Well this is rich (no pun intended). Bloomberg this morning ran a piece detailing UBS’ apparent plan to ramp up the hiring of quants.
Now as you’re probably aware, HFT and systematic strategies (risk-parity, CTA, vol. targeting) are suspected to have exacerbated the 1,000 point flash-crashing madness that occurred in the wake of the Chinese yuan devaluation. Algos just scan headlines and systematic strats are literally forced to sell into selloffs. That’s the VaR shock episodes we’ve been warning about. As long as volatility stays low, these funds can lever up, but as soon as volatility returns they have to delever – in a hurry. If you’re running a leveraged long fixed income portfolio, but suppressed historical vol allows you to lever up in equities too, well then if you get a spike in volatility you’ve got yourself a problem.
We really don’t need more of this, nor do we need more sub-penny scalping, frontrunning algos all chasing the same headlines at the speed of light. But UBS apparently doesn’t care about any of those concerns. Here’s Bloomberg:
“UBS Group AG doubled the number of quants working for Chief Investment Officer Mark Haefele in the past two years. He wants to hire even more.”
“Haefele recommends customers use investment strategies that apply techniques designed by quantitative analysts, such as complex data visualization, rather than rely on their instincts. One in 10 managers who advise clients now use such tools, said Haefele who oversees a team of more than 200.”
“The models “help us understand when it’s time to dial up the risk of the portfolio or dial down the risk,” Haefele, who helps oversee $2 trillion, said in an interview in Zurich. “In many cases the answer is yes, a machine could do a better job than a person.”
“Money managers including UBS, Credit Suisse Group AG and GAM Holding AG are betting that the strategies widely used by the hedge-fund industry will help convince clients spooked by market volatility to invest their money instead of keeping it in cash.”
Now for one thing, the models don’t really “help you understand” when to dial risk up or down. They just do it based on the incoming data. Haefele’s statement is like saying “the washing machine helps us understand how clothes get clean.” No it doesn’t, you just turn it on and it when it’s done, the clothes are clean. Same concept here. The strategies auto adjust and you hope that at the end of the day “the clothes are clean” (to borrow our own analogy).
But the real punchline is the bit about hiring more quants coaxing retail off the sidelines. Can you imagine that pitch?
“Well, Mr. and Mrs. Johnson, we know you lost half of your retirement in 2008, but now is a good time to get back in the market because we’re trading at the oh-so-cheap price of 18X and oh by the way, let me introduce you to who will be managing your money”…