Premium revenue was up 18% with a 9% increase in volume and a 9% increase in average revenue per car.

Domestic Intermodal volume increased 7%, driven by continued demand for tight truck capacity and strength in parcel and LTL shipments.

Auto parts volume growth was driven by over-the-road conversions and production growth at key locations.

International Intermodal volume was up 12%, as new ocean carrier business continued in the third quarter, coupled with strong import and export shipments.

Finished vehicle shipments were up 8% due to strong truck and SUV sales, increased production at UP-served plants and growth with new customer wins.

Although the SAAR was down 1% for the quarter, the light truck segment was up 7%.

Looking ahead for the rest of 2018, our ag products growth continues to face uncertainty in the export grain markets from foreign tariffs.

However, we are seeing some positive indications in the market due to crop issues in South America and other countries, which have made U.S. grain more competitive in the world markets.

We anticipate continued strength in biodiesel fuel and renewable diesel fuel shipments due to an increase in market demand for renewable fuels.

We also expect tight truck capacity, combined with the value of rail, to continue our penetration growth across multiple segments of our food and beverage business.

For Energy, we expect favorable crude oil price spreads to drive positive results for petroleum products.

But tough year-over-year frac sand comparisons, coupled with local sand supply and softer market conditions, will impact sand volumes.

We also expect Coal to experience continued headwinds for the remainder of the year, and as always for Coal, weather conditions will be a key factor for demand.

For Industrial, we anticipate upside in plastics as production rates increase.

Metal shipments are expected to grow due to strong construction and energy markets coupled with tight truck capacity.

In addition, we anticipate continued strength in Industrial production, which drives growth in several commodities.

For Premium, over-the-road conversions from continued tight truck capacity will present new opportunities for domestic Intermodal and auto parts growth.

Despite challenges within the international Intermodal market, we anticipate growth year-over-year for the remainder of the year, resulting from new business wins.

The U.S. light vehicle sales forecast for 2018 is 17 million units, down 1% from 2017.

However, production shifts and new import business will create some opportunity to offset the weaker market demand.

So before I turn it over to Tom for his operations update, I'd like to share a few observations on the progress we're making commercially as we put Unified Plan 2020 into action.

We are working very closely with customers to lower our car inventory levels and remove excess cars from the network.

This includes both private and system equipment.

In the near term, we plan to make adjustments to our accessorial charges to incentivize greater car and asset utilization across both our carload and unit train networks.

More importantly, we are and we will continue to critically evaluate every carload on our network to determine if it fits into our operating strategy at the right margin.

In closing, I'm really proud of our commercial team that's communicating with our customers at every turn and, in some cases, having difficult conversations with them.

We are proactively engaging our customers so that ultimately, we can provide them with a safe, reliable and mutually efficient service product.

And with that, I'll turn it over to Tom.

Thank you, Kenny, and good morning.

I'll get started with a quick update on our safety performance for the first 3 quarters.

Our reportable injury index was 0.77, an improvement of 1% compared to last year.

The reportable rate of our rail equipment incidents, or derailments, was 3.20, an increase of 8%.

In Public Safety, our grade crossing incident rate was 2.66, an increase of 6%.

I want to note that although we have a tremendous amount of transition and change happening with the implementation of Unified Plan 2020, safety is still job 1. Our goal is that all our employees return home safely each and every day.

That goal has not, and will not, change.

So that's a quick update.

Now let's turn to the changes we are making as we implement UP 2020.

As I noted in last month's conference call, Unified Plan 2020 is fundamentally an implementation of Precision Scheduled Railroading principles in a manner that fits our network and the needs of our customers.

It is a change from operations that shifts from focus on moving trains to moving cars.

The planned changes are designed to increase car velocity, resulting in reduced locomotive and car dwell.

In addition to improved equipment cycle times, the plan is also designed to better balance our resources across the network.

The outcome is a more simplified network that improves reliability for our customers while reducing operating costs and investment requirements.

We're about 1 month into implementation of the Mid-America Corridor.

To level set everyone, the Mid-America Corridor, as the map on our slide indicates, encompasses large north-south traffic flows on the eastern end of our railroad and about 50% of our daily carloads touch this corridor.

Our progress thus far.

In total, we anticipate more than 150 design changes to our transportation plan this quarter alone, and we are well on our way to implementing those T-Plan changes.

I am pleased and encouraged with the initial results.

In fact, let me give you a few concrete examples that represent the types of changes we are making.

For 1 customer on our Little Rock service unit, we changed where we build blocks of cars and how we move those blocks to destination.

As a result, we've significantly reduced the freight car dwell and their cars are now arriving at destination up to 3 days sooner.

Another customer on the North Platte service unit now pre-blocks cars at their own facility.

As a result of this change, the cars are now bypassing UP's origin switching yard altogether, eliminating 24 to 36 hours of terminal dwell and providing the customer an overall faster transit time to destination.

In other instances, we have adjusted train schedule frequency to better balance our resources and smooth customer demand.

These adjustments improve asset utilization by establishing a more precise schedule across our network, reducing the number of locomotives and crew starts required to manage the business.

I want to point out that in all of these cases, we work closely with the customers to end up with a win-win solution.

We were able to improve operational efficiency for Union Pacific and service reliability for the customer.

And although we have made dozens of these types of changes in a short period of time, we are just getting started.

There is more to come.

As we implement the UP 2020 operating model, we will begin to focus on new performance indicators or KPIs.

These KPIs will align with the operating goals we are trying to achieve and our financial targets.

We are in the process of evaluating some of the appropriate indicators at different levels of the organization and for the operating functions.

However, at a high level, the measures on this slide are appropriate KPIs to gauge our progress as we implement UP 2020.

For each measure, we are showing a pre-unified plan baseline for September, the current value and our goal for the end of 2019.

I want to note that the 2019 goals are not the end state.

These are interim targets we -- as we implement the Unified Plan across our network next year and as we continue supply and refine PSR principles beyond 2019.

To begin with, freight car velocity is measured in daily miles per car and is consistent with focus moving from cars versus moving trains.

Operating inventory is a subset of the weekly number we publish with the AAR, but excludes cars in storage and cars placed at customer facilities.

Measuring operating inventory is appropriate as we -- it will decrease as we successfully increase car velocity.

Cars per carload brings together car inventory and volume, acknowledging that inventory levels will fluctuate with seasonality and as we grow the business.

Locomotive productivity is a measure in gross ton-miles per horsepower day.

Stated another way, it is the number of gross ton-miles that we move each day for each unit of horsepower in the active fleet.

This is an all-inclusive locomotive metric, including shop time.

Car plan compliance is a measure of how well we are serving our customer compared to our service schedule.

And workforce productivity measures daily car miles per full-time employee and will give us a good indication of how efficiently our employees are working.

As I stated earlier, this, by no means, is an exhaustive list as there are number of other measurements we monitor on a daily basis.

Following the announcement of Unified Plan 2020 last month, I have had some -- there has been some speculation of what we're doing is a light version of PSR or that UP is not fully committed to making the changes necessary to achieve PSR benefits.

I can assure you that is not the case.

We have a -- we do have a planned phase approach to implementation, and we are working to accelerate the changes when and where we can.

We're also fully committed to basic tenets of PSR, including increasing car velocity, minimizing car dwell, classification event reduction and locomotive and crew requirements.

This will be achieved through simplifying the rail network and better balancing resources.

While our approach may be different, the fundamental PSR operating principles are the same.

Our implementation -- as our implementation progresses, we expect to realize benefits other railroads have achieved, including service reliability, labor productivity, better asset utilization and reduced fuel consumption.

And we have started to see some positive results.

Since the beginning of August, we have removed over 625 locomotives from the active fleet and we have line of sight for another 150 locomotives to take out of the network by year's end.

On the previous slide, our car operating inventory has come down 6,000 cars since September, and we currently have action plans to reduce this inventory by about 10,000 more in the near term.

On the TE&Y front, our September workforce was down 2% versus August on flat volume.

While I realize it's only 1 month, the numbers are moving in the correct direction.

We are encouraged by the initial results, considering the impact of most of the 150-plus service changes, that I spoke of earlier, are just beginning to be felt here in October.

I'd like to highlight some of the changes we're making in our operational organization as a part of Unified Plan 2020.

Just this week, we announced the consolidation of our operating regions from 3 to 2. As a part of the regional consolidation, we are also reducing our operating and service units from 17 down to 12.

These changes will better align our management structure and decision-making processes with the new operating model providing more speed and agility as we implement the Unified Plan.

In addition, we are closing our locomotive repair shop in South Morrill, Nebraska, in January '19.

As we begin to implement the network designed changes, car flows and traffic patterns will shift.

As a result, we are working through a terminal rationalization process.

Further in an effort to streamline 2-way communications with our customers and drive faster resolution of service issues, we are moving our customer care and support function, including car management, for marketing and sales to the operating department.

This realignment will be crucial in our initiative to reduce operating car inventory.

Additionally, we're in the process of consolidating and restructuring our engineering functions to drive better accountability and increased productivity.

Finally, we are making some near-term reductions in our management workforce beyond reductions related to the consolidations and closures I described.

These reductions will occur in the fourth quarter of this year and we expect additional workforce reductions as the Unified Plan is implemented into next year.

Wrapping up, we're building a new culture here at Union Pacific that will enable successful implementation of the Unified Plan 2020.

We're off to a great start, and I am confident that we will complete the full year implementation of our plan late next year.

In fact, we are about to begin the next phase of implementation on our Sunset Route, which is earlier than we anticipated.

As we make progress across our network in coming months, the result will be a simplified network that operates more safely with greater reliability and efficiency.

I am excited about the changes taking place, and I am encouraged about the early results.

Change is never easy and there will be challenges along the way, but I am confident that our talented and motivated team is up to the task.

With that, I'll turn it over to Rob.

Thanks, Tom, and good morning.

Let's start with a recap of our third quarter results.

Operating revenue was $5.9 billion in the quarter, up 10% versus last year.

Positive core price, increased fuel surcharge revenue and a 6% increase in volume were the primary drivers of revenue growth for the quarter.

Operating expense totaled $3.7 billion, up 10% from 2017.

Operating income totaled $2.3 billion, a 9% increase from last year.

Below the line, other income was $48 million compared to $90 million in 2017.

And as a reminder, third quarter 2017 results included a large land sale and the favorable settlement of a litigation matter.

Interest expense of $241 million was up 34% compared to the previous year.

This reflects the impact of higher debt -- total debt balance, partially offset by a lower effective interest rate.

Income tax expense decreased 39% to $483 million.

The decrease was primarily driven by a lower tax rate as a result of the corporate tax reform, partially offset by higher pretax earnings.

Our effective tax rate for the third quarter was 23.3%.

For the fourth quarter of this year, we expect our effective tax rate to be in the mid-23% range.

And going forward, we now expect that our normalized tax rate in the future quarters will average around 24%.

Net income totaled $1.6 billion, up 33% versus last year, while the outstanding share balance declined 7% as a result of our continued share repurchase activity.

These results combined to produce an all-time quarterly record earnings per share of $2.15.

Our operating ratio of 61.7% was flat with the third quarter of last year.

And as a reminder, last year's workforce reduction program and Hurricane Harvey had an unfavorable impact of 1.1 points on our third quarter 2017 operating ratio after adjusting for the change in pension accounting.

The combined impact of fuel price and our fuel surcharge lag had a 0.3 point negative impact on the operating ratio in the quarter compared to 2017.

Freight revenue of $5.6 billion was up 10% versus last year.

Fuel surcharge revenue totaled $482 million, up $255 million when compared to 2017 and up $70 million versus the second quarter of this year.

The negative business mix impact on freight revenue for the third quarter was 2 full points.

Decreased sand volumes and an increase in lower average revenue per car Intermodal shipments drove the mix change in the quarter.

Core price was 1.75% in the third quarter.

Pricing continues to be a challenge in our Coal and International Intermodal markets.

And excluding Coal and International Intermodal, core price was 2.75% in the quarter.

For the full year, we still expect the total dollars that we generate from our pricing actions to well exceed our rail inflation costs.

Turning now to the operating expense.

Slide 21 provides a summary of our operating expenses for the quarter.

Comp and benefits expense increased 2% to $1.3 billion versus '17.

The increase was driven primarily by volume-related costs, network inefficiencies and increased TE&Y training expenses, partially offset by lower management costs as a result of our workforce reduction program that we initiated last year.

For the full year, we still expect labor and overall inflation to be under 2%.

Total workforce levels were up about 1% in the third quarter versus last year.

Employees not associated with capital projects were up approximately 2%.

The increase was driven by our TE&Y workforce, which was up 8% due to higher carload volume and more employees in the training pipeline.

Partially offsetting the increase in our TE&Y workforce was a 6% reduction in management employees and employees performing capital project work.

Fuel expense totaled $659 million, up 46% when compared to last year.

Higher diesel fuel prices and a 5% increase in gross ton-miles were the primary drivers of the increase in fuel expense for the quarter.

Compared to the third quarter of last year, our average fuel price increased 34% to $2.38 per gallon.

Our fuel consumption rate also increased during the quarter by about 4%.

While there was some adverse impact from mix, the predominant driver of the increased C rate with the service-related challenges that we experienced.

Purchase services and materials expense increased 3% to $632 million.

The increase was primarily driven by volume-related costs, higher prices for purchase transportation services and increased locomotive and freight car repair costs.

Turning to Slide 22.

Depreciation expense was $547 million, up 4% compared to 2017.

The increase is primarily driven by a higher depreciable asset base.

For the full year 2018, we estimate that depreciation expense will increase about 4%.

Moving to equipment and other rents.

This expense totaled $272 million in the quarter, which is down 1% when compared to 2017.

The decrease was primarily driven by lower freight car and locomotive lease expense, offset by increased volume-related and network congestion costs.

Higher equity income in 2018 also contributed to this favorable year-over-year variance.

Other expenses came in at $287 million, up 25% versus last year.

The primary drivers were an increase in environmental costs as well as higher state and local taxes.

For the full year 2018, we expect other expense to be up about 10% compared to 2017.

Productivity savings yielded from our G55 + 0 initiatives were largely offset by additional costs as a result of continued operational challenges.

The impact of these operational challenges totaled just under $50 million in the quarter, which is down from the $65 million that we reported in the second quarter.

The additional costs were primarily in the compensation and benefits cost category, although purchase services, fuel and equipment rents were also impacted.

Looking forward, we expect our G55 + 0 initiatives, including Unified Plan 2020, will not only eliminate these failure costs but will put us back on track to achieve significant productivity savings in 2019 and beyond.

Looking at our cash flow.

Cash from operations for the first 3 quarters totaled $6.4 billion, up about 18% when compared to last year, due primarily to higher net income.

Taking a look at adjusted debt levels.

The all-in adjusted debt balance totaled $24.8 billion at the end of the third quarter, up about $5.3 billion since year-end 2017.

This includes the $6 billion debt offering that we concluded in early June, partially offset by repayment of debt maturities.

We finished the third quarter with an adjusted debt-to-EBITDA ratio of around 2.3x.

And as we mentioned at our Investor Day, our new target for debt-to-EBITDA is up to 2.7x, which we will achieve over time.

Dividend payments for the first 3 quarters totaled $1.7 billion, up from $1.5 billion in 2017.

This includes the effect of 3 10% dividend increases over the past year, including the fourth quarter of 2017 and the first and third quarters of this year.

We repurchased a total of 44.7 million shares during the first 3 quarters of 2018, including 2.2 million shares in the third quarter.

This includes the total -- the initial 19.9 million shares we received as part of a $3.6 billion accelerated share repurchase program that we initiated in June.

We expect to receive additional shares under the terms of the ASR as the program reaches completion before the end of this year.

Between dividend payments and share repurchases, we returned $8.7 billion to our shareholders in the first 3 quarters of this year.

Looking to the remainder of the year.

We expect solid volume growth to continue in the fourth quarter.

For the full year, we expect volume to be up in the low to mid-single-digit range versus 2017.

We will yield pricing dollars well in excess of our inflation costs.

With respect to capital investments, we expect full year 2018 spending to be around $3.2 billion or about $100 million less than our previously announced $3.3 billion plan.

Previously, we guided to an improvement in our full year operating ratio compared to 2017.

While we believe this goal is still achievable, we're starting to see some risks.

Implementation of Unified Plan 2020 in the Mid-American Corridor is on track.

However, we continue to see elevated levels of spending, even as our service metrics slowly improve.

While volume growth is still strong, we're not seeing the normal seasonal ramp-up in our export grain business due to tariffs and foreign competition.

We therefore challenge the entire organization to accelerate productivity gains by ramping up our G55 + 0 initiatives and these actions are now starting to gain traction.

And as Tom just stated, we've removed over 625 locomotives and over 6,000 cars from our network since August 1. We're simplifying our operating leadership structure by eliminating 1 of 3 regions and 5 of our 17 service units.

We've announced the closure of our South Morrill locomotive shop, and we are working through a terminal rationalization process.

We're taking steps to reduce our management workforce with approximately 475 positions being eliminated by the end of this year.

In addition, another 200 contract positions will be eliminated.

This is the first of what will likely be additional reductions as we continue to drive productivity within our management workforce.

While we are confident the actions that we're taking will produce near-term results, the timing of these initiatives may not support an improved operating ratio performance in 2018.

And as we look ahead to next year, we expect to get back on track making significant progress reducing our operating ratio and driving toward a 60% OR by 2020.

While we expect positive volume growth and core pricing increases to be major contributors to our operating ratio improvement, productivity gains will play a key role.

Unified Plan 2020 will be implemented across our network with the first phase expected to be completed by the end of this year and the following phases by the end of 2019.

As we make progress in the coming months, we will see lower costs.

Although we haven't finalized our financial targets for 2019 and many aspects of Unified Plan 2020 are still being worked out, it is not unreasonable to expect that we should yield at least $500 million of productivity in 2019.

We will see this productivity in the form of lower compensation costs as well as savings resulted from operating smaller locomotive and freight car fleets, including equipment rents and purchase services, materials and supplies and fuel.

While we have not yet finalized our capital spending needs for 2019 at this time, we do expect investment dollars to be less than 15% of revenue.

We will continue to provide periodic updates to our financial goals and guidance as we make further progress implementing our G55 + 0 initiatives, including our Unified Plan 2020.

So with that, I'll turn it back over to Lance.

Thank you, Rob.

As we discussed today, we delivered record third quarter earnings per share, driven by strong volumes and solid top line revenue growth, while we recognize that opportunities still exist to improve our network performance, we are encouraged by the progress that has been made so far.

Furthermore, I'm pleased with the strength of the economy and the positive impact on most of our business segments.

Looking ahead, I'm confident that the recent progress we've made on our Unified Plan 2020 will accelerate in the near term.

As we move forward with the implementation along with other G55 + 0 initiatives, we will regain our productivity momentum and improve the value proposition for all 4 of our stakeholders.

With that, let's open up the line for your questions.

(Operator Instructions) Our first question is coming from the line of Amit Mehrotra with Deutsche Bank.

Just a follow-up on the $500 million of productivity next year.

So that translates to maybe a little bit more than 200 basis points of margin, given our revenue forecast at least.

I'm not sure -- it's always been hard to, kind of, translate productivity numbers for the rails into maybe a net savings numbers and more specifically, how do you contemplate the risks around the frac sand outlook?

If silica is ramping up in Basin Permian mines, I think they've taken off 12 million tons and maybe another 10 million to 15 million tons to go.

So just given the contribution margins of those carloads, what's the impact to productivity savings next year if those volumes basically go away?

Amit, as I said -- this is Rob.

As I said in my comments, we've not finalized our 2019 and all the issues and challenges and questions that you've asked are clearly in the process of us working through, and we're not prepared to give guidance on all those items.

All our calling out here is that we are feeling confident in the early innings of our traction that we're gaining with our G55 + 0 initiatives and more specifically, our Unified Plan 2020.

And as we look at that, we know there's going to be -- we think there's going to be a positive volume environment, but exactly what that volume looks like and what the mix looks like and how sand or other commodities turn out remains to be seen at this point.

But despite all that -- all those moving parts, we think we'll be able to drive at least $500 million of productivity from the initiatives that we have well underway right now.

Okay.

Maybe just one follow-up on the CapEx, because you did provide maybe a 2019 framework for that, at least under 15%.

If I look back several years, the company has spent, on average, almost double depreciation levels.

And I think that's generally the case with the rail sector as a whole.

It is not the case with almost any other industrial sector that's even capital-intensive and even has long-life assets.

So I was hoping you could help me understand that.

Cost inflation, I don't think, explains it fully.

There may be some opportunity, I wonder, if this is to be much more efficient with the cost of CapEx projects, because it does seem the sector as a whole -- not just you guys, but the sector as a whole, there's just a lot of money that's being left on the table with respect to efficiency of CapEx projects.

Yes, I mean, I think, you're right.

I mean, I think, the rails -- and I think, the simple answer, the short answer to why is the capital spending in the rail industry so much greater than depreciation?

Really the answer lies largely, I think, in the long-lived asset investments that are contained within that investment.

So there's a lot of moving parts there, but that really is the simple answer.

But I will tell you, as we look at Union Pacific's capital spending, we haven't finalized for 2019.

But you're right, I did call out that we're tightening our guidance to be less than 15%, and I'm very proud as we've worked over the last several years to be as disciplined as we can in driving the improvements that we've seen in our operating ratio over the last decade or so and more to come.

We've been equally focused on being disciplined and thoughtful and lined up around being disciplined around our capital spending, and we were proud to, kind of, walk that down as we have over the years, and we're just walking it down as a percent of revenue even further as we look to 2019.

Do you think there's an opportunity to get it to 13%, 14% where CSX is today?

Or is that just a little bit too low, given the growth opportunities you're seeing?

Stay tuned.

I mean, we'll see.

Again, I would say that it depends on a lot of factors, but I will tell you that we're as disciplined as we can be around our capital spending.

So we'll see.

The next question is from the line of Ken Hoexter with Merrill Lynch.

Two questions.

Tom, you mentioned the change from the test phase to now.

If you go back to when you were just starting this and you saw some changes necessary, how do you tell if it's working?

Just looking at the stats, some of the KPIs, cars per carload or trip plan compliance seem to have gotten, I guess, against what you would have thought given the early changes.

And how quickly can you make adjustments as you go through the plan?

So our railroad is going through a lot of changes right now.

The -- on the metric side, the -- that's not the primary reason for the failure cost you might have seen before.

We see improvement, as we turn the assets quicker.

That could create capacity, which is going to improve our reliability and ultimately go out to the efficiency side.

It's difficult to say or parse out, what part of our improvement is due to the Unified Plan versus other initiatives that we have.

Yes, but to your question, Ken, this is Lance.

Clearly, it's very early innings, right?

We just started implementing Phase I on October 1. What we're very optimistic about are seeing movement in some of the KPIs that we would consider the first early indicators, like car terminal dwell.

We're seeing that move in the right direction.

We're going to see, and are seeing, locomotive productivity as we're taking locomotives out moving in the right direction.

That car trip plan compliance, that has a lot of moving parts to it and it's hard to move it early, right, because it's combination of what are you doing to and from industry, which is first and last mile?

What are you doing in the terminals?

What are you doing over-the-road?

And so, the fact that it's treading water right now.

I don't think that's unusual or unexpected, but over time, we do expect that to move in the right direction.

And then, the other part of your question is, how quickly can we make changes?

This is turning into a way of life for us.

So while the phased implementation of the first iteration of Unified Plan 2020 or PSR on our railroad is, kind of, well conceived, inside of that, there are already opportunities to try something that's part of that plan, and if it doesn't work out just as thought, reconstructing it and doing it again.

That's part of what Tom touched on when he said, we've, kind of, reorganized some of our management structure on the operating side so we can make those kinds of changes more rapidly.

Appreciate that.

And just a follow-up.

Lance, your thoughts now on volumes after being up 4% to 6% in the last 2 quarters, seems to be where we're going a little slower to get started this quarter.

Is there -- is that just tough comps?

Or are you seeing anything that would suggest a slowing economy from your perspective?

I'm going to start with that, Ken, and I'm going to turn it over to Kenny to give us a little more technicolor.

But I think, overall, there's not many indicators that we see in our served markets, kind of, broadly that tell us of a global slowdown.

Now there's plenty of risk, right?

Chinese tariffs pose a risk and they can disrupt trade flow.

We see international economies, particularly Europe, looking like they're slowing down a little bit.

So I don't think it's all rosy on the horizon, but we just don't see any specific markers that tell us that growth isn't going to happen.

And Kenny, can you fill in a little bit?

Ken, this is Kenny.

First of all, we do have tougher comps in our sand volumes.

And so, we are seeing less sand moving in that Permian Basin.

The other thing that Lance alluded to is on our grain side, we are seeing, what I'll call, a delay of what could be moving out or exporting out here, and so we'll watch the soybeans and the grains and keep an eye on that.

Now having said all of those things, I'm bullish and the commercial team is bullish that we have some areas that we're expecting to grow into the quarter.

Our petrochem business will grow.

We're expecting the wins on our International Intermodal sector to grow.

So we're still feeling that it will be in the positive.

The next question is from the line of Jason Seidl with Cowen and company.

I wanted to talk about the plan and how that could impact both volumes and pricing going forward.

How do you envision that?

As you start improving service to the customers, is this a volume grab?

Is this going to be a -- your ability in some areas to increase pricing in some of your products considering that your core pricing, even excluding some of this other stuff, still seems to be lagging some of your competitors?

Yes.

So I'll start with that, Jason, and then, I'll pass it on to Kenny and Tom for their thoughts.

So the way we think about this plan and the impact on volumes and pricing is first, our approach to pricing is not changing.

And that is, the business has to be reinvestable.

It has to be attractive to us from a margin perspective.

Now there's a new filter put on that as we implement this plan and that is, it has to fit within the network that we've designed.

If you go back, one of the -- not one of, the single biggest prompt for us to switch how we're designing our transportation plan is that the way we had been doing it resulted in many different boutique services that became very complex to execute and, as a result, weren't feeding what the customer wanted broadly, which was consistent, reliable service.

That's why we're making this switch to Unified Plan 2020.

It will generate consistent reliable service as we focus on car movement and focus more of our attention on moving into manifest network versus boutique unit train networks.

And so, the net effect is, there is another filter we put on price or on business and that is, will it fit into the network as we've designed it?

In terms of the volume side, as we produce a more consistent, reliable network, that should generate more opportunity for us from customers.

I anticipate that's going to be the case.

I don't know exactly when that shows up.

And between here and there, there are probably risks to the volume, which are customers that either are concerned about the way we're designing their transportation plan and want to try other alternatives or customers' businesses that don't fit well with the existing plan and they want to try other alternatives.

So with that, Tom or Kenny, you got any other observations?

Yes, just real quickly, Lance.

You hit on everything.

I'll just say, I've been really impressed with the commercial team, because they've been out early and proactively talking with customers about the transportation plan that we're implementing.

And I can also tell you that as we look at the business, we are putting that filter on to make sure that it does fit with our network.

In some cases, there have been a couple of examples where we've elected not to support the business, because it does not fit into the network.

So if you're asking the question, are we prepared to walk away from business if it doesn't fit in the network?

Then that answer is, yes.

But like I started off saying, we're really focused on trying to grow our volume by educating our customers.

That was good color and, sort of, as a follow-up to that, on your -- you mentioned you have -- you're doing some terminal rationalization plans, you're closing a locomotive repair shop, so it's starting to look more like, I think, some of the investors expect from precision railroading.

I was just curious, have you guys brought anybody from the outside in on a consulting business for this program?

Or is this all being done internally?

Yes, so I'll take a stab at that and then, Tom might be able to give a little more detail.

So when you think about the expertise that we have in-house on PSR, we do have employees, some at high levels like Cindy Sanborn, who is now going to running the entire northern region of the network, along with others, particularly in the operating team and also in the network planning and optimization team, who have experience and, in some cases, deep experience in PSR.

Having said that, and understanding that we have complete confidence in the existing team and they're doing a hell of a job both designing from ground up because it's not being done in Omaha, it's being done by the people that have to execute the product on the ground.

We are always looking for ways to increase our knowledge.

One way is, we've spent a great deal of time with railroads that have implemented PSR to understand from their perspective, what worked, what they wish they would've been differently, what they wish they would have accelerated during their initial phases.

So we've learned from that and it's reflected in our own planning and we're always out in the marketplace looking for talent that can add to the team that we've got.

The next question is coming from the line of Scott Group with Wolfe Research.

This is Ivan Yi, on for Scott Group.

First question, just wanted to first clarify your 10% labor productivity target for 2019.

Does this roughly mean a 10% spread between headcount relative to volumes?

And what volume growth are you assuming for 2019?

And if volumes are weaker than expected, can you still get to 10% labor productivity?

Rob, can you handle that for us?

Yes, the -- you're referring -- on the 10%, that would not -- I would not take that as an overall headcount.

That's an efficiency measure, on that slide that Tom showed earlier.

That's not saying that we're going to have 10% fewer headcount.

What will dictate what the actual headcount number is, is -- productivity is a big piece of it obviously, but so will volume.

So that did not contemplate volume which, at this point, we think will be volume growth.

So there's a lot of moving parts on that.

I think the message there was, that's one of the key KPIs from an operating standpoint as we roll out Unified Plan 2020 that labor efficiency around running an efficient operation is something we expect to improve upon.

And we put those percentages plus/minus out as order of magnitude what we're expecting to achieve.

Great.

And my follow-up also on labor.

How many contractors do you currently have on the network?

And how much do you expect to reduce them by as part of the Unified Plan?

Well, that's a great question.

We don't know -- well, we don't have the numbers handy right now as to the total number of contractors on the network.

I'll just remind you that not every contractor looks the same.

So we said we're in the fourth quarter taking out 200 contractors.

Those are very much focused in the IT world, where they work full-time on projects that we give them and they're supervised by somebody else in a different part of the world.

We have many other contractors that do different things on our property, because they can do it either more effectively than we can or because the asset investment to do the work doesn't make sense for us to make because it wouldn't be deployed all the time.

So in that case, you think about something as simple as snow removal and grass cutting and then you go all the way up to contractors that are running some of our Intermodal ramps.

So there's a lot of moving parts there.

You've targeted properly, Ivan, that, that is an opportunity cost.

And to the extent that we can find low-value added work or ways to do that more efficiently, that's another bucket of cost that we're targeting.

Next question is from the line of Justin Long with Stephens.

Wanted to start with one on pricing.

On core pricing, could you just talk about what drove that deceleration sequentially?

I know, you've discussed pressures in Coal and International Intermodal for a while, but even if you strip that out, we saw deceleration of about 25 basis points.

And then, looking ahead on pricing in 2019, maybe I'm reading too much into this, but you said pricing gains in 2019, but you didn't specify above inflation.

So could you just talk about your expectation for pricing relative to inflation in 2019 as well?

Justin, this is Rob.

Let me start with that and Kenny will probably give a little bit more of a sense of what he's seeing in the market.

Number one, you know how -- I'll repeat for you and everyone else, how we calculate our price, which I'm very proud of and we've done for the 15 years since I've been CFO, and that is conservatively calculated on an all-in yield basis.

So it's not a same-store sales kind of number, it's an all-in yield.

And that's a very conservative way of looking at it, but I think it's the most accurate way of looking at what did you really take to the bottom line from your pricing actions.

And I would say that, as you look at the difference between our 1.75% all-in reported this quarter versus the 2% that we reported in the second quarter, that's kind of splitting hairs.

It wasn't really a big change because of the rounding mechanics is part of it.

But also, the way we calculate price, volume is a factor.

As a simple example, if we take a 10% increase on a piece of business and the volume is down, we yielded fewer dollars.

So we don't count that as a 10% increase, we count that as whatever the dollars that we actually yielded.

And so mix, which played against us this quarter, mix clearly was a factor in a conservative way that we calculate price.

Looking to 2019, as you know, Justin, we're never going to give precise guidance on absolute pricing numbers.

But to your point, I can tell you that we do expect it to be above inflation dollars.

I mean, so the dollars that we yield from our pricing actions in 2019, we're just as confident as we head into '19 as we were in '18 that those dollars will yield above our expected inflation dollars.

Kenny, do you want to?

Yes, you summed it up good, Rob.

I will say Rob laid out our methodology and we moved less sand revenue sequentially third quarter versus the second quarter, which had an impact.

But having said all of that, if you look at some of the other markets, I'm really pleased and impressed with our ability to price to the market.

You look at some of our markets like our domestic Intermodal market, you look at our carload business that's right in competition with truck, very pleased with our ability here to walk up the pricing to price to the market.

Okay.

So really said another way, outside of mix, you would say the pricing environment is stable or better than it was last quarter.

Is that fair?

That's a good way to look at it.

Okay, great.

And as my follow-up, maybe to ask the productivity question a little bit differently.

How volume-dependent is that $500 million goal for next year?

There seems to be more concern in the market about the macro environment and the cycle for both industrials and transports.

If we make the assumption, the freight environment weakens in 2019, do you still think this productivity target is achievable?

Justin, this is Rob.

As you know, we're never going to use and haven't used the lack of volume as an excuse for not achieving productivity.

Having said that, clearly positive volume is our friend, because it gives us more optionality operationally to squeeze out productivity.

But as we look to 2019, again, we are thinking that in our -- while we haven't finalized it yet, our early look is that volume would be on the positive side of the ledger and that certainly is embedded in our more than $500 million productivity.

So depending on what actually plays out, I will assure you if volume were down, which is not our outlook, but if volume were down, we would just as aggressively go after that $500-plus million number, but absolutely, what it turns out to be remains -- we'll see.

Next question is from the line of Brian Ossenbeck with JPMorgan.

So just wanted to go back to, I think, Slide 14 on the metrics and the perspective and the goals are helpful.

But just to get some sense of the benchmarking, the network performance has been challenging for some time, as you mentioned.

So where are these key metrics relative to when the network was a little bit more fluid?

And can you give us a sense of how these were benchmarked against your peers' operating PSR to the extent you were able to get that, sort of, detail and recognizing this is not an endgame between '19 numbers you have, but just put some perspective on how that looks versus UP network and maybe over a year or 2 and then also you peers would be helpful.

Yes, so this is Lance.

I'll start and then I'll turn it over to Tom.

So parsing that question out, the first thing to note is some of these KPIs are KPIs that we have not tracked historically and it would be a real job to, kind of, create them from a historical perspective.

So we're not likely to invest that time.

That time is like time not well spent.

Some of them we can track back historically, like locomotive productivity or car dwell.

And if you look at just overall, and those are published numbers, certainly car dwell is, you can see that some of our best car dwell was way back, probably about the 2013 kind of number or timeframe.

But car dwell for us, depending on how we design the network, ranges maybe at best 26 hours plus to in the 30-hour plus range when we're congested.

As we looked in the second quarter, one of the reasons that our inefficiency costs were reduced in the second quarter was that through the second quarter, we turned a corner.

And we didn't choose September for any reason other than it's essentially the first month before we started in earnest implementing Unified Plan 2020.

So there's really no magic about using that as a baseline.

We chose to use that as a baseline because some of these measures like the crude productivity number based on car miles, the car miles per day.

Those things, we're just starting them out.

So we figured we'd just start them out in September.

Tom, what would you like to add to that?

Really, Lance said, on the focus on car dwell, it minimizes that car dwell, [spinning] the assets quicker, to help drive our service product back up.

Okay.

And then, just to follow-up on that real quick.

The benchmarking versus [PSR of peers]?

Oh yeah, I'm sorry.

Brian, I'm sorry.

We did and have actually benchmarked ourselves against every other PSR railroad and we do that periodically against every other railroad.

And what I'd say is, the target ranges that you see on that metric slide are both engineered.

They're derived from the network as we think it will be designed on us, right, because the design is not complete.

But it's -- we took a very large first step in the Mid-American Corridor, and so we have a sense for what to expect in the remaining phases.

That's item number one.

Item number two, we've pressure tested that against what we see in absolute performance and improvement from starting point on the other railroads.

Now I would remind you, there's probably one kind of glaring difference in one way and that is, we're starting the project with an OR in the 62% to 63% ballpark, where some of the more immediate implementations have started at a higher OR level, so there's that.

But we also can see where everybody else started in terms of their locomotive productivity, their car dwell, their fill-in-the-blank.

And the other thing I would say is, we expect significant -- as we point out in that slide, significant improvement.

We're already starting to see early returns that grow that confidence in it.

But every railroad is a little different, right.

We have -- we go over the Continental Divide, we have some pretty significant rise and fall that probably means that's going to look a little different than somebody who doesn't.

Okay.

The follow-up is a shorter, quicker one.

The last time, we saw mix and price, this dynamic where mix was a bigger drag than core price as you calculated, it was 4Q of '15 and the first quarter of '16 versus beginning the last time frac sand market collapsed.

So Kenny, can you walk us through any of the offsets that you might have in those markets?

You mentioned petroleum and LPG growth as well as the crude oil out of the Permian.

Are those able to fully offset what you might be facing in terms of mix shift from the in-basin sand?

Yes, so thanks for that question.

First of all, I will just say that there are other shales out there that have a good pipeline of opportunities, and so we are pursuing those opportunities and getting wins there.

At the same time, you are aware of the crude oil opportunities that are out there, both out of the Permian and out of Canada.

We don't go into a lot of the details.

We don't talk about customers, but I can tell you that we have wins in both of those markets and we feel good about the volume that's coming out there.

It won't completely offset the sand, but we feel good about the volume, and we'll see where that upside takes us.

Next question is from the line of Ravi Shanker with Morgan Stanley.

If I can follow-up again on price and thanks for the clarification on price versus mix.

But just looking at the continued, kind of, pressures in the International Intermodal business.

I'm wondering if there's a possibility at all that you guys maybe pull back on that a little bit, kind of, if that is being a drag to overall returns and then pricing?

That's the first question.

Second is, Rob, you sounded extremely confident about getting price versus -- positive dollar price versus mix in '19.

We've seen some of the rail inflation benchmarks really spike going into the fourth quarter of this year.

I'm wondering, kind of, when you think of that positive price mix spread.

Is that because you expect that inflation number to come back down again?

Or do you think you can get the pricing to more than offset inflation if it's like 4% or 5% next year?

Rob, you want to take that price versus inflation and then Kenny will...

Yes, Rob here.

Let me just state, first of all, at the outset that and this ties into your question and the previous that we never give guidance on mix.

And I have always viewed it as a huge advantage of the UP franchise that we play in so many different markets that there's a lot of mix moving parts and there's mix within commodities and there's -- so there's a lot of moving mix things, which is why we don't even attempt to try to project that or guide to it.

We play the hand that the economy deals us.

But I will assure you that commercially and across the entire organization, we are focused on driving price where we can on every move, whether it's a positive mix or a negative mix, when you add it all up, it might be a short-haul move, for example, but we're just as focused on driving price and margin improvement on that short haul-move as we are a long-haul move.

So I wouldn't take mix as implying price, is my point there.

Number two, to your question on the spread versus inflation.

Let me step back and give you the guidance that we've given and that is that our dollars that we yield from our pricing actions will be greater than the dollars that we expend on inflation.

So you're right.

I would expect at this point that inflation will be higher in 2019 than we have seen in 2018, but the dollars that we yield from our pricing actions should still exceed.

Now it may not -- I'm not saying by what gap but it should still exceed the dollars that we expend on the inflation costs, albeit inflation likely will be a little higher.

And Ravi, I'll be pretty succinct here.

Each commodity market is different, each deal, each customer is different.

I can tell you when you see us win in the marketplace, it's at acceptable margins.

Great.

And just a quick follow-up.

I think on the slide, you had listed Coal as a headwind going forward.

I'm a little surprised to see that, because I think our team internally at least expects supportive nat gas pricing at least for the rest of the year, if not into '19.

Can you just talk about again remind us, kind of, what is the, kind of, baseline breakeven point you'd expect to see on nat gas?

And kind of, where do you think that goes in '19?

Yes.

So first of all, there's a number of things going on in play there.

Natural gas prices were down 3%, kind of, year-over-year by the quarter.

There are some retirements in there.

There's a book of business on the Coal side that's always coming up for renewal.

And as I mentioned in my earlier statement, we lost some business.

So we're not going to get every win there, and we focus on the ones that we win -- have to be at acceptable margins.

Next question is from the line of David Vernon with AllianceBernstein.

Rob, I wonder if you could talk about the $500 million and help us understand how we should use that for modeling and maybe how it compares to prior levels of productivity.

I'm just trying to think about how much of that $500 million we should expecting to drop down.

I'm not looking for specific guidance on what the earnings numbers are going to be next year, but as you think about the way you measure productivity in the past, how much and how effective has that number been in terms of dropping down to the bottom line?

And how does the magnitude of $500 million compare to maybe the last couple of years' run rate?

Yes.

That's a great question.

In this year, we've been working through the failure costs or the inefficiency costs that I've talked about.

So this year is not a good benchmark, but I think it helps inform and influence and set us up for a strong year in 2019, because we have every expectation that we're going to remove those inefficiency costs that we've seen throughout 2018 and then do way more than that with the Unified Plan 2020.

So it's a big number.

I can't tell you how to model other than we're showing confidence in our ability to drive productivity and, as we've said here, we're confident that -- or our belief at this point at the early planning phase is that volume will be on the positive side of the ledger and we're just as confident as ever on our ability to drive price above inflation.

So from a modeling standpoint, you'll have to, kind of, take those parameters and do what you think the economy is going to look like.

But I would say that the $500 million, and again, I said at least $500 million, is a big number.

I mean, you can look at historically maybe if you went back over a long period of time, you'd see UP in the 250-ish kind of range.

So we think $500 million in any 1 year plus is a big productivity achievement.

All right.

And maybe, Lance, just as a quick follow-up.

In looking at the rails over the years, it's always seemed to be conventional wisdom that the franchise at UNP is great, the traffic mix, the length of haul, the business foundation, which the network has built is really, really good.

As you guys are going down this path of Precision Scheduled Railroading, is there any reason to believe that maybe you can't do better than other implementations have been based on the quality of the business franchise?

Yes.

I think the best sum for that, David, is we anticipate we should be the best-performing railroad exactly for the reason of our franchise and team and many other assets.

And we should have the best financial performance.

That's what we're targeting, that's what we're shooting for.

And as you think about, sort of, executing against that vision with the board's, kind of, direction, are you -- is that where you're pushing the organization and you're willing to make the changes that you need to get there?

Like, I'm just trying to get a sense for that level of commitment, if you could just comment on that.

I believe, we've shown a appropriate level of commitment in this first month, if you will, of activity.

Consolidating regions, consolidating engineering functions, moving something like customer care and support from marketing and sales into the operating team, finding 475 jobs that we think can be done more efficiently, streamlining the organization.

There are many first steps that are literally just first steps.

And yes, that level of commitment necessary to ultimately achieve a 55% operating ratio at some point in the future, that is in place.

Next question is coming from the line of with Bascome Majors with Susquehanna.

Not to beat a dead horse on the operating targets, but I thought it was interesting with the KPIs that you laid out and roughly a 10% productivity improvement ratio on those between now and the end of 2019.

If we want to track this with the data that's publicly available to us and, kind of, measure your progress in more real time, what would you suggest we look at?

I mean, clearly, if you're successful, the OR should improve as you go down this path.

But what could we look at weekly or monthly that you would point us to?

Yes.

So you see a couple of those KPIs that are currently published, terminal dwell is one.

That will be a good one to watch.

That's a good overall indicator.

We publish a freight car number, I think, is a big more inclusive number, but that should show some movement as well.

And the other thing to note is, those KPIs are now being built in to how we monitor our business.

And we will have an opportunity to review them with you on a quarterly basis as well.

And that's actually a really good segue into my follow-up.

Lance, after some changes last year to your incentive plans, right now it looks like the senior management is driven by a combination of operating income and operating ratio with some ROIC targets driving the longer-term incentives.

Based on your discussions with the board around your new operating strategy over the last few months, is this the right incentive package to drive outcomes you're looking to achieve with the new operating plan here?

And I guess, if the answer is no, we might tweak that, how might things might look differently from incentive standpoint next year?

Yes, that's a great question, Bascome.

So we do believe, at this time, that our current incentive packages are appropriate and driving appropriate behavior.

And it's for this reason, in the big picture, if we think about what's important in this -- in our industry and our company, first and foremost is making sure we get our investment and invested capital right and that -- from that perspective, I mean, spend it at the right time, at the right amount, in the right place, and generate the returns necessary to attract capital.

And so, the long-term plan being built around ROIC makes all the sense in the world, right?

We're a very capital intensive company, and we got to keep an eye on that.

We've modified that so that the management team gets punished or rewarded incrementally based on relative performance on operating income.

So that's a relative operating income.

So that's an important long-term incentive plan.

I think it looks right.

The short-term is built on 2 big measures.

One is, operating ratio, which is how efficiently we are taking the top line and translating it into the bottom line.

That's critically important for us.

And operating margin, which is -- or operating income, again, which is, are you growing and are you growing as expected.

So for right now, we think those make sense.

But the other thing to note is our compensation committee of the board is always evaluating what are the right incentives and how do we align management with what makes best sense for rewarding shareholders.

I think we've got that right, right now, but that does change over time and it could change in the future.

Next question comes from the line of Allison Landry with Crédit Suisse.

I think you mentioned earlier or gave an example of bypassing classification yard.

So just wonder if that means we should maybe expect a rethink of the hump yards.

And maybe if I could ask it this way, how many cars per day does one of the hump yards need to process in order to justify the economics?

And how many of your top 10 classification yards meet this threshold?

Well, I appreciate the question, Allison.

As I referenced, we are going through a terminal rationalization process.

That is a part of our function as the traffic levels or traffic shifts through our new programs and our new plan.

As far as hump yards or switching yards, we are looking at what cars specifically need to be there and assessing if those cars, if we can shut down that facility and become more efficient with surrounding areas.

As far as a hump yard goes, they vary a little bit, but 1,000 cars-ish at a hump yard is what we're looking at from a rationalization process, but it's going to be predicated on how the traffic flows end up in our organization.

Yes.

Allison, I wouldn't use that as a hard and fast rule, right?

You know, I think that hump yards are very different around the network.

Some of our hump yards are designed to run at 2,000-plus cars a day.

And some are designed to run at 1,200 cars a day.

The endgame really is, if the cars need to be switched and need to be switched there, is that the lowest cost option?

And if it is, we'll keep using it.

And if we get to a point where utilization drives an alternative decision, we'll do that.

Okay.

That's helpful.

So maybe the number per day isn't the only focus and mix and interchange and that sort of thing may play a role.

Yes.

Basically, where you have to switch the cars, that will dictate -- because one of the things we're doing is we're reducing the times we touch a car, which is great overall.

But we still have to touch it somewhere to get it in block.

Okay.

And just based on what you're seeing or, sort of, the high-level view of all the yards, does it seem like there's a fairly significant opportunity to consider shutting down some of these yards?

We're working through that process right now.

We don't have specifics at this point.

We're working through that.

There's not a specific yard we've targeted at this point, but we are looking closely at those opportunities of where the car should be and if there's opportunity to shutter a yard.

There'll be more to come on that later.

Our next question is from the line of Chris Wetherbee with Citi.

I wanted Rob to come back to the productivity question for a minute and I don't want to minimize the $500 million plus because it is a big number.

But when I look over the last 3 years or so, it's sort of been in the range, at least the projection of the beginning of the year, has been in the range of anywhere from like $300 million to $400 million plus.

So I guess, I wanted to get a sense of, you know, if this implies some back-end, sort of, waiting, I guess, that the program which would seem to make sense just given the fact that you're starting now and it will probably continue to ramp through 2019, or if there maybe potential upside opportunities from the -- on the productivity side from implementation of the plan.

Yes.

Chris, I would say that we're not finite enough yet to calendarize that $500 million in, but you're right.

We're -- we will be ramping as we go in terms of the implementation of the Unified Plan 2020.

So that likely will have some impact on the calendarization of the $500 million.

And again, I would stress that, recall I said, at least $500 million.

So we're going to go -- we're not going to -- hopefully not going to leave any money on the table here.

We're going to aggressively go after all the opportunities, and we're feeling good about this.

And oh, by the way, we have to get that kind of number plus in order to drive to our ultimately our 55% OR, which we're committed to and driving to getting to 60% by no later than 2020.

So it's going to take this kind of initiative.

And so I guess I would summarize my answer by saying, we're going to go after that.

It's going to be at least $500 million.

Likely, we'll have some lumpiness, if you will, from quarter-to-quarter or when it actually plays in and we're going to be working through that as we phase in the remainder of the network on the UP 2020.

Okay.

That's helpful.

I appreciate that color.

And then just one quick one.

I appreciate that it takes some time to kind of run through what you think the potential opportunities are and you guys are implementing that.

When should we expect, sort of, a bigger update in terms of what you can get OR wise, both near-term as well as maybe longer term?

So I guess this is a short question as, sort of, when do we get the, sort of, full set of financial targets based on the program?

Yes, Chris.

This is Rob again.

I would say that -- I mean, stay tuned.

I'm not promising that we're going to change our guidance, but I would just say that, as you know, having followed us for many years, we're not going to slow down to get to a target by the absolute end of the target date.

And I use that reference in the 60%, we're going to get to the 60% as safely and efficiently and as quickly as we can.

But at this point in time, our guidance still is by 2020.

That doesn't mean in 2020, but by 2020 is our official guidance on that.

And so, we'll update you as appropriate, if need be.

But at this point in time, that's how we're marching.

Our next question is from the line of Tom Wadewitz with UBS.

So it's -- I appreciate the information you shared in the slides, that's helpful and the commentary on the Unified Plan 2020.

I feel like I'm still a little unclear in terms of some of the things -- your activity and so forth.

Have you, kind of, come up with the new schedule and that's part of the 150 changes you referred to that there is a new schedule on the Mid-American Corridor and you're implementing that?

Or is that something that you're still, kind of, doing work and you'll come up with the new train schedule at some point in the future?

So with our 150 different transportation plan changes that we're implementing, about 125-ish of those are completed right now and we'll continue to move forward.

We've implemented the ones that were the quickest and the highest impact.

We've just cut, over the last week, our sand plan, which moved a density of unit trains that would build density over a couple of days to a daily service.

That's not without the lumps.

We've had some issues getting that going.

But overall, seems to be making the forward run.

This week, we started our Intermodal plan of running longer Intermodal trains up and down the corridor here.

So that's just getting started.

Yes.

So, Tom, I think that to directly answer your question, it is a new transportation plan.

And what it means is, Tom's team started at ground level with the cars to and from industry and built from that where's the most efficient place to switch those and do work, what's the most efficient way to advance them deeper into the network to where they want to go or to interchange and then reconstructed a transportation plan that does that.

And so, when Tom's talking about the sand plan, in your mind think, there's a number of origins up in Wisconsin and Minnesota and they accumulate cars and they fill those cars with sand over time and in a unit train, it takes 3, 4, sometimes 5 days to accumulate the cars, create a unit train and ship them down.

In today's plan, instead, we pull cars daily from those origins.

We accumulate it into a manifest network and ship them down.

What that does is, it saves days on car trips.

And you expand that into all the commodities and all the customers that we do across the network and that's how you get time savings on car movement and dwell savings in yards.

So is that just directionally, is that driving a reduction in train starts and an increase in train length or not necessarily?

So the first things to focus on are, it will drive a reduction in the time it takes a car to go from origin to destination on average.

It'll reduce the time a car is sitting in a terminal on average.

It will reduce the amount of locomotive power required on average.

I think, we'll see some train start impact, it should.

And train length, it could also continue to help us grow train length.

Those moving parts are a little less clear in terms of ultimately we'll consume fewer crews, because the product is more efficient and reliable.

How much of that is because the train -- physical train starts, that's hard to tease out right now.

So a lot of this is about how we're balancing that network one end to the other, a to b, b back to a. That drives efficiency in our over-the-road crew starts.

So you should see that -- that's how we're coming about our [crew] -- our TE&Y.

Okay.

And then, just 1 brief follow-up.

This type of approach generates improvement in labor productivities.

Do you see headcount go down?

That's been the experience at other railroads.

Are you reducing your hiring?

And is the training pipeline coming down, given it seems likely you would see some labor productivity and maybe headcount reduction going forward?

We are.

What you said is true.

The pipeline for new hires is starting to come down.

It was doing that at the tail end of the third quarter, and we anticipate that to continue.

The next question is from the line of Brandon Oglenski with Barclays.

I know it's been a long call, so I'm just going to ask one.

But Lance or Rob, I guess, as you guys look out and you've committed to keeping CapEx below 15% of revenue.

If we put that in context, if you look at some of your West Coast railroad competitors, some of those have spent at levels quite a bit higher but have actually seen a lot of volume and top line expansion as well and it's pretty notable that you guys are still below your 2014 peak in revenue.

So I guess, how central to the plan is developing top line and volume in the network?

And do you think it can be done at these lower capital levels?

Yes.

Great question, Brandon.

So bearing in mind that winning in our world is generating growing and highest amount of operating income and cash from operations, so that we can reward our shareholders, attract capital and put the capital back to high-return projects.

I think the short answer is, yes.

The way we think about capital, we build it up from the bottom.

So we don't start with a target number but that looks to us like less than 15% of revenue.

And I think as we get more efficient, what I expect would happen is, we're going to have capital that is physical existing structure, infrastructure, available to use.

Now our capital spending might target a little differently.

Prior to implementing this Unified Plan 2020, we might have spent in this one area and once it's implemented, we might have work happening in a different area that we'd like to either debottleneck or support with incremental capacity.

But I don't think that changes overall the net-net large number.

And to achieve what we try to -- what we need to achieve, a 55% operating ratio, attractive ROIC.

Growth will help, and I anticipate we will see growth because of more reliable consistent service product.

But I don't think capital is going to retard us from being able to do that.

And I think if I heard you correctly, the priority though, first and foremost, is earnings and returns within the organization?

Our priority is always generating growing operating income and growing cash from operations.

Next question comes from the line of Walter Spracklin with RBC Capital Markets.

So Rob, I guess I'd like to come back to the OR shift in terms of your confidence level about an improvement and suggesting that something has changed here that will obviously give you less confidence.

What's happened over the last few months that's led you to, kind of, flag this OR guidance for this year to be at risk?

Yes.

That's a good question.

And as I pointed out in my comments, we're still focused on eliminating or reducing the risk as best we can.

But you're right, we did flag the full year operating ratio year-over-year improvement in '18 versus '17 as a slight risk.

And I would say, why is that?

Well, what we're seeing is the inefficiency costs that we're going aggressively after.

We have every expectation that they will be behind us, certainly, as we head into '19 are still lingering.

Some of the revenue shift has not been our friend.

Notably, as Kenny commented, the grain markets have not been our friend.

But I'm not going to use that as an excuse, as we don't, because volume has still been fairly strong.

But that in fact is one of the challenges.

And on a smaller scale, some of the costs associated with the right decision to reduce some of the headcount.

The timing of that, we're obviously going to have a little bit of a -- perhaps a small headwind in the fourth quarter on that.

But you add all of that up, those are the things that have changed.

Okay.

Going forward to next year, and you gave us the $500 million, but you did not give us an OR and going back to a previous point, just holding margin constant and then reducing by $500 million productivity after the revenue change should give you around 200 basis points.

So are you effectively guiding to a 61% OR for next year?

And if not, where would the math be wrong there in terms of that calculation?

Yes.

I mean, we're not giving -- because we're just not to that point yet.

We're not giving guidance on a full year operating ratio in 2019.

But again, we'll see how it all plays out.

We're confident in $500 million plus productivity.

At this point in time, we do see positive volume.

We're confident in our ability to generate price.

So you add it all up, I would expect that we will make nice improvement in the overall margins.

What that number ends up being, stay tuned.

But clearly, we are focused on a meaningful move in the right direction on the operating ratio.

I guess, where I'm asking, is $500 million not equating to about 200 basis points, all else equal, given your current assumptions?

Yes, I mean, your math is not wrong.

I mean, that just, kind of, in isolation is a fair statement.

Okay.

And then, you came back, Rob, again, to sort of reiterate the 2020 guidance.

But Lance and everyone here and Tom, everyone has talked about this substantive opportunity that the Unified Plan is presenting.

And I just -- I'm struggling with why you're not changing your guidance for 60%?

Is it that the 60% was no longer achievable and now it's achievable now that you've got a new plan?

Or you haven't calculated exactly the impact or got a full sense of what the impact could be?

I'm just really struggling with why you would keep the 2020 target given the substantive change that your organization has put forward with the Unified Plan?

Yes.

I totally get it, Walter.

And I would say that the answer to your question is several things.

One is, we are in the early innings of the implementation as we've talked all morning here of Unified Plan 2020, gaining traction, feeling really good about it.

So that -- we're feeling very good about that.

We haven't changed that guidance, but I would just call out that I wouldn't read too much into that.

I wouldn't read that, oh geez, they're not confident, because we are confident.

But I would just highlight that, at this point, that is a by 2020, it's not a end of -- last day of the year of 2020 kind of thing.

So we're going to get there as safely and as efficiently and quickly as we possibly can.

And stay tuned, I guess, in terms of what kind of pace we're able to make as we get more innings under our belt with the implementation of Unified Plan 2020, I think you'll start to see and will talk more granular, I would expect, about where are we on that path of improvement.

Our next question is from the line of Keith Schoonmaker with Morningstar.

Compared to other rails that have implemented precision railroading, you pointed out that UP already is bringing some really attractive operating ratios, probably 10 to 15 percentage points better than where some others begin.

But it strikes us that another big difference is the scale whereby your network is a multiple of the size of the others in many dimensions, track, miles, power, revenue, volume.

Could you please elaborate on what complications this brings, especially to those of us that are not railroad operators?

Sure, I'll start and then Tom can add additional technicolor.

You're right.

Our scale is quite a bit different than the other railroads that have implemented PSR.

And I mentioned earlier in the call, as we've engaged with our peers in those other railroads, we don't really see anything there scale-based that tells us there's incremental risk to us.

One of the things we've done is, instead of implementing a brand-new transportation plan across the entire network all at once, is we broke it into phases.

That's a recognition that scale, kind of, does matter in terms of magnitude of risk and if our game plan early on was going to break some customer relationships or if we got some aspect of the design wrong or some such thing, we didn't want to do that system wide.

We wanted to learn that in a smaller chunk.

And that informs, kind of, why we went down that path.

But Tom, is there any other observations you want to make there?

Yes.

So we've turned our operating model upside down.

And we want to make sure we're solid on that as we move forward, and designing the plan and working with our customers to be proactive to improve that service reliability.

The next question is from the line of Matt Reustle with Goldman Sachs.

Just one quick one for me.

Rob, you mentioned you have more debt to raise, rates are moving quite a bit and particularly since the Analyst Day.

Just given the size of that, that you'll be raising here, these swings do have an impact on the bottom line.

Is that something that you're considering in terms of the time line for raising that next big chunk of debt?

Yes.

I mean those are all factors that go into our planning.

We don't have anything to announce here, but yes, I mean, I think, all the things you highlight are -- go into the mix of us determining what's the right business decision to make and when.

Okay.

So is that something -- I mean, has your expectation moved up in terms of maybe accelerating when we can see that next leg?

No.

I wouldn't say -- I mean, we haven't put a date on that, but I wouldn't say it's materially changed our thinking.

We have nothing to update here.

This concludes the question-and-answer session.

And I will now turn the call back over to Lance Fritz for closing comments.

Thank you, Rob, and thank you all for your questions and joining us on this call today.

And we're looking forward to talking with you again in January.

Ladies and gentlemen, thank you for your participation.

This does conclude today's conference.

You may disconnect your lines at this time, and have a wonderful day.