Amtrak posts white papers on financial performance, revenue/cost allocation

Amtrak Unlimited Discussion Forum

Help Support Amtrak Unlimited Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
How does this differ from RPAs white paper? Does RPA still support its own version, it’s sure been quiet concerning it? Does RPA agree with these new “facts” from Amtrak?

So many question so few answers.
 
How does this differ from RPAs white paper? Does RPA still support its own version, it’s sure been quiet concerning it? Does RPA agree with these new “facts” from Amtrak?

So many question so few answers.

Maybe you could help answering your own questions by reading both and then sharing your thought with us? [emoji848]

As a starter my take on this is that it is a good thing that Amtrak published this information in a consolidated form in a single document. There is relatively little that is new in this, but at least there is a single reference document to point to now. In particular it bears out my earlier statements about the way in which the system was arrived at mostly by FRA/Volpe collaboration and the fact that Amtrak is not in a position to unilaterally make major changes without breaking the claim for the big "it is certified by an auditor" claim, unless further changes are equivalently certified, and that gets us back to FRA/Volpe thing.

Roughly speaking this document lays out what the architecture of the cost and revenue allocation system is. It does not discuss the specific rules beyond a few illustrative examples. The RPA paper is more about the observed results from the use of the currently used system that is conformant to this architecture with illustration of some of the not very credible outcomes in actual allocations.

In my opinion, based on building other large and complex systems, when building a complex system like this one has to start somewhere and it is reasonable to start with something plausible, use it and make modifications and corrections to get better and more credible outcomes. It is the latter part where things seem to have broken down. Just because a bunch of auditors at some point anointed a bunch of algorithms does not mean that they should not be fixed when they produce absurd outcomes, like allocating snow removal costs to Miami.

Note that the last paragraph is an opinion based on incomplete knowledge, so is subject to revision as more becomes known. AFAICT (and maybe I have missed it skimming through the document) there is no mention of how the implementation of the architecture is maintained and updated, and what the mechanism is for fixing the architecture where it may be found to be wanting.
 
Last edited:
Amtrak very briefly describes what the APT system is *supposed* to do, in a 6 page letter with two blank pages and very few citations.

https://www.amtrak.com/content/dam/...k-route-financial-performance-calculation.pdf

It is good to know that Amtrak and RPA agree on what APT is *supposed* to do. (RPA believes that the fixed vs. variable costs should also be broken out and published separately, as required by law; Amtrak thinks it can get away without publishing those costs, despite the law.)

However, The Rail Passengers Association White Paper proves clearly that the APT system *does not do what it is supposed to do*. APT is simply not fit for purpose.

https://www.railpassengers.org/happ...e-accounting-fatally-flawed-misleading-wrong/

Amtrak does not actually provide any evidence disputing this; so I guess they agree that APT is broken. Snow removal costs in Miami, track maintenance costs in Michigan billed to the Lake Shore Limited (which does not go through Michigan), station costs being double-allocated to the Lake Shore Limited versus Empire Service frequencies (this is NOT the explicitly stated allocation method) -- these are just flat out errors in the system.

Amtrak's White Paper states that these are not what APT is supposed to do. Since these are in fact what APT is doing...

APT is full of these errors, as documented by RPA, and would not pass a real audit.

----
On another topic, I'm glad to see that newspaper reporters are starting to understand the variable-costs vs. fixed-costs issue -- Bloomberg's Justin Fox discusses it.

https://www.bloomberg.com/opinion/a...isn-t-about-to-turn-a-profit?srnd=politics-vp

He points out that Amtrak CEO Moorman told Congress (accurately) that cutting the so-called long-distance train would require about $423 million in additional Congressional subsidy in the first year alone. (In other words, the so-called long-distance trains as a group actually make $423 million per year in profit based on short-term avoidable costs. Sounds about right to me.)

The supposed losses of the long-distance trains are entirely an artifact of cost allocations, primarily due to allocating roughly a billion dollars in fixed costs to the long-distance trains. (Fixed costs should not really be allocated to anything.) The company as a *whole* requires federal subsidy because the total profit from all the trains isn't enough to cover the fixed costs, but cancelling nearly any individual train would increase the need for federal subsidy (with a very few possible exceptions).

The fact that the cost allocations are actually complete unsupportable garbage which would flunk any real audit just makes this worse. The underlying problem is that the allocations shouldn't be used at all. From a business decision point of view, fixed costs and variable costs have to be addressed using different techniques, separately.

---
It's funny that Amtrak mentions that states have an incentive to force Amtrak to fix its defective cost allocations. They do have that incentive! Having been unable to get Amtrak to fix its accounting, they are instead talking about hiring a different company to operate their trains.

It's funnier that Amtrak says that they have an incentive to not overallocate to the long-distance trains. Sadly, it's now been documented that *some* members of Amtrak management seem to have an incentive to overallocate to the long-distance trains -- so that they can badmouth the long-distance trains which they dislike and make excuses to end them.

Moorman told the truth when he said it would cost $423 million in Congressional subsidy in the first year if the long-distance trains were cut, and yet today's Amtrak management doesn't seem to want to admit this in its published "allocation" accounting. Shows that someone in Amtrak management has an incentive to overstate the costs of the long-distance trains by about a billion dollars.

I really think someone is pulling the wool over Anderson's eyes; Moorman, Boardman, and their predecessors all knew what the real avoidable-cost accounting looked like, regardless of what garbage APT put out. Anderson is acting like he doesn't actually know what the real accounting looks like, and so is proposing dumbass stuff which would increase the Congressional subsidy requirement. As a former private-sector businessman, he can't *intend* to look like a dumbass who doesn't understand economics -- someone else is making him look like a fool.
 
Last edited:
The new CFO is SO new that we need to send her RPA's White Paper as a courtesy. RPA should do it. I would do it myself but I am very busy and facing medical flare ups right now.
 
Back
Top