Commentary: Hiding in plain sight – railroad KPIs suggest continuing weakness

This market view column is about using different rail freight performances as outlook markers towards 2021 rail freight business.

(Photo credit: CSX)

KPIs are shorthand for linked key performance indicators. Not all railroad industry KPIs are internalized data. In fact, the best railroad supply chain KPIs come from non-railroad sources.  

Internal railroad data is important. But its value appears when cross-checked with other sources. Here, the prime research source is the Association of American Railroads, or AAR, monthly report card.

While a trade association is normally a promoter, the AAR has a highly educated and skilled economics group that probes both internal industry data (much of it confidential) and then directly addresses that data against a broad range of non-industry data. 

The result is a monthly report – Rail Time Indicators. This report provides hard trend evidence that I have shared during my long engineering and management career with students.


Let me share it with you.

Over decades, AAR has created excellent statistical and graphic coverage of railroading as a business. To subscribers it then relates that to the rest of the commerce data.

No other global or national association of railroad operating companies provides an equivalent continuous measurement of rail industry-wide commercial activity straight up against other trending and changing economic indicators. Not on such a rapid after the last month fashion. This AAR resource makes studying North American freight an interesting and up to date opportunity.

Therefore, when you are asked to comment about the business volume outlook, this is the place you should start.


Here is this week’s first question. How does AAR see the market trend for North American freight?

Let’s start with AAR’s end of October 2019 status report, which makes this clear statement: “A combination of a weak domestic manufacturing sector, feeble economic growth abroad that’s limiting exports, continued trade spats that are disrupting global supply chains, and general economic uncertainty are creating strong headwinds for U.S. rail volumes.”

Few economists disagree.

Here are the internal railroad industry KPIs. In round numbers:

October total U.S. rail carloads were down 8.4% year-over-year. That marks the ninth straight monthly decline.

Only four of the 20 carload commodity types that AAR tracks had positive gains in October. The markets with the largest volume declines were coal (down 15%); motor vehicles and parts (down 10%, due primarily to the United Auto Workers strike at General Motors); crushed stone/sand/ gravel (down ~ 7%); and grain (down 6.0%).

Take coal volume out – and U.S. October carloads were still down by just over 5%,

Here is a graphic summary of the years through October between 2013 and 2019. 


How bad is this?

This newest score carload drop is “the biggest percentage decline since January 2016.” These multiple three- to five-year slope changes are important to watch. Do not get lost in the weekly change numbers. That is “noise.”

Meanwhile, the next “star” for strategic rail traffic growth was supposed to be intermodal. However, intermodal continues to stumble. That is bad news when the keystone of growth hits a wall.

October 2019 intermodal volume was down just under 8%. A combination of pricing and competition from trucking KPIs has stalled the rail intermodal “take trucks off the highway” plan.

Over a three-year period, that’s the biggest percentage decline in intermodal since the great recession period of January 2009. Ouch! Are we paying attention? Follow the changing multiple year slope changes in this AAR seven-year graph.

Given this graphic’s shifting patterns of trend lines, what should your tactical car buying decision be – remembering the implications of long-term strategic cost of capital? Please think in both timelines.

Focusing on the U.S. railroad view

The AAR summarized the state of the U.S. economy by quoting a New York Times analysis from November 2: “The October jobs report and other recent data paint a picture of a job market that is weathering the storm of trade tensions and a cooling global economy. The manufacturing sector is in a slump, business investment is falling, and overall economic growth has slowed this year.”

Upon further review, this railway economist agrees.

The AAR believes the Purchasing Managers Index provides a longer term specific KPI to watch for freight.

A slightly different view is acknowledged by the AAR research staff in this graphic – the offset is that U.S. consumers are still spending, while employers are still hiring.

But can consumer consumption support rail freight? And if so, for how long?  Those are big and unanswered questions.

Will there be a full-blown recession in the United States over a period of two quarters or more? The railway data does signal “a recession ahead.” But is rail freight as relevant a recession predictor as it was in the past decades?  Maybe. Maybe not?

The third quarter U.S. GDP might have dropped to about 1.9%t from the previous two quarters’ higher percentages. Will the fourth quarter 2019 GDP drop below the third quarter? We may not know the answer until late January 2020.

Tick… tick… tick… with 13% of the weeks to go. Practically speaking, there are only six to seven good weeks left in the year for rail traffic to show signs of an economic recovery.  

Courtesy of the AAR, here are a few of the most critical economic graphics that experience suggests show the trending economic outlook for the U.S. economy. 

When you look at these, please pause. Eyeball the changing long-term versus most recent two to four period patterns of change. Imagine the slope of the declining/increasing trend line. 

That pause is your moment to capture the forecast insight for your judgement.

The past is not necessarily prologue. True. But ignore the past and the slope pattern at your due diligence risk.

Solutions to improving both share and earnings?

How might the U.S. railroads react to these trade wind doldrums?

Start here. In a down market with hard movement assets on hand, Economics 101 principles suggests marginal price selling.

Translation? With nearly 23% of rail freight cars in storage, one action to generate modal shift demand would be to selectively discount traffic rates on higher margin lanes.

To execute this, rail costing systems should calculate the difference between empty car movements and loaded car movements at a rather precise movement level.

Not every railroad company cost system does this easily. However, the objective is simple.

Try to maximize car loaded time and loaded distances – plus increase the total revenues earned by fleet space technically available to sell. 

That generates more net cash flow if done accurately. Why? Because freight cars owned (or leased) are in short-run periods almost liked fixed costs. The equation is to attack the variability function of marginal added units that could be loaded, multiplied by added revenue per fixed cost car platform.

Rod Case of Oliver Wyman has repeatedly emphasized this in his Rail Equipment Finance presentations. His strategy is complemented by Alta Rail Technology’s Dany Maron. Mr. Maron’s profitability perspective in part looks at stored or fixed fleet size assets as opportunities for generating higher cash flow. 

Freight cars are really the platform required to move more freight. North American railroads (railroads in the USA, Canada and Mexico) have spent decades improving track asset capacity. But the asset that can generate more volume and more marginal return is the freight car.   

Heavier loadings per year from the current cars directly increases cash flow. Cash flow is a true measure of earnings even in these tough times.

Graphic courtesy of Ata-rt.com Dany Maron

Otherwise, railroaders will sit, watch and wait for s random trade winds shift. 

Wrap-up

Experts calculate that the nominal North American freight railcar fleet replacement is in the 40,000 to 45,000 railcars built per year rate. That’s a sound expectation.

The most recent lowest level of railcar replacement occurred back in the Great Recession, when railcars delivered dropped to about 18,000 units during 2010. 

Susquehanna Financial Group’s Bascome Majors sees the current 2019 rail market this way. 

Going into the fourth quarter of 2019, the traffic level is clearly down.

Admittedly the full-year 2019 might still record a rather healthy delivery of 60,000 freight railcars. Why so high? Because the manufacturers are working off previous years’ backorders, when projections by the railroads were more optimistic.

Looking ahead into 2020, the actual new railcar orders might drop below the calculated fleet replacement level.

Two exceptions exist. One is grain car orders, which could possibly surge as early as the first quarter of 2020 if rumors of “huge Chinese grain orders” in fact occur due to a new U.S.–China trade agreement.  

But beware – huge gains in grain orders is currently speculative. If a railroad orders a railcar with a 35- to 45-year lifespan, the potential grain deal with China should have a similar “long life certainty.” Why? Because the capital investment or financing for that railcar will last a long time…   

The second upside exception is that specialty tank cars to move chemicals, plastics, liquified natural gas, crude oil, etc. could be the next sector favored for high railcar orders. Indeed, that is a likely rail growth sector.

The downside? TTX, a pool railcar company, still faces a dilemma in determining the future of both the slow-growth boxcar as well as the strategically expected high growth intermodal car market. As of the fourth quarter, it’s hard to see the collective railroad company owners of TTX approving any significant new intermodal or boxcar orders for 2020.

Until loaded car cycles per year increase by a magnitude of 20% or greater, the future of the boxcar fleet remains negative. 

There is no high rate of return for boxcar owners. Perhaps it is time for boxcar-dependent users like paper products shippers to invest in managing their own shipper boxcar pool? Buy a large percentage of the TTX boxcars and create a shipper-managed and financed pool of boxcars. That might protect the shippers’ needs.

Finally, as a reminder, stored railcars require both continuing debt and interest payments by someone. Even leased railcars have monthly payments, as well as a “parking fee” now.  Entrepreneurs turn such lemon-like circumstance into business lemonade.    

Who does this well and how do they do it? A FreightWaves SONAR graph illustrates the differences in freight car values that, regardless of good or bad times, is financial risk.

Source: FreightWaves SONAR

Acknowledgements

For more about the AAR monthly graphs and critiques, contact Dan Keen in the Policy and Economics Department of the Association of American Railroads.

For more detail about maximizing rail car assets reach out on LinkedIn to Dany Maron at Alta Rail. Or separately, reach out to Rob Case, a partner at Oliver Wyman. 

Exit mobile version