

Hiromichi Mizuno had spent more than a decade growing London-based secondaries firm Coller Capital into a global investment platform when he became the first chief investment officer of the world’s largest pension fund in January 2015.
Mizuno was brought to the Government Pension Investment Fund of Japan to impart a new culture to the institution – the culture of global investment banking – and to help it evolve, industry insiders tell us.
Verdicts on Mizuno’s tenure at the more than ¥150 trillion ($1.4 trillion; €1.2 trillion) GPIF are unanimous: the CIO is driven and incredibly smart.
He is also shaking up the monolithic establishment. On taking over at the beginning of 2015, he was thrust into dealing with the pension’s historic policy shift of increasing investments in riskier assets, such as stocks over traditional Japanese government bonds.
In his four-year tenure Mizuno has created a blueprint for GPIF’s entry into alternatives, grown its investment staff from zero to around 15 as of 2018 – and created gamechanging innovation in using artificial intelligence.
A Tokyo-based investor who has known Mizuno since he was with Coller says the CIO has very distinguished views – thinking on a more global standard – on various industry matters, including ESG and the controversy over executive pay that hit Japan Investment Corporation in December.
“He has a good reputation in the market and has built a good and important network with both Japanese and non-Japanese players in the community.”
According to Mizuno, the biggest surprise for him when he came on board was finding the investment team – while it had a lot of talent – had no experience in alternatives. He was alone at the top, tasked to build a team from scratch, and with the gargantuan job of deploying up to 5 percent of its total portfolio – more than $50 billion – in private equity, real estate and infrastructure.
“The industry kind of underestimated GPIF in a way, because the team had no prior experience investing in alternatives. I came in with the expectation that we really had to build the team and bring in new hires,” Mizuno says.
The second surprise was the market expectation of GPIF’s “transformation”, and the increasing public scrutiny – both domestic and global – on how it would re-organise its portfolio of assets. “Every time we did some new investments or some changes to our portfolio, we attracted much more media attention,” he says.
For example, when GPIF incurred a ¥14.8 trillion quarterly investment loss in the quarter ended December 2018, opposition parties criticised the pension in local media and called for the fund to shift back to its previous bond-heavy asset mix.
A ‘TOP CLASS’ FUND OF FUNDS
Early on, Mizuno had a gut feeling the investment team’s success would depend more on his ability to attract talent than on his banking skills.
“When I became CIO the first thing I did was analyse GPIF’s business model and other business models used by asset owners, on whether we would build an in-house investment team or continue to outsource to external asset managers,” Mizuno says.
While all the big and leading asset owners were taking more investments in-house, GPIF was restricted from investing directly. The hybrid model (half of the investments outsourced, and half done in-house) was discussed initially but didn’t get enough nods. Mizuno notes he himself didn’t believe in that model. “This dilutes the limited partners’ bargaining power over the asset managers, if we show them we will eventually take their jobs and become possible competitors,” he says.
“I also found that it’s not practical and reasonable to believe that we can build a class A in-house investment team within the regulatory restrictions of our remuneration and our location.”
This led to the decision to come up with a new model based around full outsourcing and long-term partnerships.
“We have also decided to use it as our bargaining power when we discuss the investment opportunities with managers. I always feel that whether they like us or not, we are guaranteed to be the asset management industry’s biggest customer,” he says.
COMMITTED TO ESG
Under Mizuno’s tenure, GPIF has made undeniable progress in the wider adoption of ESG considerations across its portfolio.“ESG has now become the central pillar of our activities and we use ESG to send our message through our investment chain,” Mizuno says. “By demanding asset managers integrate ESG and engage with portfolio companies’ executives on material issues, we are actually making the investment chain more long-term focused.”
ESG issues don’t have impact on a business’s valuation tomorrow, but it will affect its valuation for the long term, he adds.
Mizuno considers private equity to have the best governance among asset classes, since GPs own the company and make sure the executives will always act in the interest of shareholders.
This means private equity managers are also in the best position to implement “E” and “S”.
The pension giant has high expectations for all its asset managers. It signed up to the UN Principles for Responsible Investment in September 2015 and has since asked its managers to make the same move or explain why they are not doing it, Mizuno says. “We think those who don‘t agree to the principles of PRI do not share our investment philosophy.”
Mizuno stresses GPIF is not pursuing the “Canadian model” of originating, evaluating and investing in private assets by themselves.
“We have no plans to take any investment in-house. Our aim is to become in a way a top-class fund of funds manager.”
But building a “top-class” fund of funds is a huge challenge and a lot of complexity for a new investor in the asset class, especially with GPIF’s size and scope.
The first challenge is the limited number of professionals who have experience in alternative assets within the Japanese asset management industry. Compared with GPIF’s public markets investment team, it was much harder for the private markets team to find talent.
The second challenge was operational. With GPIF’s legacy of investing only in public equities and fixed income, its entire regulatory framework, internal rules and data management and IT system were not designed to facilitate alternative investments.
Despite all this, Mizuno and his team hired a consultant three years ago and developed a blueprint on how it can effectively enter the market.
The first strategy is the fund of funds manager mandate. In April 2017, it kicked off a search for private equity, real estate and infrastructure asset managers to establish separately managed accounts for the pension. For private equity, it was seeking fund of funds managers targeting North America, Europe and Japan, and focused on several strategies such as buyouts, growth capital, private debt and venture. At press time GPIF had not revealed its selection for private equity.
The investor awarded infrastructure mandates to Pantheon, DBJ Asset Management and StepStone Infrastructure & Real Assets in the first four months of 2018. Los Angeles-based CBRE Global Investors received GPIF’s first global real estate mandate in September 2018.
“We started with the direction of appointing FoF managers so they can go on autopilot and commit capital to leading GPs. Given limited resources on hand, we just cannot afford to select funds one by one,” Mizuno says.
“We gave [the fund of funds managers] quite significant cheque sizes so that they can continue to look for good GPs and keep deploying money.”
Mizuno admitted that early on he was frustrated with the speed of GPIF’s progress in alternatives, but realised it was wiser with its scale to work step-by-step.
GPIF now wants to revise its alternative investment blueprint, scaling up from its “meaningless” private equity allocation: just 0.21 percent of the 5 percent target has been allocated to alternatives, as of December 2018.
One way of doing that is by finding partnership opportunities with peer asset owners. Mizuno says many leading asset owners have approached GPIF about potential strategic partnerships for co-investments. The pension is considering these, he tells us, and wants to take advantage of its position in the market as the biggest asset owner in the world.
Mizuno notes there is a lot of expectation from the industry to deploy a lot of capital into alternatives, but GPIF wants to make sure that when it does team up with other asset owners, it will “contribute to make that partnership work well”.
It is a missed opportunity, as Mizuno sees it, that LPs don’t seem to communicate with each other as much as they should.
He is a big advocate of partnerships in the investor community and has made it a top priority for GPIF. Its global asset owners’ forum, now on its fourth year, has seen participation from US pension giants California Public Employees’ Retirement System and California State Teachers’ Retirement System, as well as APG and GIC. The investors meet in Tokyo to discuss common issues, from managers’ fees to environmental, social and governance integration and collaborative engagement with corporations, says Mizuno.
“These things are common challenges asset owners are facing. We have just been used to having too little dialogue with each other.”
REDESIGNING FEE STRUCTURES
Mizuno says he also wants a “very open dialogue with GPs on how they should structure their fund – starting from the fee structure, the vehicle set-up, as well as their governance mechanism”.
People often expect GPIF will demand an attractive fee and carry structure, he tells us, but the pension is more concerned with keeping the LP-GP alignment of interest.
“In reality our focus is not in reducing the actual amount of fees we pay, rather we just want to redesign the conventional fee structure, so we have much more confidence in the LP-GP alignment of interest.”
Mizuno, however, declines to share details on how GPIF wants to redesign the fee structure with its managers.
He says: “Overall, I think the fee and carry structure in private equity is much better than public equities because of the GP commitment, which at least gives investors confidence that they have skin in the game. But the fee and carry balance can be revised and discussed, as well as the catch-up rate and clawback.
“GPIF is not going to be a GP. We will continue to be an LP. The ideal relationship between a GP and ourselves is one in which we are confident and have faith in the alignment of interest.”
GPIF doesn’t need to micro-manage or micro-monitor; it would rather let the managers do their job, he adds. As an alternative asset class, private equity is only useful if it beats the performance of GPIF’s public equity portfolio, Mizuno says.
“That’s the private equity’s industry’s responsibility to prove – that even with the higher fees and inconveniences that are put to investors, it still delivers extra returns.”
HUMANS OVER ROBOTS
GPIF is leaning into artificial intelligence. In December 2017, the pension partnered with Sony Computer Science Laboratories to investigate harnessing AI in its investment decisions and manager selection. A team from GPIF and Sony CSL analysed the pension’s manager structure and developed a proof-of-concept prototype system to test the use of AI to detect the investment style of managers from trading behaviour data. A paper on the study was published in March last year and GPIF concluded it was equipped with AI and intends to move forward with more advanced technology.Alternatives, however, are one area where it will take much longer for AI to replace investment professionals, Mizuno says.
“Alternatives require a lot of dialogue, emotions and a lot of nuances. I’m not saying that the private equity industry shouldn’t adopt AI, but we see the human strength in private investments compared to public equities investments.”
ESG is also another area in which humans have more competitive advantage than robots, he adds.
“AI doesn’t care what world our grandchildren will be living in, and ESG requires some of the value judgment only humans have.”