Tuesday, January 10, 2012

Car Service Rules

I’m not an expert on the Car Service Rules of the AAR, but have been asked about the main features often enough that I thought a summary might be useful here.
     These rules were adopted and modified over the years by a process of consensus among the member railroads of the AAR (and the ARA prior to 1934), through the committee process of the associations. The primary purpose of the rules was to increase efficiency in utilization of the national freight car fleet. The core idea is simple: reduce the amount of miles run by freight cars during which the cars are empty.
     The simplest way for a railroad to handle the freight cars of all other railroads (called “foreign” cars) would be for the road to return them, when made empty, to the owning road, while loading its own cars exclusively. This means, nationwide, that empty miles for every car would be equal to loaded miles, because every other railroad would be loading its shipments in its own cars and getting them back empty. But since many cars are functionally equivalent, such as general-purpose box cars, why not load that fresh empty and send it homeward, but loaded? This obviously reduces empty mileage considerably, and improves utilization. If Road A loads a car to Road B, and B then loads the same car back to A, there is no empty mileage at all. Even if the car only moves, say, halfway homeward with a load, the ratio of loaded to empty miles is improved.
     There was eventually an extensive set of rules, with detailed interpretations, special exceptions, and so forth, to try and implement the simple-sounding idea of loading foreign empty cars in the homeward direction. But here is a compact summary, developed by the AAR and reproduced in some issues of the Official Railway Equipment Register. The rules are listed in order of preference (click to enlarge).

The Districts referred to were defined in a map, the details of which changed from time to time, but here is the map for 1954.

Some of the “home district” definitions may seem odd, but these depended on the presence of major junctions or interchange points, not the primary operating areas of individual railroads.
     The intent of the rules is for the owning road of a car to benefit from car movement under load if possible (meaning freight revenue), thus the rule directing that car should be loaded for movement via the home road, even if destined beyond that road.
     A further point to recognize is that although empty cars normally were returned homeward via the route they took when loaded, so that each road benefiting from handling the car under load would also bear the burden of moving the empty car without revenue, there was an important exception. If the road on which the car was made empty had a direct connection to the owning road, the car would be moved homeward via that direct connection, regardless of the route under load.
     [Much more on this topic can be found in the very interesting book by E.W. Coughlin of the AAR, entitled Freight Car Distribution and Car Handling in the United States (AAR, 1956). It can usually be readily found on the Internet through used book dealers.]
     Railroaders will usually comment, when addressing this topic, that the AAR list of rules fails to show one rule, sometimes called the “boss rule,” or “rule zero,” or the “big rule.” That rule is stated something like, “Protect the shipment.” That means, provide the shipper with the car he needs, even if you have to break all the other six rules, and you can be sure your supervision will not criticize you for doing so.
     That leads to the question, “just how seriously were these rules taken? are they really rules or just guidelines?” Several AAR studies found that the Car Service Rules were obeyed around 80 or 90 percent of the time (before World War II) in loading empties, but more like 65 percent thereafter, and recollections of employees of the transition era are consistent with such percentages. So although these were not really mandatory, they did tend to be generally followed.
     How does all this affect modeling and operations on model layouts? You can, of course, entirely ignore them and spot cars for loading however you like. But if you want to capture some of the prototype approach, at least try to load foreign empties in a homeward direction, and move direct-connection empties directly homeward.
     What is perhaps less obvious is that loaded cars arriving at your industries from distant origins should also have been loaded according to the Car Service Rules. That means that the originating railroad in some distant area, say New England for my SP layout, would probably not use its own cars, nor empties of railroads in its immediate area of the country, because those would go home on direct connections. Instead, they would be trying to load cars typical of the area you model. As an SP modeler, this means I expect to receive loads from the far corners of the country in SP, UP and Santa Fe box cars in many cases.
     In times of economic weakness, when the car supply tends to be in surplus, foreign empties probably do go home much of the time. But in boom times, when car supply can get rather tight, the Car Service Rules are usually bent to suit circumstances. This can be an “out” for those who don’t want to bother about car service; just say you are modeling a time of prosperity. Or you can do a little research and find out if your chosen modeling period, say fall of 1950, was in fact a boom time or not.
     But whatever you choose to do on your own layout, knowing something about the Car Service Rules helps you choose how prototypical you want to be in operation.
Tony Thompson


  1. Great post Tony. Just to clarify in my mind, in order for a very foreign car, say from the PRR, to show up on the Coast Line it would probably have been loaded someplace back east with a cargo for the west coast because no west coast cars happened to be reasonably available at the time. Is this true?
    Sort of makes you wonder how any est coast cars ended up on the west coast yet the study you sited earlier says they did, in roughly the proportion that they made up of the national car fleet.
    Somehow it seems to me that for the study to be correct then there must have been a lot of loading going on where cars from the far reaches weren't. I think I'm missing something here.
    Thanks again for an informative post.

  2. Jim, you ask a reasonable and not necessarily easy question. That PRR car might have been loaded from the east coast to Seattle, then loaded (vaguely homeward) to Los Angeles. Or maybe it was loaded in utter disregard of the rules, as you say, when no other car was available. That could be an instance of "Rule Zero."

    As I mentioned in the post, in the 1950s, studies showed that only about two-thirds of car requirements were fulfilled in accord with the rules. That alone accounts for lots of "unexpected" car movements; and in times of car shortage, the rules in many cases were quietly ignored, and you loaded whatever cars you could lay hands on.
    Tony Thompson

  3. The (per diem) cost of keeping a foreign car around waiting for a home-bound load must also be considered. That is, I doubt foreign cars were kept waiting for a load. If a manufacturer served by the Pennslvania needed to ship his product to a distributor in Los Angeles, the Pennsy would use a foreign road car if it had one available, but certainly would not pay per diem to keep the car available for a loaded return. This would or should decrease the chance that a foreign road empty car (from the region - SP, UP, or ATSF) would be available for loading, and increase the chance of the Pennsy using it's own cars, right?

  4. Railroads varied in how seriously they took per diem, and of course yards at or near interchange points had a different viewpoint than yards far from an interchange. There were certainly railroads which lived or died by how well per diem was managed, and others which really didn't focus on it.

    You are right that roads would not keep foreigns around "just in case," but I believe it's impossible to make a general statement. In times of car shortage, ANY car in the yard would be loaded if suitable for the cargo, regardless of the Rules, whereas in bad economic times, when cars were in surplus, much more attention could be paid to the Rules. In your Pennsy example, many different foreigns might be used for a West Coast destination, especially if the car were routed over the owner's line as part of the movement, or if the owner was somewhere in the vicinity of the West Coast. So Midwestern and other intermediate-region roads, Pacific Northwest roads, Southwestern roads, are all within the Rules, not just SP, ATSF, UP. If the Rules are followed, PRR cars would be a last resort.
    Tony Thompson

  5. Interesting! Good point on the home-road car being a last resort if the rules were followed. It never occured to me, from reading the rules as you'd presented them, that a car from a railroad that did not operate in a zone would be considered for a shipment within the zone - i.e. a Great Northern or Southern Rwy car for a shipment to Los Angelas.

    I really appreciate your time to research this, and to explain it so well.

    Thank you very much!

  6. You're welcome, Arved. For the point you raise, read Rules 4 through 6, which provide options if a car can't be routed directly to a Home District of its owner. I once heard someone summarize the Rules as saying, "Load foreign cars when you can, and send them somewhat homeward if you can." I'm sure any former Car Clerk reading this is rolling his or her eyes at such a statement, but I think you CAN read the Rules that way.
    Tony Thompson

  7. Dear Mr. Thompson,

    I know it's been over a decade but I recently found your blog and have learnt a lot from reading it!

    Also, as you might have already discussed, I presume that it was very likely for eastern cars to show up on western roads, given how much larger eastern roads were—NYC with 86,483 cars (another 110,000+ if counting those of her subsidiaries) and UP with 41,313 cars according to the ORER of 1917, not to mention the mighty you-know-who with 169,706 cars.

    1. You are entirely right. The analysis that supports this point is by Tim Gilbert and Dave Nelson, and often called the Gilbert-Nelson idea. You can use those names as search term in my blog, and you will find numerous discussions in my past blog posts of exactly the point you make.
      Tony Thompson

    2. Me too! It's a timeless topic for modelers. Great reading and thanks for putting it out there!