You don't have to invoke magic or time travel to explain Renaissance Technologies and their amazing track record. First-mover advantage suffices. (1/N)
Consider this sequence of events:
- Rentec discovers a persistent source of mispricing in the market.
- Rentec trades *fast* to capture this mispricing.
- Rentec's activity forces prices to converge. (2)
To external observers, the mispricing never arises. Nobody even sees the opportunity; they all think "oh, that (slice of the) market is already efficient". Back-testing on this opportunity reveals no alpha to be captured. Everybody moves on. (3)
Meanwhile Rentec harvests consistent returns year after year, on trades that nobody else would ever think to do. The only catch is that there's a limit to how much money they can make -- but they're okay with that. (4)
All you need for this equilibrium to hold is for Rentec to have been in business longer than most (true), with historical data deeper than most (true), execution that's faster than most (true), and a willingness to cap their AUM at the size of the opportunity (true). (5)
Everybody thinks Rentec is unfathomable voodoo. But what you see is precisely what you'd expect if they were just a little bit smarter, just a little bit earlier, than everyone else in quant finance. (6)
1/ Let's talk about quant investing! People tend to label a lot of things "quant", but this muddies the very significant differences between various quant approaches.
2/ An options market-maker, a systematic long-short fund, an HFT platform and a trend/reversal macro trader are all quants, but they do very different things.
3/ Instead of lumping them all into the same bucket, I find it useful to think about various categories of quant investing in terms of their "edge". What advantage does a specific quant strategy rely on?
I see so many startups whose business plan involves "monetizing the data". Sure, they have a product, and some revenue, but the *real* payoff is the data they're collecting. Or so they say.
PSA: it's not that easy. /1
Selling the data directly is almost always a non-starter. Repeatable, scalable, high-value data sales require a set of conditions that are exceedingly rare. /2
First, the data has to be reasonably comprehensive: covering enough of the domain of interest to be statistically significant and economically actionable. This is where most startup datasets fall down: they're simply not big enough. /3
As someone who was trading professionally (and successfully) in both 1999-00 and 2007-08, I have to say it *finally* feels like we're in the late stages of a bull market. I'm not talking about valuations or fundamentals; I'm talking about the zeitgeist. /1
The defining feature of late stage bulls is not price action; it's craziness. Think GameStop, and negative oil, and TikTok investors, and Davey. /2
This craziness is often driven by retail. Retail investors have more buying power, higher risk appetite, and fewer inhibitions than professionals. When retail enters the market, other investors get run over. /3
1/ It's been 6 months since the low point of US markets and economic activity. Ordinarily, we'd see the first academic papers on the COVID recession emerge right around now. But thanks to new sources of data, researchers are way ahead of schedule.
🧵THREAD👇
2/ Let's begin with spending. Chen et al use daily transaction data -- bank cards and QR code usage from UnionPay -- to track the decline of consumer spending across 214 cities in China, one of the earliest indicators of pandemic-induced changes: papers.ssrn.com/sol3/papers.cf…
3/ As early as March, the BEA was using credit card transactions processed by Fiserv to track COVID's impact on consumer spending in the US, per Dunn et al: bea.gov/system/files/p…
Kevin has a great 2x2 where he points out that most well-known VCs are in the "successful + brand-network-effect" quadrant, for obvious reasons -- they need the inbound deal flow. And Sutter Hill is interesting because it's in the "successful + low-profile" quadrant.
This is actually a quadrant I'm quite familiar with -- most hedge funds fall here! As a junior trader I was told: play dumb, stay quiet, keep a low profile, protect your edge, never reveal your positions or plans.
1/ Pricing curves for data are dramatically different from pricing curves for software, hardware, services, or consumer products. A thread of some things I've learned:
2/ Price per data point first increases, then decreases with quantity. Small datasets are usually worth less than big ones. But beyond a certain point, adding more data points doesn't add marginal information, and hence the price plateaus.
3/ Price first decreases, then increases with adoption. Unique datasets are worth more than commoditized ones, but once a dataset becomes "table stakes", price goes up again, especially if there's a single dominant supplier.