From a David Weber post to Baen's Bar Snerker's Only dated August 15, 2009:

Costs of the Prolong Process

    The cost of prolong (or the thread which would not die). Does anyone recall that I specifically said in my earlier post about the costs of prolong that I wasn't going to give a specific number for exactly how much it cost? I used the number $500,000 in an example to demonstrate that even at that cost the investment would be fairly easily recoverable over the lifetime of the recipient, not because that's how much it actually costs. If you will recall, that specific number was cited in a previous post in the thread, which is the reason I used it. I did not say at any point in my response that the actual cost on a per-recipient basis was remotely that high. I said it was high, and that I didn't have any problem with comparing it to the start up costs of a restaurant, but I didn't give a specific number, I didn't say what the start up cost of the restaurant would be, or even what sort of restaurant I was talking about, or anything of the sort. In case I'm not being adequately clear here, the cost is not $500,000 per recipient. It is expensive, and the therapies - while not required for the entire life of the recipient - do have to be administered and monitored over the space of a couple of decades if they're going to have maximum effect. So, yes, what Anton was offering Steph represented a very large number, especially for a seccy from Mesa. And, yes, the total cost, out-of-pocket, for third-generation prolong, and the support therapies that go with it, for a noncitizen of Beowulf, would approach the cost of opening a small restaurant. I'm not talking about a five-star French restaurant in Manhattan, obviously. I'm talking more about opening a Ruby Tuesday's in St. Louis, say. Or even a small family restaurant, like my favorite Thai restaurant right here in Greenville, South Carolina, which was opened by a very hard-working family of immigrants (like Steph) on (well) under $100,000.

    Now, the argument has been made that given the obviously stupendous cost of prolong, providing it to the Talbott Cluster's population is going to bankrupt the Star Empire, especially in the middle of a shooting war.

    Go back and read the books. There is some extensive, specific discussion in At All Costs about one specific star system - Dresden - where it is not available at all. I never said that it was available on Nuncio (which it isn't, if you go back and read Shadow of Saganami), but it is available - now - on all the worlds of the Rembrandt Trade Union, on Kornati, on Montana, on New Tuscany, etc. I'm not saying that it's universally available on all of them (although it definitely is on the RTU worlds and Montana), only that the basic infrastructure already exists on those planets and can be expanded upon. In the case of Dresden, where the capability doesn't exist at all and they have several billion people to deal with, there was extensive political horse trading involved - as you'll see if you go back to the discussion between Gwen Archer and Helga Boltitz - which ended up with the understanding that Manticore would get prolong available on Dresden as rapidly as possible and that the deal would be sweetened with extensive economic incentives from the rest of the Cluster (and the Star Empire) specifically because it wasn't going to be possible to do it overnight. In other words, this isn't a case of the Star Empire suddenly having to provide prolong to all of the teeming billions of the Cluster, nor a case of the Star Empire having promised it would wave its magic wand and make it available to the people who don't already have it First Thing Tomorrow Morning.