Monday, May 21, 2018

Railroad Renaissance or Dead Man Walking

There's a lot you can find out by just looking.  The problem is knowing where to look.

A recent thread on a Trains magazine got me thinking.  Conventional wisdom says that railroads are back.  That they're having a renaissance.  But, one fellow there says "not so fast", and he trotted out some data to show it.

Hmmm.

He had two data points about a decade apart.  I wondered what the rest of the data set looked like, so I dug into it.  I particularly wanted to know about NS - for obvious reasons.

Step one.  Get the data.  Edgar has it as NS published annual carloads and revenue by line of business in the 10-K.  Great!  Just copy and paste it into Excel, no sweat.

Actually, some sweat.  From year to year, NS arrayed the data differently.  In the older reports, they covered 5 years.  Lately, 3 years.  So, I grabbed and pulled and copy and pasted and "paste special, transposed" and "find and replaced" and cntl-c, cntl-v'd my way to 133 rows of data covering 1999 through 2017 by year.

Some basic results:


Well, okay.  Doesn't seem awful.  Coal is dying.  We already know that.  Intermodal is zooming.  We know that, too.  But Merchandise traffic seems sick.

  • Every line takes a dip in 2009 for the recession.  
  • Coal dives after 2011.  
  • Merchandise never gets back to pre-2007 level.
  • Intermodal is off to the races.
What about revenue?


This is better! 

  • Merchandise revenue is growing nicely
  • Intermodal growing, too, passing coal in 2014.
  • Only coal is falling apart.
Here's the skinny on rates:


It's RPU that's driving the Merchandise revenue.  Look at the inflection point about 2004, when the "Thoroughbred Operating Plan" took hold - NS's version of scheduled railroading.

Also, look at how flat the RPU for Intermodal is.  Clearly, it's volume that's driving the revenue.

I'm going to add one more set of charts.  A set that might indicate the overall direction of NS.  I normalized the data with two factors.  One is GDP.  As the economy grows, one would expect rail traffic to grow proportionally.  The other is inflation.  It's always good to put money inconstant dollars so you don't mistake inflation for growth.  Here's what unit and revenue look like normalized for these factors.


If the lines are flat, then business is keeping up with the economy - growing just as fast.  If the line is going down, then the business isn't keeping up.  If the line is going up, then the business is growing faster than the economy.

Parsing this out.  Merchandise seems to by dying.  Plain and simple, although it seems to have plateaued the last five year or so, it hasn't come close to the pre-recession levels.

Intemodal is growing.  Maybe.  The overall trend is up, but the past four years are flat.  This likely means that conditions aren't great for truckload conversions.  Lack of investment in new, speedier lanes, perhaps?  There are lots of trucks on I-85 and I-81 that aren't being "converted".

Coal is just plain dead and I wouldn't plan on any resurrection.



(Note that I put Merch units on the RH y axis so it was easier to compare)

If you put it all together and look revenue, it's even worse.  Merchandise isn't taking in any more money than it did 20 years ago.  Intermodal, despite the volume growth, isn't growing revenue faster than the economy.  It doesn't look like it's going to be a replacement for coal, unless the company makes some investment for a cost and speed competitive Crescent Corridor or pulls a rabbit from their hat.

And, coal.  It's just dead.  It's okay to ride it to a stop, but if it keeps declining at the current rate since 2011, it'll be all gone in 25 years.


So, is NS in a stalled renaissance?  A flat-line existence where staying in business means finding cost efficiencies every year? Or, just having a long, protracted "going out of business" sale.

Don't listen to what they say.  Look at the data!  And, look where they spend their profits.  Hint:  It's not the Crescent Corridor.

These charts make me sad.

P.S. The only factor I didn't normalize for is revenue tons per car.  If car capacity is growing, that could bend some of these lines up a bit.  I believe it is, but I'm skeptical of the leverage...and I'm too lazy to dig out the data.