Chat with the Chief: Hiromichi Mizuno
Posted on 05/21/2018
GPIF’s head of investments tells SWFI what’s really behind the Japanese pension giant’s new fee structure, why the asset management industry is in need of a paradigm shift, and how to create a sustainable business model that works for the small-scale asset owner.
It’s been nearly four years since Hiromichi Mizuno left his position at one of London’s premier private equity firms to help reshape the world’s largest public pension – first as a constituent of its eight-member advisory committee, then as its chief investment officer. Since returning to his native Japan – where people 65 or older already account for more than a quarter of the population – Mizuno has been on the front lines, leading the country’s massive Government Pension Investment Fund in a fight against time that is being waged by asset owners in aging nations around the globe.
To prepare Japan’s welfare apparatus for future generations, Mizuno has had to aggressively redeploy GPIF’s US$ 1.45 trillion in assets away from the government bonds of years past towards higher-yield bets in stocks both foreign and domestic, as well as alternatives in private equity and infrastructure; recruit talent from the private sector capable of executing the migration into uncharted asset classes; and roll out an innovative new set of Stewardship Principles for sustainable investment and company engagement that has been felt throughout the country’s famously staunch business culture. Now he’s working on redefining how the asset management industry should be judged and compensated for their efforts – all within the budget afforded to him and the 100 or so civil servants who work together at the fund’s headquarters in Tokyo’s bustling Toranomon district.
Course-correcting such a monolithic public institution – especially one responsible for the financial security of more than 70 million stakeholders from all walks of life, including the many Nikkei-listed firms in which GPIF invests – has not been easy. A universal owner in the truest sense, the fund relies upon a host of external managers and index vendors to provide for the retirement of more than half of Japan; taken as a whole, its movements have consequences for the economic wellbeing of the entire country, if not the world. Like a flagship at sea, its fortune is inextricably tied to the supporting fleet of vessels arrayed in formation around it, and the steadfastness of their crews in the face of disaster.
At its helm, Mizuno’s every move is put under intense scrutiny in Japan – often times by those who measure success in terms of news cycles and quarterly reports, rather than generations or the systematic integrity of global markets. Even with what is perhaps the most drastic portfolio restructuring in the history of the industry almost behind him, appealing to the concerns of a public made wary by decades of deflation without giving away the fund’s investment strategy to its competitors is likely to remain a full-time responsibility for the pension chief – no matter how effective his policies prove to be over the long-term.
Despite the pressures of the job, Mizuno’s abiding optimism remains firmly rooted in the ingenuity of the human spirit he sees expressed in rapidly developing technologies such as artificial intelligence and big data analytics – disruptors he hopes may one day soon be harnessed to propel the financial world into a new age of investment in which sustainability, in every sense of the word, is universally valued as the key to a brighter future that is ever on his mind.
In this interview from early April with Spencer Ward, research associate and writer here at the Sovereign Wealth Fund Institute, Mizuno provides insight into what went into GPIF’s recently implemented performance-based fee structures, why he believes asset managers need to be closely reevaluating their business model, and how small-scale asset owners can cost-effectively focus their stewardship activity to systematically affect change across the investment chain.
SPENCER WARD: Why don’t we start by telling me about how GPIF has evolved as an organization since you first started there?
HIROMICHI MIZUNO: Well from an organizational and cultural perspective, we still carry the image of a bureaucratic organization. But everybody who has recently joined our team here has been very surprised, I think, by how open and dynamic our organizational culture is. We have been encouraging members of our team to find new and innovative ways to contribute, not only within the framework of conventional asset management, but also in terms of how to add value to the sustainable performance of GPIF throughout the investment chain, and more recently about how to impact social responsibility for the benefit our pensioners.
I think that this shift in mindset has resulted in a growing sense of pride in what we are doing here. It wasn’t too long ago that we were regarded as these bureaucratic investors. But now we have a lot of opportunities – thanks to people like yourself – to have our work recognized, not just as a slight variation or copy of what other asset owners are doing, but what I think is our own quite creative approach to the markets. It’s difficult to explain to someone from the outside how much GPIF has changed and evolved as an organization since I first began. But when I get feedback from people coming to us from the private sector, from which we are heavily recruiting, they are always surprised by how much more dynamic we are from their previous organizations.
We are not competing against other asset owners, and because GPIF is the textbook definition of a universal owner, we can’t beat the market either.
WARD: Now you’ve played a big part in that shift. You’ve become a vocal advocate for the integration of ESG criteria into long-term sustainable investment, which has gained a lot of traction in the private sphere over the past couple of years, and you’ve worked with stakeholders across the investment chain to help shift their focus as well. When did you start including sustainability as one of your responsibilities as chief investment officer (CIO) at one of the world’s largest public pensions?
MIZUNO: Initially it was a reflection of my analysis of the industry. When I came into this role, I tried to analyze the business model of asset managers, as well as asset owners, and when I looked at the two of them side by side, I found that most asset owner’s business model is just a copy – in many cases a low quality copy – of the asset manager’s. So I realized that we needed to find the proper business model that works for the long-term investor, particularly asset owners, with the resources available to them.
The initial reason I pushed for ESG integration was that I questioned what exactly we were trying to achieve as fiduciaries. If I were an asset manager, I have to rise to a certain level of competitiveness, right? I either have to beat the market, or beat the competition. But asset owners have no competitors, in a sense, because our sole responsibility is to our pensioners and our government’s fiscal obligation to the health and wellness of future generations.
We are not competing against other asset owners, and because GPIF is the textbook definition of a universal owner, we can’t beat the market either. We own the market through a globally diversified portfolio that is largely managed passively, and as a truly long-term investor, we have very low chance of actually divesting from those assets, so we need to be paying more attention to how to make the system more sustainable – and our performance with it – over the long term.
To my thinking, ESG is one of the most effective tools for communicating our deep concerns about sustainability to our investment chain. In other words, GPIF is less concerned with how we beat the market, or with how to underweight or overweight specific companies. We don’t benefit from zero-sum competition within an industry. If one company took the whole market, well yes, their stock may appreciate. But we would lose as much, if not more, from our holdings in their competitors. So what we need to do is ensure that capital markets as a whole become more sustainable.
It’s a very high level concept for most people in the industry, I think, and at the beginning I struggled to convey it to our partners throughout the investment chain. But when I was first introduced to ESG, this concept began to fall into place for me as a way to bridge the conceptual into something tangible that could be acted upon. So rather than simply telling our managers, “We want you to manage our money with more long term focus,” it became far more effective to say, “You need to integrate ESG factors into your investment process,” in order to get them to pay closer attention to sustainability, and frame it as a goal that was within their reach.
WARD: Of the three sustainability factors – environmental, social, and governance – which one came for you first personally?
MIZUNO: I of course have my own personal interest in the different issues, but as an organization we never prioritize one issue over another for a number of reasons. The first is the multi-stakeholder model of the investment chain that you mentioned earlier. Given our organizational structure, we really struggle to focus on any one issue. Unlike the asset manager, who manages money strictly for their clients, we have a fiduciary responsibility to all Japanese individuals, well as Japanese corporations that contribute to our funds. So it just didn’t appear to be wise for GPIF as an organization to pick one issue over another.
What we decided in the end was to promote ESG as an integrated whole. But on the other hand, in our Stewardship Principles we make it very clear that our asset managers should integrate those factors into their investment process, and that for critical issues they need to proactively engage with their portfolio companies. So far, we’ve refused to spell out which of these issues are critical, and have left it to our managers to decide that for themselves.
Now this is somewhat of a tactical approach, in that rather than raising the issue with all our stakeholders and debating which are most important, we’ve outsourced that discussion to our managers. As a matter of fact, I think I speak not only for GPIF, but for the asset owner community as a whole when I say that we have a shortage of resources compared to managers. With the exception of some very fortunate funds like those in Canada, asset owners are almost always short on staff and expertise. In contrast, asset managers have a high concentration of resources, and that gives them the freedom to find out how to best engage on these issues on our behalf as they see fit.
I think this approach has worked out well so far. Right now we are checking with all our external managers on what ESG factors they’ve identified as critical, and making sure that they really are engaging with their portfolio companies those issues. It’s a bit of a chess game, going down the chain one by one, but the important thing for us is verifying that these changes are moving the system in a more sustainable direction.
Here at GPIF, we are trying to develop a model for demanding best-in-class corporate governance that is more suitable to the public investor, and I hope that this kind of approach can be adopted by small-scale asset owners – whether it’s in Japan, emerging markets, or even in the United States.
WARD: GPIF has been somewhat instrumental then in using its status as a universal owner to drive asset owners towards ESG integration and engagement in Japan. What kind of lessons do you think other asset owners around the world – particularly in emerging markets – can draw from your experience?
MIZUNO: From the beginning, when we first started analyzing the business model of asset owners, one of the challenges I identified in the community is that there are actually very few who are fortunate enough to have the resources to apply a more direct engagement model and become what is essentially an impact investor. In California and Canada, for example, they are shifting their operations to take on more of the responsibilities of an external manager and engaging directly with companies. But given regulatory restrictions, which prohibit us from direct investing in specific companies, we were not able to go in that direction.
So we began exploring how we could focus our stewardship activity on the asset manager, rather than on specific companies, as well as how to promote ESG integration holistically, rather than targeting some particular issue. One of the things we found in our analysis, is that the industry as a whole does not pay enough attention to the importance of the index vendor – many of whom are shifting from a simple market cap weight, or mechanical index building, towards more qualitative and adjustive indices that factor in ESG considerations.
We now probably spend as much time examining and working with index vendors as we do with our asset managers, which is unconventional in this industry. But I’ve always thought that if we can work with vendors to change their metrics, which would in turn directly affect external asset managers’ investment decisions, then we can be a role model for change to smaller investors throughout the global asset owner community who cannot afford their own in-house investment team, and so are forced to outsource everything in the way that we are required to do by law.
Many of the practices that the fashionable asset owner is doing these days cannot be replicated with the resources available to the average public pension. GPIF has about 110 people, and together we manage about US$ 1.45 trillion, which is quite a bit larger than most asset owners, but we still face many of the same challenges as other public investors in terms of insourcing. When we put together our Stewardships Principles, we took a somewhat innovative approach of sampling all the other major asset owners, looking at their guidelines for direct company engagement, and then adapting them to apply instead to the external manager. So rather than outlining how to engage with companies as many asset owners have, our set of Stewardship Principles explains how we engage with our managers.
Here at GPIF, we are trying to develop a model for demanding best-in-class corporate governance that is more suitable to the public investor, and I hope that this kind of approach can be adopted by small-scale asset owners – whether it’s in Japan, emerging markets, or even in the United States.
WARD: Well it seems like you’ve developed a strategy for bringing about that best-in-class governance by making the best of what you have available to you. You’re putting the responsibility on your external managers and index vendors to utilize their own resources – which are considerable – to meet your needs as a customer, and in doing so instigating a shift in perspective across the industry at the systematic level, which is in contrast to what some other asset owners have done by striking out down the direct investment path.
MIZUNO: You know, the keyword I use when I discuss our approach with my team is “systematic.” We simply cannot afford to optimize our portfolio by tackling individual investment decisions. At the end of the day, we are both a beneficiary as well as a potential victim of the global economic system as a whole. If the system fails, we also fail. That’s the concept I believe everyone should be aware of, but I’m afraid that there are very few players in the industry who actually understand that.
Many of them continue to act as though we, the asset owners, are trying to beat the market, and I still very much appreciate the importance of active managers and the value they bring, especially because they are keeping a certain level of pricing discipline in the capital markets. But we are still, like I said, more invested – as both a beneficiary and victim – in the sustainability of the system as a whole.
WARD: Let’s move on and talk about your new performance-based fee structures. Now, you’ve been very careful to emphasize that the goal of these changes is not to reduce the total fee amount, and in fact under the new system you’ve actually provided a large incentive to your active managers to generate alpha by removing fee-caps that existed under the previous structure. Based on what you’ve seen so far, how have your managers prepared to meet these changes? And what do you think will separate those who thrive in this new environment, from those who struggle to earn their keep, so to speak?
MIZUNO: First of all, I cannot say this enough: We have zero intention of reducing our total fee payment. Actually, I’d be happy to pay more, because that means we’re getting greater returns from our portfolio. To those who say that we are no longer willing to pay – and there are players in the industry who always criticize our fees as being too low – I would respond by arguing that our management fees have always been low, even before these changes.
But if you look back on our performance net of fees over the past decade, the alpha has been almost zero. My question for the asset management industry, particularly active managers, is this: How can you expect us to pay you more, when you didn’t leave us any money in our pockets the last time? I believe that the active management industry has to change their business model, because over the past decade, after accounting for costs, they didn’t bring any value to us as a customer. First of all, how can they survive like that? And second, how can they expect us to pay more? It’s a very simple observation of what we have experienced in the past, but it’s a crucial starting point.
I’ve always said publicly that I would like to increase our active portfolio. But I can’t do that if we are continuously facing a situation where the alpha net of fees added by our active managers is zero, and what I’ve told them is, “You are the ones who need to prove that GPIF has the basis to increase its active portfolio.” We are trying to send a signal to active managers that they have to do better. If they can do that – and I think they can – they will earn more, and GPIF can allocate more money towards them through our active portfolio.
The second point I’d like to make sure is understood clearly, is that incentivizing asset managers to do better is not my main focus. I came from the asset management side before I became CIO, and I know how hard they work. I really don’t think that if I just give them more of an inventive that their performance will improve overnight. If that were the case, then that would mean they are lazy, and I know our asset managers are not lazy. So I really don’t believe that these new performance-based fees will incentivize them to work harder, because I think they’re working hard anyway.
We are challenging the traditional concept that active managers should be paid according to their costs, when they should be paid according to the value they bring to the customer, and that’s what I really want to make crystal clear.
What we are trying to achieve with these performance-based fees is a change in concept around remuneration. With our passive managers, I really don’t mind discussing their costs, because they cannot really prove they are adding value to us other than through qualitative analysis. I have been advocating that our passive managers be more active owners in terms of engagement, and am happy to discuss raising their fees to cover operational costs if their business model is compelling.
But for active managers, what I would like to change in terms of concepts around remuneration is this: If they don’t deliver the value they proposed, which is to produce a certain level of alpha, why should we pay them more than our passive managers? Sometimes they even underperform the benchmark, and then we end up paying fees on poor performance, and that makes no sense to me.
In the investment industry, active managers have traditionally charged fees to cover their operational costs. But in any other industry, customers don’t care about the cost of the producers; we only care about the added value their product delivers, right? So I’ve told our active managers, “I really don’t care how expensive it is to hire analysts, or researchers, or fund managers; that’s your problem. We only pay you if you add value to our business through alpha.”
So our new fee structure is less about incentivizing them to do a better job, and more about making it clear that for GPIF, the active manager’s value is alpha, and if they fail to deliver that, then they shouldn’t be paid any more than our passive managers when we can achieve the same results with them at a much lower cost. We are challenging the traditional concept that active managers should be paid according to their costs, when they should be paid according to the value they bring to the customer, and that’s what I really want to make crystal clear.
One piece of technical jargon in this industry that I never really thought made sense is capacity. Everyone conceptually agrees – and this is supported by academic research – that if the manager’s portfolio grows past a certain size, their capacity to generate alpha will diminish. In practice though, all we can do is ask the asset manager how much money they are willing to take on for us.
From their side, there are a lot of different incentives – especially if the fee is fixed – for them to take more and more of our assets under management, even at the expense of a deteriorating alpha. From the asset owner’s side though, we are working with an asymmetry of information that doesn’t put GPIF in a good position to fairly judge their capacity limit. So let the asset manager be responsible. With this new fee structure, asset manager’s shouldn’t be taking on more than they can manage, because then it becomes punitive as alpha starts deteriorating.
Part of the issue is the alpha target that asset managers pitch to me and my investment team, and that we agree upon at the time of hiring. Unfortunately, in the history of GPIF’s management, less than 20% of managers have delivered the target for alpha that they have set for themselves. If a company, in this case a fund manager, is delivering on their contract only 20% of the time, the customer should get a refund, right? So we made it very clear to our managers, that if you agree or propose an alpha target, that’s the break-even point, and the new system is actually designed for our active managers to receive identical fees if they achieve their target.
This is not a reduction in fees; it’s a change in the fee structure. If they perform as well as they propose, they will receive exactly the same amount. If they perform above their alpha target, of course they will be rewarded on top of that – and we put no cap on that bonus. But if they don’t achieve what they propose, then they’ll be penalized.
It’s surprising to me that they invest in all these tech companies, and they know how dynamic their portfolio companies are in adopting new technologies to improve their business model, and yet they don’t seem to be applying them to their own business.
WARD: And how have they responded so far to this change in concept?
MIZUNO: Well their initial concern was with whether they could continue to cover costs, and I reiterated my point to every single one of our asset manager’s that I sat down with, and it’s that I really don’t think that the cost-transfer model for remuneration is acceptable in business anymore. It’s only the utility and asset management industry who still think they can transfer costs to the customer, when what we should be moving towards is an environment of competition centered around bringing value to the customer.
Probably the biggest problem has been the misperception among our asset managers that GPIF is trying to reduce fees, or incentivize them to perform better, when that is not our goal. All we want to do is try to challenge the norm of how fee structures should be designed and who they work for, and what we’ve paid the most attention to throughout this process is alignment of interest between us and the asset manager, as well as creating self-governance mechanisms that let the asset manager be responsible about what they can accomplish during the negotiation process. So alignment of interest and implementing mechanisms for self-governance have been the two key objectives of this new proposal.
WARD: Now let’s back up a bit to your point about how incentivizing better performance isn’t your goal. You already know how hard your asset managers work, and they work hard. But you know, as they say, if you can’t work harder, work smarter. In previous interviews you’ve talked about how adoption of technologies like artificial intelligence and big data analytics are already fundamentally changing the investment world, particularly in areas of short-term trading. Do you think the asset management community is doing enough to adopt these new technologies, and how do you think asset managers will need to adapt their investment philosophies, strategies, and models to keep pace?
MIZUNO: In our discussions with asset managers on our new fee proposal, we spent quite a bit of time with them going over their cost structure, and what kept coming up was automation. In every industry today, when the customer is not willing to pay a given price, and companies are competing to reduce costs, they automate some of their operations, right? So I asked them why they are not doing the same. You cover this industry, so you know how much pressure asset managers are under to reduce costs, and in my opinion, what they should be thinking about is how to automate their process. That’s the case for any industry facing consumer pricing pressure.
From the beginning, I’ve held the view that the asset management industry needs to adopt technology much faster than its current pace. It’s surprising to me that they invest in all these tech companies, and they know how dynamic their portfolio companies are in adopting new technologies to improve their business model, and yet they don’t seem to be applying them to their own business. So I’ve been somewhat frustrated by asset managers who come to see me that are only focused on the macro-economy and capital markets, when I want to be hearing more about how they are improving their business, including how to bring these technologies in-house.
I think fin-tech is most likely to change how we do indirect financing, and the banking and brokerage sectors in particular are feeling a lot of pressure right now. But I’m surprised by how much the asset management industry seems to feel like they are secure or isolated from these disruptive technologies. This is one reason why I’ve been pushing GPIF towards research in artificial intelligence (AI), to sort of stimulate the industry to take a serious interest in its development. Because we are such a data and intelligence-centric industry, I think it’s natural to believe that AI will reshape how we do business.
The more I learn about AI, the less optimistic I become that people in short-term trading and investment will be able to survive in the future. So I’ve been asking our asset managers, as well as my own team, how we are going to stay in business when AI begins to take over short-term investments. I could be wrong, and I’m really looking forwards to the ideas that people come up with, but so far I haven’t heard much.
I think that what humans should be focusing on are long-term issues that AI wouldn’t care about, like systematic social and environmental sustainability. ESG is one of those areas that requires a human value judgment, at least that’s my thinking. All I can do is raise these issues when I communicate with our asset managers, but so far I haven’t really received any compelling input from the industry, so we’ve decided to move ahead and start a joint research program with Sony Computer Science Laboratories, which is obviously a non-conventional player in the investment management chain. But we’ve been discussing with them a lot of new ideas about how to use AI systems to assist in making our manager selection and monitoring process more dynamic.
I don’t think that AI will replace asset managers for the time being, not because their performance is not good enough, but because people’s perceptions will prohibit us from fully replacing human management. But I think that getting AI’s assistance will be critical moving ahead, simply because so many investors – including GPIF – are short of staff. So we feel that we should be pushing AI into this business, and that’s why we are leading on this issue.
WARD: Well, humans are nothing if not resilient.
MIZUNO: Absolutely, but we have to work hard to prove that we are better, and not allow ourselves to become entrenched in our thinking. I don’t think you can assume that AI won’t at some point become overwhelming, which is why we have to find out what particular areas humans can do better, as well as what we can do much better with the assistance of AI. So GPIF is now working with Sony to see if we can use AI to analyze all the investment data that we have been collecting over the decades from our asset managers and the companies in our US$ 1.45 trillion portfolio.
If you are the CIO of an external manager, you really don’t have to care about how your investment strategy will be perceived by different stakeholders, because at the end of the day you’re judged only on your performance. But even if we made US$ 100 billion in profits, we just can’t make everybody happy.
WARD: Let’s shift from the future back into the present. You’ve talked about your desire to have a more active portfolio, of allocating up to 5% of your assets towards alternative investments in infrastructure and real estate, and over the past couple of years you’ve overseen a pretty massive shift in the fund’s fundamental portfolio structure. Bringing about that kind of change to a government institution responsible for the financial security of 70 million people – especially in Japan, which is facing a number of demographic and economic challenges – is no easy task. As we wrap up here, can you tell us about how you’ve navigated and grown from that process, and what else you hope to bring to the fund moving forward?
MIZUNO: The last time we decided on some of these big policy changes to our asset mix in 2014, my involvement in the process was still only as a member of the investment advisory committee. It was only afterwards that I was made CIO and began to implement what was probably the most drastic portfolio rebalancing in the history of the industry.
Unlike a sovereign wealth fund, GPIF and other public pension funds almost always have to work to meet certain transparency and accountability requirements that are required by law, and GPIF’s requirements are sometimes almost impossible to cope with. GPIF also has many different stakeholders, so that’s made my job even more difficult.
I joke with the CIO of the U.K.’s environmental agency pension plan, because she always tells me that she never has any problem introducing concepts like ESG to her employees, and I tell her, “Of course you don’t! Their job is to be conscious about the environment, so nobody’s going to argue that incorporating the environment into investment decisions isn’t a part of your fiduciary duty!”
But in our case, we have to manage the interests of every single Japanese beneficiary and company that contributes to our fund. So making these kinds of drastic policy changes at GPIF requires a lot of stakeholder management and massaging of the media, who tend to be very tough in their coverage of what we are trying to achieve and don’t usually give us enough in the newspapers to explain fully why we’re doing what we’re doing. It’s very difficult, and will continue to be difficult.
If you are the CIO of an external manager, you really don’t have to care about how your investment strategy will be perceived by different stakeholders, because at the end of the day you’re judged only on your performance. But even if we made US$ 100 billion in profits, we just can’t make everybody happy.
What I would really like is for our stakeholders and people in the industry to know more about what we are trying to achieve in how we approach our portfolio. For instance, we’ve set up a manager registration system that’s open and dynamically responsive 24-7, and one of things we’ve began offering as part of the new performance-based fee structure is multi-year contracts to public equity and fixed income managers, which is somewhat unconventional as you know. It’s similar to the private equity world that I came from, where managers will receive a 10-year mandate. Nobody offers anything like that in public equity, but we would like to give away the option of firing a public equity manager at our discretion within a short cycle.
What we are essentially telling our managers with this is that we want them to manage our money with a long term perspective, and one of the ways we would like to do that is through a performance fee carry-over mechanism. We don’t allow them to vest all of their performance fees each year. Rather than focusing on getting a big check at the end of the year and risking a crash from focusing too much on the short-term, we want them to look at their performance over the long-term, and I hope that’s something that is being observed alongside everything GPIF is doing to make sure our asset manager’s investment process becomes more oriented towards sustainability.
SWFI’s Chat with the Chief interview series aims to present the investment perspectives and experiences of some of the foremost C-level practitioners in the sovereign wealth fund and public pension community. This interview was conducted over the phone on April 5, 2018, and has been edited for clarity.
Keywords: Government Pension Investment Fund Japan.