@Black_Elk I don't think this topic really fits the Maps & Mods category, as it is not about a specific map. What do the admins think? Should this thread be moved somewhere else?
Regarding the silly phenomenon of having "those patchwork checkerboard expansion patterns", I agree that, for example, it is really silly that in v3 the Italians may conquer a piece of Russia, and that they go ahead declaring it Italian territory, cause they took it (I can quite see Adolf Hitler face there). This would never happen, in practice, that the Italians go and conquer a piece of Ukraine and declare it Italian territory. The matter is even more laughable in maps like World At War, where you can see Russia turning into a chaotic mess of German, Italian, Romanian and Finnish occupation zones, like if it was the Scramble for Africa, and often the Germans getting a rather small piece of the cake.
However, I don't really see a premium production for contiguous direct ownerships as a good, or even workable, solution to this issue, and it really wouldn't address the actual problem at its roots.
Of course, this matter is, by definition, solved in the moment in which, for example, you make Romania a minor of Germany, so that all that Romania produces are German units, that will simply go and conquer in the name of Germany (and you cannot even tell them apart from the actual made-in-Germany units). This is substantially a lame solution, or not really a solution at all, that is closer just not having such other players at all, like in the basic games, where Germany would just be all that is Axis in Europe and Africa. Consequently, this is not going to solve the issue in the moment you want to define the various Axis powers more in detail, as you are probably not going to make Italy a minor of Germany, having and producing only German units.
Maybe someone here would suggest creating predefined spheres of influence, so to practically have up to two instances of original ownership, one for each alliance. So that, for example, no matter who conquers Ukraine, it will always go to the Germans (it would be practically like saying that Ukraine is originally owned by both Russians and Germans, all Allies liberating it for the Russians, and all Axis "liberating" it for the Germans). However, I don't like creating pre-defined sphere of influence, to determine what conquers what, because, at the end, as we have seen in Europe, it is mostly who takes takes (and then some bartering). I wouldn't even trust the Germans actually liberating Italy, if they would take it back from the Allies (pretty sure they would set up a nominally independent puppet state)!
I see two possible solutions only, to address the matter to a good degree:
The first, and likely more complex, solution would be to allow for multi-powers attacks, where units coming from a territory owned by an allied can attack together with such ally, the owner taking for itself whatever gets conquered out of it. This would likely need to be coupled with rules enforcements against can openers (for example, blocking moving-through territories until the start of the turn of the new owner), so that you would, for example, use the Italians to reinforce German attacks, the reverse being impossible as long as the Italians don't have any ownerships next to a Russian territory (so this would strongly or totally limit the "patchwork" phenomenon). Of course, this would still allow players to turn the Germans as Romanian reinforcements, and conquer all Russia on behalf of Romania, instead, if starting at a point when Russia and Romania share a border. This could be solved by just defining hierarchies, in that, for example, Romanians can reinforce Germans, but not vice versa (I cannot see Hitler putting his armies under supreme Romanian command anyways).
The second, and likely easier, solution would be simply having an option or property (or both) for determining that the conquering player is only secondarily the attacking units owner, but primarily the owner of the territory whence those units came from. This way, for example, if Italian units conquer East Ukraine coming from a German West Ukraine, East Ukraine becomes German too; or if German units conquer Egypt coming from an Italian Libya, Egypt becomes Italian too. This would need to define what happens in case not all units come from territories having the same ownership or all of them are sea borne (I would simply say that in this case the conquered territory goes to the units owner). Of course, this would create arguably wrong situations, as well, as there would be cases in which it may make some, or even more, sense for those units to conquer the territory for their own power, rather than for the one they were guest of, but changing the engine behaviour to this dynamic (as an option, of course) probably would make much more things right than wrong, thus improving historicity and realism, especially for maps having many contiguous allied players, like World At War or New World Order (of course not talking about changing those, but allowing new maps of such a scope to behave more sensibly, instead).