On Medici and Thiel

We should radically scale genius grants

I think the single most misdirected bit of philanthropy in this decade is Peter Thiel's special program to bribe people to drop out of college.

Larry Summers

Peter Thiel has a lot of ideas. One of them, from slightly more than a decade ago, was that entrepreneurship is a viable alternative to college. So he decided to spend around $2m annually to sponsor 20-30 kids to drop out of college and do something.

It's been a decade, which means around $20m invested in total, which gives enough room to perhaps draw some conclusions from this experiment. What did he get for spending that money? To start with, he found these chaps:

  • Vitalik, founded Ethereum

  • Austin Russell, Luminar CEO

  • Kaushik Tiwari, Better Financial CEO

  • Dylan Field, founded Figma

  • Ritesh Agarwal, founded Oyo

  • Alex Rodriguez, Embark CEO

Now these are only a few, out of 217 that I found, but it's a pretty great selection ratio. Just measuring with the regular VC yardstick, that's an enviable hit ratio even keeping Vitalik aside. There's also Loom, Scale AI, Upstart. And a whole lot more once you look at the rest.

Considering it's a $100k grant per fellow, that's remarkable success. Even taking Larry's point onboard from the top, which can charitably be taken to mean that this doesn't help everyone, you would be hard pressed to get a similar ROI if you surveyed an Ivy League class.

My first question was, why is this such an isolated example? Surely there are more than couple dozen smart people every year in college who can be induced to drop out and build things.

My second question was, isn't this the same as old timey patronage? What happened to that anyway?

A short detour. A few hundred years ago, Cosimo Medici kickstarted a financial empire. At that same time he began a grand tradition of patronage, of Donatello most notably. He kickstarted a tradition where his kids (including Lorenzo the Magnificent, which has to be a nickname he gave himself) provided patronage to Leonardo Da Vinci and Michelangelo.

One could also draw more farfetched analogies. Of how Lorenzo also tried (and succeeded) to rule Florence through payoffs and strategic alliances. And of how he built and expanded a giant financial empire!

Now the dynasty sizes differ, but the actual patronage itself remains eerily similar. I mean, I don't think Vitalik was living with Peter for a year or three, but $100k and some prestige sure could be the modern equivalent of room and board and a stipend.

There are oddly few modern examples of individuals funding individuals, despite it being a far more decentralised and human-scale version of patronage that's easily achievable by large groups of people. You don’t even need to set up a thinktank!

I'm not saying that the Renaissance geniuses who came together in great concentration during Medici's reign are the same as the Thiel Fellows. What I am suggesting is that the idea of patronage seems to not have transmitted all that much through the centuries. What seems the simplest trade-off; I pay some of my enormous sums of money to get exposure to smart people, and increase my social stature and secure posterity by helping them achieve greatness, seems to have disappeared.

Part of it is that we've conflated richness and achievement. Patronage would seem counterproductive if you didn't believe in the genius of others. In fact you have to believe that your job, so to speak, is to encourage others, and not to guide or direct or actually produce. Elon Musk taking on the challenge of creating tunnels is great, but it's hardly patronage if you do it yourself.

It is also true that it's such an absurdly small amount of money that Thiel has spent. Altogether, for the 217 participants, Thiel shelled out around $22m. Add another 10-15% for administrative expenses, call it an even $25m, maybe more for kombucha and massages so $30m. To compare this, he individually contributed $20m just recently to the Senate races in Arizona.

For ROI on spend, what's even remotely comparable? Having the creator of Ethereum and a bunch of unicorns see you as their patron, or being one of the contributors to a challenger's Senate race?

For comparison, a top 20 funds chart from a couple years ago. I’m sure plenty funds are doing better than this today. But Ethereum alone has almost a $300B market cap, and owning a couple percent of that would’ve itself given Thiel Fellowship of that year a 3000x return. I don’t think we’re anywhere near the efficient frontier here.

The idea to split us all into mini-Medicis through platforms like Patreon, if anything, has only further diluted the actual value. It's not just about the fact that they financially support you (that's called a customer, or a customer who tips generously), but that they actually help grow your potential by using their status and name and resources (which might include money). Mini-Medicis don't fulfil any of this purpose.

Whether it's the rise of Substack or Tiktok or Youtube or the general rise of the creator economy, none of this fills the actual role a patron would've performed in the past. Patronage doesn't seem like a tradition that can be easily disaggregated. Once atomised, you end up with work that's much less iconoclastic, deteriorating the action into tipping.

So while artists like Randall Munroe from xkcd exist, helped along by the power of fans coming together, they're still being paid for creating something the consumers value. There must be so many other Munroes who remain unfulfilled in their NASA jobs but would never embark on this adventure because their talent for stick figures will, alas, never be recognised. It's incumbent on us to fight against this market failure.

This phenomenon also seems true because we’ve stopped providing patronage to selected individuals, and instead started funding institutions or entire movements, like the Koch brothers or Soros. In trying to scale it up we've robbed it of its actual purpose.

Lorenzo Medici, it must be noted, spent the modern equivalent of $450m over the years in charity, buildings and taxes. That's a hell of a lot of money to spend, including on buildings and institutions and museums and other aspects of public goods. I don't know how much of that was eaten up by Leonardo & co on food and housing and marble, but it can't have been more than a tiny fraction.

What about institutions like YC that exist today? They surely exist, at least at the earliest instances of a company formation, and provide roughly the same amount of money ($125k). Shouldn't that count?

Well, it does, sort of, with a couple caveats. What they provide is financial backing and an imprimatur of legitimacy for a team with an idea. Which is great, but its not the same as identifying individual genius to try and foster it. Institutions do grow and over time provide legitimacy and certification to those it selects, but that's not the same as a patron giving a relatively free rein.

Individual patrons are far more risk seeking than organisations, especially organisations at the later, scaled, stages of its lifecycle. If there's anything that's particularly emblematic of this problem it would be the Ivy League universities. Extraordinarily prestigious but extremely ossified. YC probably isn't there yet, but then they're only a decade and half old!

For another example, take Emergent Grants led by Tyler Cowen, which burst onto the world with high impact last year. Using speed as its defining characteristic, they have tried to redefine what grant provision actually means. Several brilliant projects and several brilliant researchers have all benefited from this approach, with many a life saved during the Pandemic.

It’s obviously fantastic, but if anything this doesn't go nearly far enough. The fellowship is loose by nature (and by virtue of the pandemic), and the amounts are more encouraging for the individuals than an invitation to quit pursue their dreams, some being awards more than an invitation. It's often closer to looking at those who are already pursuing their dreams and providing them with a stamp of approval. We need a hundred more of these!

Also, frustratingly for us and pushing ahead the great stagnation, none of this is even remotely similar to the actual way that Galileo or Kepler got their patronage. An identification of their talent with an implicit understanding that pursuing their work is what's asked of them is key. They had duties for sure, tutoring the occasional princeling, but hardly something that fundamentally distracted them from their research.

The moving of art and science from the realm of spirited genius amateurs to accredited professions meant that the funding institutions matured too. No longer did they need to rely on the largesse of the occasional duke to get things done, they could rely on actual public government funding and longstanding institutions to pursue things like art and science.

The modern version also seems to revolve a fair bit around putting your name on buildings, mostly museums and university departments, but little beyond that. There is no movement from the larger scale to the human scale. There's little interest in the actual details or the specific people who are supposedly leading the research work. The aura that comes from being a renowned patron has now separated from the actual patronage.

It is precisely this increased institutionalisation of what used to be more eclectic that's rubbed away the corners of that made it interesting. A substantial number of top scientists across fields claim that they would never get past their respective doctoral/ departmental/ review committees today, and this reduction of iconoclasticism is a miss.

The closest we've come to breaking the mold might be Entrepreneur First, helmed by Matt Clifford and Alice Bentinck. They are a "pre idea" accelerator, bringing smart people together and encouraging or helping them to start companies.

This is great, though it still acts primarily as a sort of summer camp for entrepreneurship hopefuls, with a tight deadline in which they need to form teams, crack ideas and go to work. Goal orientedness is kind of the name of the game and the grant itself is de minimis.

If the program was ostensibly longer and/or more diverse in its intake this could be the very ticket!

There's a very clear distinction often between the types of capital people often are able to gather from their work. They usually get either Social capital or Financial capital. This isn't some great wall separating the two, but they're two sides of a impermeable membrane. People translate social capital to financial (ex politicians) and financial capital to social (VCs ...), but still the difference between the two types is often stark!

You could quibble with the placement of various orgs on that chart, but the key problem is that the top right quadrant is frustratingly empty! We need a Givewell for top talent.

Thoughts like Navals are popular, especially amongst those who actually do have money. Maybe they would be better spent using some of that money to try identify talent who would come up with these world changing ideas.

In 1981, MacArthur Foundation created a fellowship, that boasts the following:

the fellowship is not a reward for past accomplishment, but rather an investment in a person's originality, insight, and potential

That does sound exactly like what we need. With a large enough prize of $625k over 5 years, and the social prestige it confers, it provides a perfect route to pursue any ambition its recipient could want.

MacArthur grants come closest, but it's still fairly opaque and, more to the point, tiny! They’re the same size as Thiel Fellowships, 20-30 people annually, which means the cumulative number of Fellows is still sub 1000, and that’s over 40 years! And it still boasts Stephen J Gould and Tim Berners-Lee and Susan Sontag and Irving Feldman and dozens more who are also incredibly accomplished.

What we need is the YC of talent identification and encouragement. We either need Thiel to step up and 100x his fellowship, or a 100 other billionaires to step up and stop whining about talent.

We need more actual patronage, ones that provide both social and financial capital, and encourages more talented folk to pursue their vision. We need to try and figure out how good our talent recognition is and can be, and work monomaniacally on making it better. Any vision of the future that doesn’t have it would be one where we’re squandering our potential.


A few possible objections I looked at below:

Objection 1: We already do this, in all investments for instance

Yes, we do. But when talent becomes part of the process, as opposed to most of the process, we end up making decisions on their proposed theses as well as the talent in executing those thesis. Which is a sufficiently different ballgame! It also pushes the talent to come up with ideas that are fundable by the the money providers.

There's another program that's somewhat similar, called SciFounders (which actually seems awesome). But even they have an idea + talent view of things, making the grant more akin to YC.

Also, there's a reason tenure as it stands is not dependent on the specifics of what they're going to work on in the future.

Objection 2: We suck at finding talent, see our university admissions/ startup funding/ politicians/ novelists/ elites

That's totally fine. Even with it being horrible, we don't need a 100% hit rate. All you need is for the PV to be positive, and there's an enormous amount of room to play between Thiel Fellowships and hitting zero.

So a whole lot more room for talent recognition exists until we hit our societal limits. When we hit that Pareto curve and start funding mediocre corp dev folks from IBM well have gone too far. Until then we have room to play!

Objection 3: We'd just be encouraging indolence if we fund talent without an output

For some folks, probably. But almost by definition, you have to swing hard for the fences to get out of distribution outcomes. Ask any early stage VC.

Plus what gets measured gets managed. Have you seen the hundreds of articles on what it takes to get into YC, what type of job market papers to write to get published, what sort of novels get published, and more? Artificially restricting even a proportion of our human capital to try game the system seems silly.

In sciences, and literature, and cinema, and TV, and entrepreneurship, we have the problem of lower than full capitalisation levels. While we don't need to maximise every metric, the consensus seems to agree with the data that there's more room to maneuver here.

Also, laziness brings creativity, which kind of is the point of the whole thing.


If you liked this you might also like: