Speaking as a member of the the "batch that broke YC (s12)", I think YC's resources are best utilized by continuing to improve the quality of the program at its current size. Read: http://ycombinator.com/w13smaller.html
Obviously the quality is already very high, but still quite a ways to go for each batch consist of "50 AirBnBs", which to me is a far more impressive goal than having 1000s of lower quality startups. It's also better that YC is a signal of success than "not-YC" being a signal of failure. Franchising would dilute the brand and more importantly be a distraction from (as YC would always say) making and improving something people want.
YC is relatively "open-sourced" and there's nothing to stop others from creating accelerators with similar models around the world (and in the valley). There are already hundreds of imitators and these catch many of those that YC misses.
I'm saying this politely - so don't bite my head off - but
a.) you would say that, wouldn't you, you'd like as much of PG's attention as possible
b.) it's quite presumptious to state that the 1000s of others would be lower quality. Are the less than 1000, mid 20s, mostly male, mostly american, applicants to YC the best candidates on earth?
There are probably 1000x, if not more, capable founders of all age groups, nationalities, and genders that PG could get a slice of. He's built a brand and should seek to capitalize on it before someone establishes a superior brand.
Off course, this all about what PG does for fun, he's already rich enought to never work again. It might be that he enjoys mentoring a few founders a couple of times a year and doesn't have global ambitions.
b.) The startups being of "lower quality" has only to do with a watered down selection process. If you were to go to a YC dinner or demo day it would be immediately obvious that while there are plenty of mid-20s, male, american founders there are also plenty that don't fall into those categories.
Harvard "built a brand", but I don't see Harvard establishing franchises around the world to capitalize and satisfy it's "global ambitions". It's satisfied with trying to be a better Harvard, which is why it's brand persists.
In the same sense, YC could scale by getting more of those 1000x capable founders to apply the regular way and continue to improve on picking the best of them.
My argument was against "franchising" in the sense that the franchises would be top-level equivalents of YC, a demand on the time of YC partners and a dilution of the network.
I think YC does extend its' brand in a different way: by funding companies that create new avenues for startup funding/networking such as FundersClub.
I see your point, PG could focus on relentlessly increasing the quality of each batch - if his aim is too just establish a high quality brand for no other reason than he likes to (the elite college model).
However, from a purely profit perspective, PG would stand to make a lot more profits from scaling. After all, Walmart is younger than Harvard by hundreds of years, but makes a lot more money.
YC's biggest benefits - the network and the credibility with investors - is somewhere between hard and not possible to replicate everywhere under a franchise model. Lots of incubators all over the world based themselves on what YC pioneered mostly with lackluster results.
If there was a YC here in Costa Rica the startups would be heading to SV to raise money. Multiply that by dozens of other countries that can't provide the complete ecosystem and the name becomes worthless and the network no more beneficial than HN.
Well, I don't know about that. A lot of that is based on the PG brand, which the other incubators don't have. PG and Buchheit's connections come in to play when selling shares at first round and after. Their handshakes count more than whether the startup's based in beijing, boston, or bangalore.
Obviously the quality is already very high, but still quite a ways to go for each batch consist of "50 AirBnBs", which to me is a far more impressive goal than having 1000s of lower quality startups. It's also better that YC is a signal of success than "not-YC" being a signal of failure. Franchising would dilute the brand and more importantly be a distraction from (as YC would always say) making and improving something people want.
YC is relatively "open-sourced" and there's nothing to stop others from creating accelerators with similar models around the world (and in the valley). There are already hundreds of imitators and these catch many of those that YC misses.