Omega Advisors Chairman Leon Cooperman, a vocal self-made billionaire, is often on CNBC, and he cites the proliferation of risk-parity funds as the culprit of our recent "flash crash". To use a simplistic definition, risk-parity is an investing strategy some quant funds use to limit risk by over-allocating lower volatility assets. This investing technique usually has a heavy dose of bonds over equities, and is supposed to protect investors such as pension funds in all investing environments. In theory, if stocks sell off, you make money with the bonds, and visa versa.
The problem is that both stocks and bonds are not supposed to depreciate at the same time. Enter the new "flash crash" where both stocks and bonds took a beating. A recent Reuters article gives you more depth about risk-parity funds and the possibility that they did do damage to the overall markets three weeks ago. Mr. Cooperman took exception to funds like this in a CNBC interview because they cause instability in the indexes, plus alter conventional investing tactics:
"In the world I grew up in, and the world Warren Buffett grew up in, when something went down you wanted to own more, and in the world that we're in now, it goes up you want to own more, and it goes down you want to own less, and that's just counter-intuitive. It lacks common sense."
He's absolutely right that it doesn't make sense. You've probably heard stock pickers use the expression, "It's a market of stocks, not a stock market.". That's a dead chestnut in today's era of programmatic portfolio allocations. An old stock picker like Cooperman said it almost correctly:
"I think the machines seem to be taking over."
The machines don't seem to be taking over, they have taken over. That's why it's much more advantageous for individual investors to be in S&P 500 index funds such as SPDR S&P 500 ETF Trust (SPY), iShares S&P 500 Index (IVV), or, Vanguard 500 Index Fund (VOO). Any one will do. This is especially true if you are a domestic investor. It's like John Henry versus the steam drill. John Henry and his hammer won, but only to die at the end of the competition from exhaustion. It's much better to be on the mechanized side of the fight at this juncture. Yes, you can still pick winning individual securities, but what's the probability you'll do it consistently when competing against computers.
The Reuters article cites that the world's largest hedge fund Bridgewater Associates, allegedly lost 4.2% in August. Bridgewater's 'All Weather Fund' is an algorithmic trading vehicle, a risk-parity fund. It's supposed to make money during market sell-offs. Because you need to be a high roller to invest in hedge funds, it's the one percenters that lost money during August investing in Bridgewater's fund. Nevertheless, it's Main Street investors, either through pensions, 401K plans or individual broker accounts, that probably saw shades of the 'Great Recession' of 2008-2009 flash before their eyes due to market instability. It was the lead story on the evening news for a week.
Bridgewater Associates has $165 billion in assets under management, and this is miniscule compared to the net worth of the trillions of dollars invested in the overall markets. I doubt they were the lone wolf that caused such a big market meltdown in the most recent 'flash crash', but they could have contributed to it. However, they were asleep at the wheel during August with their proprietary trading algorithms, which presumably were being monitored and adjusted by a team of human beings. What's going to happen when the machines take over, and all trading is guided by artificial intelligence? The future is closer than you think.
"I'd rather be a hammer than a nail"
High finance is quickly morphing into a machine learning world along with the rest of corporate America. Bridgewater Associates recently formed a new artificial intelligence division spearheaded by David Ferrucci, the mastermind behind the IBM and academic engineers that created the Watson computer system. According to an article by Phoebe Venable:
"The AI unit will devise trading algorithms that make market predictions based upon historical data and statistical probabilities — and like all AI systems, it will adapt to new information and get smarter as it goes."
Warren Buffett also likes to say, "If past history was all there was to the game, the richest people would be librarians.". However, Bridgewater isn't alone in its pursuit of unlimited profits by utilizing machine learning. A blog posting by Robust Tech House lists most of the major players venturing into the new era of AI, plus a brief synopsis of their business models. The firms included are:
- Two Sigma Investments
- Bridgewater Associates
- Clonealgo
- Renaissance Technologies
- Aidyia
- Cerebellum Capital
- Rebellion Research
- Commeq
- Castilium
- Binatix
- Sinai
- KLF Capital
If you have plenty of cash to burn, a closer examination of these organizations will give you a nonstop flood of ideas where to invest if you are inclined to go with machine learning algos. The prevailing orthodoxies on Wall Street would tell you to put your money in firms like these. However, I don't know which ones are flush with cash, or which ones are skating on thin ice. With the exception of companies such as Bridgewater Associates, most hedge funds have short self lives, automated or not.
Aidyia, one of the AI Funds listed above, is thoroughly covered in Quartz by Georgia McCafferty. As she states in her posting about the future of finance:
"Most quantitative trading, as it is currently practiced, relies on a human being to develop a mathematical model to identify trading opportunities. The model is then updated by hand to adapt to new markets or changing conditions. For an AI, conversely, humans develops the initial software, but the AI itself develops the model and changes it over time."
Just like the Bridgewater algo. What I find disturbing is that if the people monitoring Bridgewater's 'All Weather Fund' can't keep from losing money in a "heads I win, tails you lose" environment, what's an algorithm going to do if they aren't programmed correctly? This could put a lot of pressure on the markets and cause a collapse. It almost happened three weeks ago with human interaction. What about on autopilot?
I'm not a Luddite, but am wary of the new era. With the proliferation of personal computers and smartphones, the age of privacy is over. This becomes more evident when you mix machine learning with cloud computing. Artificial Intelligence in all forms is here to stay, especially in finance. It's a leading edge business where technology is concerned. There has to be some government intervention to monitor automated hedge funds or else they're doomed to crash the markets.