Get Ready Wall Street, Google And Amazon Are Coming

Hiromichi Mizuno, now in his third year as chief investment officer of Japan’s Government Pension Investment Fund, the world’s biggest manager of retirement savings, thinks dramatic changes are ahead for asset managers. While the headlines blare “World’s Biggest Asset Manager Says AI Will Replace Asset Managers,” Mizuno’s own preference for incorporating ESG, or environmental, social and governance factors in the pension fund’s investments, point to a human touch – a decision not likely made by computers. In the future computers, artificial intelligence and human asset managers will more tightly integrate with one another — and “Wall Street” may be confronted by Silicon Valley for dominance.

Get The Full Series in PDF

Get the entire 10-part series on Charlie Munger in PDF. Save it to your desktop, read it on your tablet, or email to your colleagues.

Pension Fund Manager
kalhh / Pixabay

Artificial intelligence will integrate with human logic, says pension fund manager

Artificial intelligence is likely to replace humans when it comes to short term trading decisions, Mizuno said in a Bloomberg interview. But the real issue is not replacing all asset managers, but the integration of computer-based thinking with a human logic that can’t be replicated in a formula.

To date, algorithms are limited in certain thought process capabilities.

“Ultimately all algorithmic trading is great at responding predictably to situations within a known universe, or things which might correlate well to the event,” QuantConnect CEO Jared Broad told ValueWalk. QuantConnect provides a platform for algorithmic traders to write and test their algorithms and recently opened up their system to institutional investors looking to tap this capability.

Mizuno, for his part, sees asset managers focusing on long-term themes, aided by artificial intelligence to focus on using human intelligence for higher level tasks. “I believe in human wisdom,” he said. “I believe AI will release the human resource to do something else.”

It is this integration between a mathematical logic with discretionary logic where the future might be found. “Even fully-automated trading firms require human design and oversight,” Broad observed, looking into a future where the asset manager jobs shifting more into strategy design. In other words, humans determine “what AI should care about” and how the programs get implemented.

Artificial intelligence isn't likely to avoid "sin stocks" when making mathematical investment decisions, as Amazon and Google expected to dominate "Wall Street"

Mizuno’s focus on ESG is a prime example of human decision making overriding a pure, unemotional mathematical logic.

If Japan’s pension fund were entirely operated by an algorithm, would it incorporate ESG concerns into investment decision making unless it was programmed to do so? Avoiding investments in certain "sin stocks," for example, has dampened returns. If the computer was given a primary goal to deliver absolute returns, would even consideration for the environment or socially responsible investments enter that formula without it being forced?

Would a computer “learn” on its own to have a social conscientious?

“I think the long-term thinking and the ESG-like non-numerical, non-quantitative information will continue to require human interpretation,” Mizuno said.

“So far there hasn't been an algorithm with the ability to apply general reasoning, but this may change,” Broad noted.

Mizuno sees AI having a significant impact to the point many of the major players may change. He sees Google and Amazon entering the asset management business, creating a direct challenge to the dominance of Wall Street.

“I take it for almost granted that they will come into this market because they have cutting-edge AI technology and they now capture all the big data of what’s happening in the market,” he said.

Rather than looking at Amazon as an e-commerce firm, some analysts see it as a superior technology firm that overlays their systems on top of various industries – a move that justifies its sky-high price-earnings multiple.

When looking at the cause of the next financial crisis, Mizuno sees an issue that a computer likely couldn’t successfully analyze due to the lack of previous historical precedent: the “mishandling of the reverse QE.”

Never before have interest rates gone negative, violating rules of economic textbooks and supply and demand logic, much like never before have non-economic buyers dominated the interest rates market to this extent.

“I think people are a bit too optimistic that central banks will be able to reverse their quantitative easings smoothly because nobody has done it before and this is going to be a big, big historical experiment,” Mizuno said.