Outsourcing capital component production: Yea or nay?

This is a very special sort of question. It has a simple answer, which is "no". But then you follow the line of thought and go "...huh." And you start looking at some numbers, and it starts to have a different and much less simple answer, which starts with "this probably isn't what you had in mind at all, but-", involves a lot of spreadsheets (from which the reader will be spared, fortunately) and might end with a complete re-evaluation of your long term plans, although though you can already think of four reasons why it probably wouldn't work. But if it did....

...let's start with the long version of the simple answer.

The most dangerous and tedious part of building capitals for me is jumping materials to lowsec. Capital components have a larger volume (by about 30-50%) than the minerals it takes to make them, so getting components from a different system (say, jita) would increase somewhat significantly the number of jump freighter jumps I need to make. In fact, most capital builders compress minerals into 425 mm railgun I, allowing them to move about 10 carrier builds worth of minerals in one jump, at the cost of what is reported to be very slight inefficiency. I haven't looked into this yet because I'd need to grind standings to fix reprocessing tax, research railgun blueprints, probably train another manufacturing character....

But let's say that I did outsource component building, stopped building components altogether and just got them in jita.

And for the sake of this discussion, let's say that I'm limited to the 30 manufacturing slots I currently have, and that a capital ship hull takes 10 days to build. I currently have 9 capital ship hull blueprints, and under optimal conditions can keep them all building at all times, which comes out to a theoretical maximum of 27 ships per month. This is pretty standard for a serious capital builder; the other two I know report building 21 and 28 ships per month.

First, we have to know if it's actually possible to get components at a reasonable price. Comparing my spreadsheet to jita prices for capital armor plates, I find that plates in jita sell for 5148 thousand isk, while at jita mineral prices it would cost me 5107 thousand to build one run. This is a difference of 0.8%, which is negligible.

(related: armor plates are probably the worst possible component to look at for this purpose)

So if I were to put in a bit of extra work, I could avoid having to build components. That in itself doesn't really do anything for me, but if we take the 'buy components, build capitals' idea to its logical extreme we get to some really interesting places. What if, with no component bottleneck, I could build capital ships in all my manufacturing slots? 30 slots at 10 days each is 90 ships a month. Which is, like, a lot.

...and obstacles at every step.

Second first off, capital BPO. This isn't actually much of an obstacle, but buying and researching BPO would cost about 30 billion isk and take about a year. Isk I don't have right now, but this is eve. Isk happens.

Second, liquidity. If I have 30 capital ships building at all times, I will probably have, at a very rough estimate about 60 billion isk of material (components, hulls) wrapped up in the manufacturing process. This is doable, but probably not until I can get rid of my supercarrier BPO.

Third, buying components. Using armor plates as a standard (as mentioned, probably a very bad standard) there are about 300 units available in jita for a reasonable price, and they move about 50 units per day. That's enough to build 4 capital ships per day. I would be building 3 ships per day, which would increase demand by at least 75% (probably more, since some of the movement is traders) and probably drive prices up enough to make this idea not profitable.

There are a few things I might try to mitigate this. One of them is people who build capital and supercapital component packs -- with the supercapital nerf there's a good chance these guys are facing a decrease in buyers, and may have spare component building capacity that I could yoink for a reasonable price.

Fourth, selling. The forge market sells 280 capital ships per month that aren't freighters, orcas or jump freighters. Demand for capital ships is inelastic to price, so dumping another 90 ships per month on the market will drive down prices, probably in a significant way. I would probably try to mitigate this by spreading my operation to multiple regions around highsec -- probably the amarr, rens and jita regions, plus maybe one or two others (I hear good thingsabout lonetrek). Conveniently, this would also make it easy to spread buying of components to the other market hubs.

HNNG.

This is all pants-on-head theorycrafting, but if somebody could make this work it would be pretty awesome. To the tune of 19 billion isk of profit per month, under perfectly optimal conditions.