The goal of capital expenditures is to link investments to the quality of the transportation product, generating increased customer satisfaction. Perhaps no railroad does it better than Norfolk Southern.
The Norfolk Southern approach to capital investment is to provide the right resources at the right place so that all commodity groups—coal, intermodal, manifest—can use the same resources to add value for the customer.
Here, Chief Executive Officer Wick Moorman sets the strategic view for the $2 billion 2013 Capital Plan. Chief Marketing Officer Don Seale tells how asset-sharing and “co-production” provide more bang for the capital dollar.
RA: At the 2013 NS Shareholders meeting, you said the businesses that prosper will be those that can “anticipate, adapt, and accomplish.” How does the 2013 capital plan support these three tasks at NS?
Moorman: Our capital budget maintains the core infrastructure of the property, and supports our targets for growth and higher returns. About 75% of our 2013 capital plan keeps our infrastructure and technology current. The other 25% is for the service enhancements we need to make to provide the best transportation product that our customers need.
Hockey great Wayne Gretzky once said, “I skate to where the puck is going to be, not where it has been.” In railroading, you can think of two or three places the traffic might go and position yourself to capture that business. The obvious example for us is our capex commitments over the past few years to build our intermodal components. To the extent that we can anticipate that intermodal will continue to grow, part of the 2013 budget is aimed at supporting that growth.
Over the past couple years we’ve seen substantial changes in our traffic patterns. We have a systematic process in place for thinking about infrastructure. Where do we need more capacity? Where are the pinch points now, and where might they be in the future? Wrapped around that is a lot of analysis and modeling to understand where we should spend infrastructure money and how we’ll get the best return.
We’re doing that in anticipation that these targeted corridors will continue to grow—be it intermodal, coal, crude oil—and which ones will require an adaptation strategy. Traffic shifts of coal to oil, for example. As we’ve seen crude oil ramp up, we’ve quickly added infrastructure and rapidly made some operating adjustments. Being nimble in the way we adapt to shifts in traffic mix—as well as preparing the railroad to handle significant changes in traffic—is a big part of the way we think about capital as well.
We also think about our locomotive fleet in terms of skating ahead of the puck. We want to make sure we have enough locomotives and, at the same time, minimize the cost of maintaining that fleet. So anticipate, adapt, and accomplish run through much of our capital spend for 2013 and beyond.
RA: The 2012 capex spend of $2.2 billion was 19% of the year’s $11.0 billion in revenue. Street estimates for 2013 revenues stand at $11.24 billion, making your 2013 publicly announced capex spend of $2.0 billion about 18% of sales, up from 15% of sales in 2011. Do you see this trend of higher capex spends as a percent of revenue continuing?
Moorman: We do see higher capex spends as a percentage of sales. Part of that is the $229 million annual spend on PTC. But ex-PTC, we can support higher capex levels through increasing operating cash flow from improved operating income streams. We’re comfortable with the spending where it is, and it can even go up some. The big ticket items—rails, ties, ballast—will continue to be the largest line items.
On the other hand, we have some areas that will diminish slightly without impacting the quality of the railroad—technology, locomotives, cars. Much of our coal hopper fleet is life-expiring right now. The coal business is depressed, moderating over the past four to five years the car replacements that six or eight years ago we thought that we were going to have to buy. Nonetheless, other pieces of our car fleet are going to require substantial investment over the next few years.
We also have a very good and innovative program to rebuild road locomotives—the SD60E for example—for about $1 million each, less than half the cost of a new unit. So if we can continue to do that over the next few years, we’ll be able to compress that side of the capital budget. I’m confident we can perhaps return a little more to our shareholders in the bargain. We pay out roughly a third of net income—the highest dividend yield in the industry—and I can see that increasing over time.
RA: Coal cars: You don’t have to replace them as fast, but now you can rebuild the cars you want to keep with composites, adding ECP, and increasing payloads.
RA: So many of your thermal coal cars are utility-owned?
Moorman: Yes, a high percentage. What makes us different is the met trade (domestic and export comprise about 30% of NS’ coal tonnage). Met coal doesn’t really fit the leased-car model.
RA: Doesn’t that also support your at-pier blending ability where you can make custom mixes without having to put the coal on the ground?
Moorman: The blending capability has always been one of the great advantages at Lamberts Point. We’ve just reopened one of the piers after a 90-day outage to rebuild substantial components. We’ll spend a lot of capital on the piers over the next few years. As the largest coal loading facility in the Western Hemisphere, it allows us to do a lot of things that our competitors can’t.
RA: AAR data through May suggests NS is running about 168,000 cars on-line vs. 180,000 in 2Q 2012. Where can you go before you get clogged up?
Moorman: We’re running 165,000 to 170,000 cars and can certainly handle more than that. We look at where we have a lot of trains today, and where we think more are coming. We do an analysis to determine where we need another siding. On average we’ll do four to six pieces of double-track or siding extensions a year to remove bottlenecks so we can position ourselves for future growth.
RA: You closed Buckeye Yard in Columbus, Ohio, curtailed hump operations at Roanoke, Va., and are doubling the footprint of Bellevue, Ohio. How do these actions support the “anticipate, adapt, and accomplish” theme?
Moorman: Bellevue is an adaption, and the others are part of the same process. Prior to the Conrail transaction, Bellevue did not fit a lot of what we were trying to do. Post Conrail, with the acquisition of the Northern Region, we built out the infrastructure on both sides of the yard so that you can slide traffic off the ex-Conrail lines into Bellevue and back out to the NS lines and vice versa. Closing Buckeye was one result.
We anticipate that merchandise carload traffic will continue to grow. At some point we could run out of switching capacity. Even though Buckeye was out of route for some of the traffic, the yard gave us sorting capacity we didn’t have anywhere else. The question became, do we rebuild Buckeye, or do we expand Bellevue? We decided to do the latter because we can now do some very interesting things at some of our different terminals—blocking for western connections and reducing the number of times we handle a car. And since one of our key metrics is car handling, lowering that number increases velocity.
Roanoke fits the same general theme. We were handling a modest amount of traffic. We had yards around it that had capacity—Linwood, N.C., for example. We took a hard look at Roanoke and concluded we could do the same job more efficiently elsewhere.
RA: Another good example is what NS calls co-production. You have an intermodal facility in Greencastle. You also have an intermodal facility at Rutherford and another at Harrisburg. That gives trucks the flexibility to triangulate among the three places and creates a competitive advantage for NS.
Moorman: We have a fairly dominant position in central Pennsylvania. It’s already a competitive advantage because of the enormous amount of truck traffic there. Greencastle provides one more outlet for some of the truck traffic that wants to flow into an intermodal terminal in that general area, extending into the Baltimore and D.C. markets.
The Harrisburg Triangle gives shippers, the dray people, and NS far more flexibility in terms of where we can take this traffic and handle it most effectively. Greencastle is aimed almost exclusively at north-south traffic while with Rutherford and Harrisburg you can handle both east-west and north-south traffic.
RA: Continuing south along the Crescent Corridor, it appears you’re putting intermodal terminals closer together, again to minimize dray and create competitive advantage.
Don Seale relates how NS uses its network to win new customers and encourage current customers to come back for more.
RA: NS runs as an integrated network with specific corridors as the backbone. Everything NS does to make the railroad run better is done with the network and its corridors?
RA: Examples of corridors and how they add value?
Seale: NS has developed four primary corridors and has been successful in creating public-private partnerships in two of them. The 789-mile Heartland Corridor was the rail industry’s first multi-state PPP. Our Meridian Speedway joint venture with KCS is the shortest, fastest rail route between the West Coast and the Southeast. The Memphis-Harrisburg Crescent Corridor covers 2,500 miles across 11 states and extends to Dallas and Mexico plus Birmingham and New Orleans. We’ve added new intermodal terminals in Rossville, Tenn. (Memphis), McCalla, Ala. (Birmingham), and Greencastle, Pa. (Harrisburg). In late 2013 we’ll open a new Charlotte (N.C.) terminal.
In Mechanicville, N.Y., the west end of our Pan Am Southern Corridor, we put an automobile terminal adjacent to the intermodal facility and we repeated the model in Ayer, Mass., to serve the Boston market. At Greencastle, we’re adding a merchandise carload industrial park adjacent to the intermodal terminal so we can serve both carload and intermodal customers from the same asset base. Co-productivity, again, and now that we have the formula, you can bet you’ll see more of the same across the NS franchise.
RA: It certainly looks like NS is betting heavily on the intermodal segment, even though it generates about 20% of today’s revenue, with coal about 25% and merchandise carload the remaining 55%. Why is this?
Seale: The additional intermodal capacity we’re bringing on will open many new business lanes for NS and will be the foundation of future growth for years to come.
We know roughly 80% of all intercity truckload freight is moving east of the Mississippi—right into the heart of the NS network. The number of truckloads we see as being ripe for conversion to rail is large enough that we can grow at a significant pace even if we gain a small percentage of that motor carrier volume. We know customers are buying rail networks and that’s what is behind our decisions to go into New England, the Ohio Valley, or even Mexico. That’s why we’re adding 34 new intermodal lanes, including 18 in and out of Mexico. In the process we’ve sped up the capital program a year. Our 2013 plan is $2.0 billion, $233 million-to-$200 million less than last year. We have a smaller budget this year because much of our intermodal franchise investments are nearly complete. That’s one-time money that’s in the ground now, and we can move on to other things.
RA: What about the single-carload service product?
Seale: The poster-child for our carload service is still the vanilla general-service boxcar. Our goal is to standardize one car type, and we’re finding that the single-door, Plate F, 100-ton boxcar offers customers the most carrying value and for ourselves the highest use rate.
However, we still see many of the older, smaller 70-ton, Plate C cars in service, mainly for paper shippers. Your typical newsprint or boxboard plant was built 50 or 60 years, ago and doors were spaced for the 50s because that’s all there were. Fast-forward to today and the trend to bigger cars moving faster made the 50-footer obsolete. But the paper plant doors don’t fit the 60-foot cars.
The day may be at hand when we railroaders have to do for the papermakers what we did for the animal feed producers. We demonstrated how unit trains of feed ingredients could save them time and money and increase margins. We collaborated on the feed mill evolution from single cars to bigger and bigger unit trains. A big motivator was the economies of leaving power with a train during loading and unloading and accelerating over-the-road transit times for more uniform flows from origin to destination.
Cycle times dropped to 8 or 9 days from twice that. The economic benefit to the feed mill made the case for plant expansion to take unit trains. And NS generated increased volumes with higher per-car yields, lower fixed costs, and lower capex.
RA: So why not with boxcars?
Seale: Manufacturing processes can change daily based on customer order quantities, raw material availability, manufacturing capacity—any number of variables. But farm animals must eat, the lights must stay on, and retailers must keep the shelves stocked. As a result, it will always be trucks-to-the-rescue when the manufacturing variables kick in, and it’s the old reliable railroad for every day.
Much of the single car business will stay single car business. At NS, we’re finding that, even with the manufacturing variables, NS’ investment in equipment utilization, network velocity, and workforce productivity makes the single-car business a good business to be in.
Chief Operating Officer Mark Manion and Chief Information Officer Deb Butler talk about how NS will add value to the transportation product.
RA: Earlier in the year, Wick said, “The fluid network leads to better service, lets us lower operating costs and adds network capacity.” What’s first on your list for developing a more fluid network?
Manion: First is finding the most efficient way to move cars between origin and destination while maximizing revenue tonnage per line segment. The 2013 capital plan includes $84 million for adding capacity to the line segments we’ve identified as needing greater capacity to handle the volumes we expect—skating ahead of the puck, as Wick puts it.
Keep in mind that we developed our network plan—The Thoroughbred Operating Plan or “TOP”—in 2001 to include the entire system. That was before we started the corridor concept and yet, as the concept has evolved, the corridors have become an important part of the network plan. TOP gives us the operating consistency we need to identify potential choke points before they become choke points, and we can apply capital dollars to add infrastructure—longer passing sidings, double-track extensions, signals, and cross-overs.
We used computer modeling to add the infrastructure changes surgically, exactly where we needed them so we were not moving the problem someplace else.
Over the last 18 months the network-with-corridors concept has really paid off, giving us the best operating metrics we’ve ever seen. A big reason behind that is we’ve got the right resources, and we’ve got a crew base that absolutely supports the network plan. We’ve got the right number of locomotives and the right infrastructure. Put it all together, and we’re now to the point where the network can handle the volume.
RA: The 2013 Capital Budget calls for $2.0 billion and has $1.1 billion in three buckets for track and structures: Roadway, Infrastructure, Facilities & Terminals. What goes where?
Butler: We expect to invest $831 million in Roadway projects for the normalized replacement of rail, ties, and ballast and the upgrading and replacement of bridges. Facilities & Terminals gets $203 million for property enhancement and new projects, such as the intermodal new terminal in Charlotte, N.C. And the $84 million Mark just mentioned is in the Infrastructure Fund—increasing mainline capacity and accommodating volume growth. Also in this $84 million bucket is money that’s our share of funding for public/private partnership investments like the Crescent Corridor and CREATE in Chicago.
RA: You say the Infrastructure bucket includes contingencies. How does that work?
Butler: You recall I said “accommodating volume growth.” That’s a contingency because we can’t say this far in advance where all the volume growth will be, and we must be prepared. This year we’ve set aside $40-50 million for infrastructure enhancements that we know we’ll have to do but where and when are more a function of traffic patterns that have yet to develop fully—crude oil, for example.
Another example is in Bellevue, Ohio, which is achieving critical mass in terms of traffic volumes. The infrastructure fund lets us be nimble in our response to new traffic patterns.
Manion: Every analysis of traffic flows across the system shows nearly everything we move in the North wants to go through Bellevue. Doing so benefits our customers because we can concentrate our major class yard activities in one place and eliminate intermediate yard dwells. Blocking for the distant node lets you bypass intermediate yards, and even run manifest freights directly to and from foreign road terminals. Think Bellevue and North Platte on Union Pacific, for example. Fluidity, again.
RA: That’s the third time you’ve mentioned fluidity. How does NS measure fluidity?
Manion: We use a three-part Composite Service Score that measures train performance, connection performance and plan adherence. We measure operating plan compliance for every train between end points. Compliance means being on-time, and at NS on time means on-time, neither early nor late.
We’ve had to make some cultural adjustments to meet our metrics, too. In the past, intermodal trains were our hot-shots, and the faster they got over the road the better. Not anymore. We run intermodal trains to make our guaranteed customer availability times at terminals. We run merchandise trains to make connections and scheduled load placements. And we score our local trains for Local Operating Plan Adherence. As a result, you might see us let a manifest train that has fallen behind plan overtake an intermodal train that was ahead of its operating plan, which leads us to connection performance. We don’t hold trains for connections because the biggest cause of late arrivals is late departures. Worse, late arrivals get power and crews out of position to meet system plan compliance.
Finally, we measure trip plan compliance for every revenue unit from the time we get it from a customer or another railroad to the time we either deliver it to the customer or hand it off to another carrier. Here again, the repair-and-replace and capacity-addition components of the 2013 capital plan give us the tools.
RA: Let’s turn to the carload side of the house. I sense three recurring themes: asset utilization, network velocity, and workforce productivity. Let’s start with asset utilization and network velocity. How are you getting more loads out of your freight-carrying assets by making them move faster?
Manion: A good example is our decision to close our Roanoke hump yard. We found that most carloads classified at Roanoke were heading for another hump yard before getting to their final destinations. By closing the Roanoke hump we forced blocking at the hump yard closest to the point of origin to the hump yard closest to the destination. We saved a day’s dwell in Roanoke between arrival in the receiving yard and actual departure and took out the odds of cars missing connections or being otherwise delayed. Car—and thus network—velocity improved.
Asset utilization is increased by having yards before and beyond Roanoke do more work while we put the Roanoke assets to other uses. We can do block swaps and stage train sequences to maximize gross ton-miles per day between yards and make better use of our mainlines. We shifted hump locomotives to more productive uses and were able to give our crews regular eight-hour jobs that did not vary with traffic volumes.
RA: Does this mean NS will be closing other hump yards?
Manion: Not necessarily, though we were able to close Buckeye in Columbus because Bellevue was a better location. Most of our carload traffic between the West over Chicago on the one hand, and the Northeast, Mid-Atlantic, or Southeast on the other, flowed through Bellevue. It was clear we had to remodel the classification yard network according to where cars wanted to go.
So we split the work that Buckeye had been doing between Bellevue and Conway – and even Enola—depending on which is the best “distant node.” We took down the second hump at Conway sometime back, which helped drive volumes at Bellevue, and we used the space at Conway for a staging yard for ethanol and crude oil trains heading east.
RA: But isn’t crude oil riding mainly in unit trains?
Manion: About a third of our crude oil is the heavier oil-sands product coming out of Alberta and heading for East Coast refineries. We put that in merchandise service as far away as Conway to keep it moving and to get better car-cycle times. Once in Conway, we build it into unit trains that go direct to destination. Our Bakken crude moves mainly in unit-train service, and we don’t see that changing.
RA: So how do you keep everything moving according to plan?
Butler: That’s in the $57 million technology piece of our capex pie chart. Our “Movement Planner” is part of TOP and is the equivalent of an air traffic control system for our railroad. It’s break-through technology that lets us run more trains over a given line segment and at faster speeds. That saves money in many ways, including capital costs.
If we can increase average train speed by just one mile per hour, we can run the railroad with fewer locomotives and serve more customers with fewer cars. You will recall that on the conference calls we’ve shown how NS has increased revenue units per locomotive and reduced crew-starts even as gross ton-miles increased.
Then we have the Uniform Train Control System (UTCS), the dispatching system that is the backbone of Movement Planner. The combined UTCS-Movement Planner technology gives us a simulation-based model of the physical network so we can set train speeds to ensure on-time arrivals.
Out on the railroad, we’ve developed a locomotive-based computer system to help crews get over the road using less fuel per gross ton-mile—and more safely. We call it LEADER, Locomotive Engineer Assist Display Event Recorder. New York Air Brake developed the system, and we use it to give our crews real-time information about the train they’re operating, recommending optimum speed given trailing tonnage, gradient and curvature, for example. This is one tool that has helped improve gross ton-miles per gallon by three percent in just one year.
RA: And, speaking of technology, what about PTC? I see you have $229 million in the 2013 capex budget. Where’s it going?
Butler: PTC will cost us roughly $1.4 billion by the time we’re done. This year’s allocation is for locomotives, wayside devices, and back office systems. Because we’ve been upgrading our wayside signal systems continuously over the last 20 years, we don’t face many of the legacy system challenges we’ve seen on other roads. The bigger challenge is the design and construction of the hardware and software needed to communicate between and among locomotives, wayside devices, and back office systems.
Right now we think we’ll want to equip 3,400-to-3,800 units, depending on the FRA rules on yard locomotives. By the end of 2013, we will have 1,700 PTC-equipped locomotives, and we’re putting LEADER on the power at the same time we do PTC. Our preference is to do it all at once, even if we have to go back and retrofit as the PTC hardware evolves. We expect to have $600 million of the entire $1.4 billion PTC commitment in place by next January.
RA: Mark, how does the change in traffic mix—more intermodal, less coal—affect your capital needs?
Manion: Fewer coal trains affect the coal corridors, mainly. On the Pocahontas Division there are some line segments, branch lines and secondary yards that have seen significant traffic declines whereas others, particularly the core routes, have increased tonnage volumes. We’ve repurposed the Norfolk-Roanoke-Columbus Heartland Corridor through the traditional N&W coal country. The resulting mix change gives us more room to increase intermodal and manifest carloads.