Skip to main content

Bill Gross Patents Way To Count

  • Author:
  • Updated:

Mazel tov to Bill Gross on his patent for using arithmetic in certain ways! Here it is in patentese; here it is in English:

Bill Gross, who runs the world’s largest mutual fund at Pacific Investment Management Co., received a U.S. patent for the methodology behind a global bond index that weighs countries by the size of their economy rather than their indebtedness. … By focusing on gross domestic product, the index avoids concentrating on countries saddled with debt and offers a greater weighting for developing markets.

So hey that's just super. I'm actually envious; I used to work in a field that back in the day would occasionally patent derivative trades for their tax/etc. advantages, but then the IRS/etc. realized that they could just look at the patents to see who was cheating - sorry, being aggressive - on their taxes. That's kind of a hard sell to clients - "every purchase comes with a painful audit at no extra charge!" - and the days of patenting derivatives are kind of over.1

But patented indices lacks that you'll-be-horribly-audited factor, and indices are a hotly competitive business for some reason. I've never entirely understood the business of licensing indices - it's, like, licensing averages! I can average prices! I'll sell you a spreadsheet for ten bucks! - but some of the appeal is obvious.2 Or at least more obvious post-Liborgate. An index maintained by a third party and licensed for profit is less likely to be manipulated, since the third party's incentives are (1) to make the index reliable so they can profit from selling it to everyone and (2) specifically not to make it unreliable so they can profit from manipulating it, since they're in the index business rather than the trading-the-thing-referencing-it business.3

That's not Bill Gross's business, though: he does trade the thing, rather a lot of it, and his goal in patenting this index is presumably less to profit by licensing it to anyone who asks than to profit by refusing to license it, making his the only GDP-weighted-in-this-particular-way bond fund in the land and thus competitively more desirable. Market-cap weighting by definition gives you average performance, and if the GDP weighting is better - which the patent application argues it is4 - then it should in expectation give you above-average performance. So you should be able to charge a premium for it, and you should want to keep your competitors from using it.

So, like most patent applicants, he's looking for a monopoly, and seems to have gotten one, though arguably of a rather limited kind since there's after all more than one way to use arithmetic and I am right now developing a bond index weighted by ... I dunno, by something historically correlated with GDP. But there's arguably another, more pleasing reason for Gross to want to keep his index restricted. If investing relatively small amounts in global bonds via a GDP weighting does create outperformance, then everyone should use it. But of course the more people who use it, the less edge Pimco will have: investment strategies that reliably outperform attract imitators until the outperformance dissipates.5 If Bill Gross's patent can delay others' efforts to pile money into his strategy, that doesn't just protect his stream of fees from investors in his funds, it also protects their performance.

Pimco’s Gross Wins U.S. Patent on His World Bond Index Method [Bloomberg]
United States Patent No. 8,306,892- Gross, et al. - Fixed income securities index [USPTO, via Joe Weisenthal]

1.Some patents still lurk in equity-linked securities, though I don't think anyone ever enforces them. Much the same can be said for silly acronyms - back in the day, I guess, before my time, there was some marketing value in telling a client "do a trade with our VIPER group!" Then everyone realized that that was idiotic.

2.Though there is some daylight between "ETF providers, financial advisers etc. should license third-party-provided indices" and "you should be able to patent those indices," but whatever. Our patent system is broken, discuss amongst yourselves.

3.Also, I idly speculate: you'd have some incentive to keep the index limited to some number that reasonably relates to the amount of stuff underlying it. One goofy aspect of Libor is that it indexes this tiny-to-nonexistent market, but there's like eighty bazillion dollars of derivatives tied to it, so you could profitably manipulate the eighty-bazillion-dollar market by trading in the much smaller cash market. (Except you don't have to because you can just make up a number for the cash market, but that's a separate issue.) Ideally you'd want to license a bond-market index, say, to institutions that represent only like a fraction of the total principal amount of the underlying bonds, rather than a huge multiple. Charging licensing fees lets you do that; arguably part of the problem with Libor is that everyone was free to reference it so the dopes setting Libor had a huge impact on the world financial system without necessarily meaning to.

4.Because it doesn't overweight overpriced assets, doesn't overweight bad credits with lots of debt, has countercyclical features insofar as debt prices often drop with GDP growth, and generally speaking based on historical data "would advantageously provide an investor with higher returns for a given risk level than would a market-capitalization weighted fixed income portfolio." Bloomberg tartly notes "The Global Advantage Bond Index is used as a benchmark by the $5.24 billion Pimco Global Advantage Strategy Bond Fund, which returned 7.2 percent this year, trailing 58 percent of peers."

5.At the extreme, if everyone uses it, GDP weighting converges on market-cap weighting, meaning that a country with ~10% of world GDP and rather less than 1% of world debt would … well, its debt would get pretty rich pretty fast. TBF the Pimco patent lets you use derivatives, presumably in part to mitigate this issue.


Bill Gross Wants His God Damn Money

Pimco's worst nightmare is back.