Berkshire wins by using a formula that relies on them owning whole businesses with sustainable, predictable returns, ensuring that they're extremely well managed, and backstopping them with a gigantic pile of cash that they can use to roll up other smaller bolt-on companies with.
YC works almost exclusively with unproven, highly speculative new firms, with unproven management teams (many of whom will, after joining YC, hire their first employee ever), owns so little equity in each firm that they don't even get a board vote, and by design avoids further capitalizing companies they bet on.
That doesn't make YC bad; the model seems to work extremely well. It just seems like a very different model.
edit: HN->YC. Embarrassing.
BH also has an extra 50 years on YC, so of course the models don't match up. And I'm not Sam, so I can't speak to whether not YC's ownership or investment models will change over the next 50 years (and yes, I think there's a good chance YC will be around in 50 years). YC is focused on testing and validating its thesis across other industries, but the thesis started in one area. Very much the same with BH.
In other words, if BH's only value is that they can successfully help grow a candy company, an insurance company, and a train shipping company, and YC is able to help grow a fusion company, a hotel industry company, and a home cleaning company, then they are both doing something right. It seems they are both heavily driven by focused principles.
Another great point is that YC and BH don't have traditional competition. See pg 31 starting
"Berkshire has one further advantage that has become increasingly important over the years: We are now the home of choice for the owners and managers of many outstanding businesses."
In what way? Yes, they both invest in companies, and the investments represent people trying to figure out how to make money and beat their competitors. But there are a lot of differences in the details (Buffet has a source of stable income that people trust him to reinvest; Y Combinator makes money sporadically when other investors decide to buy their companies). They're about as similar -- to me -- as football and bowling. Sure, both sports have balls, but there are a few relevant differences.
Compared to ordinary investors, Buffet's alpha is consistently high. He looks like a genius. But if you change the definition of alpha based on his strategy, his alpha ends up being much lower. Personally, I don't think that diminished Buffet in any way: he came up with the strategy, after all; but it does show that his investments work because he's doing something different than the competition.
And, again, Y Combinator is also doing something different than the competition (or at least, different from what the competition was doing when YC launched), but it's also doing something wildly different from Berkshire Hathaway. BH invests in a small number of mature companies, YC invests in many small and risky companies. I would be interested to see what kind of personalized alpha YC manages to get (and, even, what kind of generic alpha they have). I don't know if they publish their investment numbers, so I don't know if it would be possible to calculate.
IMHO, there is reason to believe this is plausible. Injecting my own interpretations - the proportion of startups to established businesses in an economy is dependent upon a.) the pace of change and b.) the degree of interconnectedness in the economy [because when firms are tightly connected, there is a high chance that future technological developments will invalidate fundamental assumptions that existing firms' existence is based upon, while when firms are generally siloed and vertically integrated, they are unlikely to be affected by the entry of new firms into the market]. Both of these variables are increasing rapidly today. That's going to put increasing evolutionary pressure on existing old-line industries, such that we may see a mass-extinction event in the near future where a number of prominent industries all go down en-masse.
What Berkshire and YC have in common is a full appreciation for the effect of compound interest and a willingness to employ that capital by putting their fingers in an increasing number of pies.
This is the exact opposite of YC and other startup incubators. Their approach is to invest in newer companies with great potential, but also great risk of failure. The business models of these companies are not always clear.
From a personal level (I live in Omaha), I find it interesting to contrast Berkshire Hathaway and YC from a cultural perspective. YC is located in Silicon Valley, while Berkshire has its headquarters in Omaha, Nebraska. When asked about why he stays in Omaha, Buffett said "It's very easy to think clearly here. You're undisturbed by irrelevant factors and the noise generally of business investments."
Silicon Valley’s culture is the opposite of Omaha in many respects. Companies are concerned about chasing the latest and greatest trends even if many founders don’t complete understand those trends. Startups aren’t thinking about the next 10 or 20 years, they are more concerned about their next round of funding.
The success of Berkshire Hathaway demonstrates a valuable point: Silicon Valley isn’t the center of the world. Plenty of successful companies thrive in backwoods locations like Omaha. This isn’t meant to say Omaha’s culture is superior to Silicon Valley’s, they are just different. Diversity is a good thing.
YC is just about the opposite. "Find extremely risky things that have huge upside, and invest a very small amount". Both models are great and have proven to be quite profitable. Though to me, YC seems to be much more altruistic. At least currently, they appear to be much more interested in helping people to start successful companies than they are making themselves rich(er). I don't know exactly what pg's motivations for starting yc were, but it looks like it was something along the lines of "Starting a company was a risky decision when I did it , and I want to help others to have an easier time than I did" as opposed to "Gimme gimme gimme money". As a result YCombinator and Berkshire Hathaway probably won't be in direct competition for quite some time, as they are optimizing for different results.
If anything their model seems less exploitative to me...