Japan's GPIF revamps portfolio strategy to navigate uncertain global markets
In recent years, the growing uncertainty in global markets have prompted Japan’s Government Pension Investment Fund (GPIF) to adopt a new approach to portfolio management.
“To deliver the benchmark return we need to be good at rebalancing and manage our active risk very carefully. Focusing more on that, our tracking error to the benchmark has become relatively low and we are steadily returning the benchmark return,” Eiji Ueda, chief investment officer and executive managing director at GPIF, told AsianInvestor.
“Going forward, we will be able to add small alpha on top, but I want to achieve that on top of a very steady basis,” he added.
To meet return targets, GPIF, one of the world’s largest pension funds, has revamped its rebalancing strategy while diversifying its active management across more, but smaller, external mandates, particularly in equities.
This diversification of active fund managers will help stabilize returns for the fund, according to Ueda.
“If a manager can add value to their benchmark two years out of three, we get a win in two years out of three from that mandate. If we have 100 managers who can win two out of three, and all of them, assumed, make completely independent decisions to their portfolios, the chance of years where we will go below benchmarks becomes very low,” he said.
GPIF spent almost three years developing the methodology and assessing the skills of active fund managers. It began investing with those managers about a year and a half ago and has since selected more than 60 active funds.
“The returns have been very steady since we started investing this way. By having more managers adding value to our portfolio, that increased diversification will give us a much steadier performance. And there is no room for us to miss the benchmark return due to the design of the fund,” Ueda said.
As of June 30, GPIF holds ¥254.7 trillion ($1.8 trillion) worth of assets.
BACK IN BALANCE
When GPIF aims to achieve its primary objective of generating returns, it targets both absolute return goals and benchmark returns.
From fiscal 2015 through 2019, the fund underperformed its benchmark by approximately ¥2 trillion, or $20 billion, according to Ueda.
“Due to the way GPIF is designed, there is no buffer to fill a future deficit. So, missing $20 billion means a lot. And the fund management cannot miss the benchmark targets within a five-year strategy period.” he said.
As a result, Japan’s Minister of Health, Labour and Welfare adjusted the midterm five-year objective starting from 2020, placing greater emphasis on return and designing the portfolio to achieve the target return.
“Because of our size I found that if we don’t rebalance properly, we tend to miss the benchmark return. If you look at our annual reports between 2016 to 2019, four years, the reason why the fund underperformed against the benchmark was basically that the rebalance was not efficient enough to meet the benchmark return,” Ueda said.
Also read: What GPIF thinks about crypto, other niche assets
GPIF’s portfolio is equally divided into four asset classes: international and domestic fixed income and equities. Therefore, the fund needs to rebalance when one of these asset classes outperforms the others.
“We analysed how we could rebalance more efficiently, but also on how to execute this in the markets without having any impact on the markets themselves. As a result, the impact of our rebalance has been reduced significantly so we, at least, can perform similarly to our benchmark,” Ueda said.
ACTIVE RISK
GPIF also identified other areas in need of optimization, particularly within active funds and the active risk they entail.
Active risk, as defined by GPIF, includes passive funds whose benchmarks are unrelated to the policy benchmark. Even if they are passive fund, they still pose active risk to the fund.
“When I initially looked at the portfolio, there was a lot of skews towards certain currencies, countries, risks, and asset classes,” Ueda said.
GPIF then conducted a thorough analysis of all aspects to determine whether the fund should assume those risks. Where it was deemed inadvisable, adjustments were made to align more closely with the policy benchmark. This approach was also applied to the selection of active funds.
“I think that a lot of analysis should be done to select active funds. Because traditionally, what people are looking at when evaluating the fund performance is the alpha. And that is not the only measure for looking at the potential risk,” Ueda said.
Also read: How GPIF uses active management to achieve ESG goals
He cited the example of the S&P 500, where seven tech-focused companies currently account for about 30% of the market cap.
These seven tech giants have relatively high valuations, so many active equity fund managers focusing on S&P 500 tend to invest more heavily in mid- and small-cap stocks, due to their valuation, financial outcomes and more straightforward stock prices.
“Many active fund investments are really skewed to mid- and small-caps, but the index remains as S&P 500. As a result, when the mega caps perform well against the rest, the alpha looks negative. If it is the other way around, alpha looks positive for everybody,” Ueda said.
“We found that the correlations among those funds are very high, so we started to think about how we should evaluate the manager skills properly away from these indices beta and those stuff sticking to benchmark,” he added.