Yeah, but Bungie was 3.6bln, ABK was 70. I know the girl is hot, but still…
To be exact Bungie was $2.4B. Plus up to $1.2B in bonuses that partially weren't paid because some were retention bonuses for people who left or got fired, and others should be performance bonuses that at least partially may not have got because Marathon got delayed and Destiny 2 didn't perform after acquisition as well as they originally estimated (Sony recognized an around $200M brand value difference for Destiny 2 in their finances).
The problem is twofold
1. Singleplayer games are mismanaged, thus making them riskier than need be. There is no reason why these games should take as long as they do and cost as much as they do.
AAA games of all companies (not only Sony), both MP/GaaS and non-GaaS/SP, need more workers and more development time (so higher budget) every generation because their visuals get more detailed, and the games get larger and more dense and complex.
The companies who release games more frequently is because they have more teams working on different projects at the same time (and/or more people per team) working on games. Or because they're sports games with minimal changes versus the previous one. There isn't any magic formula to ship AAA games faster.
It isn't mismanagement, it's because companies do what it sells in the AAA area and what it sold has been to keep making them with better visuales, longer and more detailed.
2. The risk of gaashit games is understated. Yes, revenue and playtime of gaas games keeps going up. But it keeps going up to games that already exist. Only a small piece of the pie growth is actually going to new games, and it's damn near a crapshoot as to what those new games will be. So you not only have to spend on a gaashit game but you have to be one of the lucky ones that finds an audience.
Yes, like SP/non-GaaS titles, GaaS compete versus new games and old games. And the room for new -both GaaS and nonGaaS- games keep getting smaller as the playtime and revenue spent in old games (mostly GaaS) released in previous years.
The difference between GaaS and non-GaaS isn't only that when succesful GaaS make more way money: the most important difference is that they keep doing money steadily and predictably during several years, while non-GaaS make like over half of their revenue in their first handful months.
This means a company that relies on non-GaaS has to be frequently releasing a higher number of successful AAAs to keep a steady volume of high revenue to keep being sustainable, while a company that relies on GaaS needs to have less successful titles to have the steady amount of revenue they need, lowering the risk (remember each AAA, both GaaS and non-GaaS, nowadays costs over around $200-300M, and that a portion of them as happened since the 70s get cancelled or tanks).
This is particularly important in a context of playtime and revenue for new games getting smaller and smaller, plus with the revenue needed to make a AAA game profitable keeps getting bigger and bigger.