Cerulli Asia-Pacific Edge: Four Sovereign Wealth Funds With Little in Common

Cerulli Associates Asia-Pacific Edge, November 2014 

Asia’s four most powerful sovereign wealth funds offer an interesting case study in diverse approaches to the same central challenge. All four – Temasek and GIC from Singapore, the CIC from China and KIC from Korea – are there to diversify foreign exchange holdings or state wealth, as opposed to the more common model of a finite amount of hydrocarbon wealth that needs to be invested for the future. But they do so in markedly different ways.

Total assets and structure

The four funds vary in assets under management from US$72 billion to US$653 billion, but in fact the total assets available for international investment are a little closer together than they would appear.

The reason for this apparently contradictory statements is CIC. At the end of 2013, CIC had US$653 billion under management, up from $200 billion upon foundation in September 2007. But although CIC’s numbers sound impressive, and appear to constitute the largest sovereign wealth fund in Asia, they are misleading.

CIC has two wholly-owned subsidiaries. CIC International is the one that all international fund managers care about; that’s the one mandated to diversify China’s foreign exchange holdings and to invest exclusively outside of China, in an asset allocation model similar to other sovereign funds around the world. The other is Central Huijin, which holds controlling stakes in most of the biggest state-owned financial institutions in China, including Bank of China, ICBC and China Commercial Bank.

The two are quite separate within the broader institution – CIC’s 2013 annual report specifically states they are “two distinct entities, with a strict operational firewall separating the business activities of the two entities” – and they report in different currencies, with their performance considered separately in annual results. But they are lumped together for the total asset figure, despite the fact that at least two thirds of the total assets are believed to be in the domestic, Central Huijin side; not much more than $200 billion is thought to be invested internationally in the usual sovereign wealth fund style.

The growth in CIC’s assets over the years has been partly through performance and partly through asset injections from the state (generally carved out of the holdings of the State Administration for Foreign Exchange), although those appear to have stopped now. Performance has varied over the years but now looks pretty good: 9.33% in 2013 and 10.6% in 2012 after some difficult early years, making for a cumulative annualized figure since inception of 5.7% per year.

The Government of Singapore Investment Corporation is the only one of the four that does not disclose its total assets beyond saying that they are “more than US$100 million” – it has been saying this for years and the figure is thought to be considerably higher than that. Indeed, in terms of assets available for international deployment, it may still be the case that GIC, not CIC, is Asia’s biggest sovereign wealth fund.

GIC exists to diversify Singapore’s foreign exchange reserves, unlike Temasek, which started out as a vehicle to own Singapore’s state-held assets and then grew into an international fund – albeit one still roughly one third invested in Singapore – from there. GIC does not disclose year-by-year performance, instead opting (like Abu Dhabi’s ADIA, with which it has quite a lot in common) to state 20-year annualised real rates of return. This figure was 4.1% up to March 31 2014, and has hovered around the 4% mark for five years. In US dollar nominal terms, the figure is 6.5%. GIC also discloses annualised nominal returns over 10 years (7%) and five (12.4%).

Singapore’s other sovereign wealth fund, Temasek, is unique among the four in that it is permitted to invest in its home market, as it is not a foreign exchange diversifier like the other three. It differs considerably in asset allocation too, as discussed below, with more of a direct investment model, and partly for this reason uses different criteria to account for performance. Last year’s growth was a modest 1.5% – that’s Total Shareholder Return, the metric Temasek uses – but looks stronger over the long term: 9% per year over 10 years, and 16% per year since inception.

Temasek changed direction in 2002 when it began to look outside Singapore and in to Asia, and it is instructive to look at performance since that change in attitude. Investments made after 2002 have delivered annualised returns of 16%, and those prior to March 2002, 11% (in the years since 2002, that is). Temasek, which started out in 1974, had a net portfolio value of S$223 billion on March 31 2014.

The Korea Investment Corporation had US$72 billion under management at the end of 2013, having started out with $1 billion in 2006. Much of the subsequent increase has come in asset injections from the central bank and finance ministry (of which more below), but nevertheless the fund has been performing well, generating a return on investment of 9.09% in 2013, and a return on its equity and fixed income portfolio of 8.67%. Over the past five years, the total portfolio has returned an annualised 8.27% (mercifully, the disastrous 17.53% loss of 2008 no longer appears in those five-year figures, but it does drag the ‘since inception’ total down to 4.02% annualised). Of the $72 billion in total assets, $12 billion are gains, made upon a total of US$60 billion of entrusted assets.

KIC is a straightforward foreign exchange diversification play: all its investments are overseas, like CIC’s and GIC’s.

Asset allocation

Three of the four funds are broadly similar in approach, with Temasek the outlier. In common to all of them is a renewed focus on equities (though Temasek doesn’t really do any other form of investment bar equities and pre-listing investment which will, logically, become equities in due course).

CIC made some significant shifts in its portfolio in 2013. As of December 31 2013, 40.4% of the fund was in public equity, 17% in fixed income, 11.8% in absolute return, 28.2% in long-term investments, and 2.6% in cash and others. It’s important to understand what ‘long-term investments’ means, as this is key to the fund’s strategy: it includes direct investment and also private equity, resources and commodities, real estate and infrastructure. “Absolute return”, in CIC’s methodology, is chiefly hedge funds and multi-asset.

The standout change year on year was that public equity increased considerably, from around 32% the previous year, while long-term investments went the other way, dropping from 34.2% to 28.2%. In fact, all other asset classes were trimmed in order to accommodate the shift into public equities.

CIC does not embrace emerging markets as fervently as Temasek does, but still puts a considerable amount of its public portfolios outside the developed world. For example, 26.8% of its global fixed income portfolio goes into sovereign bonds of emerging economies; also, 26.5% of that portfolio goes into investment grade corporate bonds, some of which may well be from emerging markets. In equities, 17.1% of the portfolio is in emerging markets. It is worth remembering that this is despite the fact that the fund cannot invest in China, which is a mainstay of equity investments by many other funds. By sector, financials are by some distance the largest part of the portfolio, at 22.9%.

CIC has a dedicated department of asset allocation and strategic research to develop investment policies. Alongside it are two other divisions: public market investment (which has two departments beneath it, one for public equity and one for fixed income and absolute return); and long-term investment (which also has two departments: private equity and special investments). The special investments division lists energy, mining, infrastructure and agriculture as priorities.

Central Huijin is a different entity entirely. It holds stakes in 18 domestic businesses from China Development Bank and the big four state banks to newer brokerages. It is not normally seen as an investor, although comments in September suggested it may invest in European banks in future.

Temasek has a dramatically different model to the other three sovereign funds in this report. It has little or no investment in bonds, or indeed any other asset class beyond listed equities and direct investment in companies. It is an active investor and not purely buy-and-hold: over the last decade it has invested S$180 billion of assets and divested S$110 billion. Last year, with S$24 billion of purchase and S$10 billion of sales, was typical. The split between the liquid and illiquid is roughly a third apiece between liquid regular listed assets, large listed blocks (a more than 20% share of a company) and unlisted assets. It continues to hold significant stakes in Singaporean blue chips, notably DBS, CapitaLand, SingTel and Singapore Airlines, but is also a major shareholder in Asian and multinational leaders such as Standard Chartered and ICBC.

Another point that separates Temasek from its peers is a significant shift in its geographical allocations. In 2004, 52% of the portfolio was in Singapore, 18% in Asia ex-Singapore and 30% elsewhere in the world. In 2014, those figures are 31% Singapore, 41% Asia ex-Singapore, 24% ‘western’ markets (North America, Europe, Australia and New Zealand), and 4% in the rest of the world (chiefly Latin America and Africa), meaning that traditional western markets only account for less than one quarter of the portfolio. Even that’s an overweight, and likely to decline as emerging markets recover; Temasek’s target is actually 40:30:20:10 for Asia, Singapore, developed world and other. Today, 72% of the portfolio is exposed to Asia.

By sector, over the last year or so Temasek has been prioritising financial services (a long-standing focus), life sciences and energy, and more recently consumer. Recent direct investments have included AIA and ICBC (an increase in existing holdings), Lloyds Banking Group, Gilead Sciences, Thermo Fisher Scientific, the BG Group, Pavilion Energy, and Nigeria’s Seven Energy.

Temasek typically states its key investment themes as transforming economies, growing middle income populations, deepening comparative advantages and emerging champions.

GIC set a new investment framework in 2013, using three components: the reference portfolio, made up of 65% global equities and 35% global bonds, in a passive manner; a policy portfolio, which is GIC’s long-term asset allocation strategy, across six asset classes, and brings most of the risk and return in the portfolio; and a further overlay of active strategies that differ from the policy portfolio (this is the bit where external asset managers come in).

To look in detail at the policy portfolio, GIC has a series of bands within which investments can be made: 20-30% developed market equities, 15-20% emerging market equities, 11-15% private equity, 9-13% real estate, 4-6% inflation-linked bonds, and 25-30% nominal bonds and cash. In practice, on March 31 2014, both equity categories were near the top of their range (developed 29% and emerging 19%), and nominal bonds actually above it (31%), with inflation-linked bonds in the middle (5%) and both real estate and private equity actually below their range bottoms (7% and 9%). This may reflect the fact that it takes time to build private equity and real estate investments to these new ranges, although in fact the real estate total actually declined year-on-year. Notably, developed market equities dropped considerably over the year from 36% a year earlier, which partly reflects a pledge by GIC to look more closely at emerging markets, and also probably the fact that in 2012-13 developed market equities looked a lot better than emerging ones did.

Geographically, GIC’s allocations are 42% Americas, 29% Europe and 27% Asia (of which 10% is Japan).  Australasia adds a further 2%.

Generally, over the years, GIC has been seen as an innovative early mover in alternatives. It is one of the largest real estate investors in the world in its own right, and it has worked with a great many of the most powerful names in private equity and hedge funds, often working alongside them as co-investors or cornerstones rather than investing in underlying funds. In private equity alone, where it started out in 1982, GIC boasts a network of over 100 active fund managers.

The separation of GIC into these various parts of the portfolio makes it much harder to read overall allocations across the fund as a whole. We don’t know, for example, what proportion of GIC’s assets go into the reference portfolio – possibly none, if it is purely a theoretical exercise for comparison’s sake, though GIC says it is “not a benchmark” – and how much into active management.

KIC, like CIC, dramatically boosted its allocation to equities in 2013; the 48.4% of the portfolio that was invested in shares by then is the highest allocation figure in KIC’s eight-year history, and was up from 45.2% in 2012. Conversely, fixed income, at 34.3%, is at its lowest ever level; in the fund’s early days, the vast majority of the fund was in fixed income as it gradually developed expertise elsewhere. Alternatives have grown steadily too, now accounting for 8% of the fund, while what KIC calls ‘other’ – inflation-linked bonds, commodities, cash and hybrids – are also becoming significant, at 7.6%.

KIC has, since its early days, divided its portfolio into what it calls traditional assets and alternative assets, and often discloses the returns for the two separately. Alternatives are, in the KIC definition, private equity (3.7% of the portfolio), hedge funds (2.5%) and real estate (1.5%). Separate from both is a long-standing investment in Bank of America, accounting for 1.6% of the portfolio, which stems from an unfortunate heavy investment in Merrill Lynch during the early days of the financial crisis, a mistake the fund is highly unlikely ever to repeat.

One trend in recent years at KIC has been direct investment in mainland China, following the launch of a dedicated division for this in September 2012.

Disclosure

CIC discloses a perhaps surprisingly large amount of information for a Chinese state entity. This includes total assets (although not the clear delineation between CIC International and Central Huijin), asset allocation, performance, some regional and sector distribution information, and various structural and governance information.

It does not tend to disclose individual holdings, nor a ranking of the biggest holdings.

Temasek discloses total assets and performance numbers, but not asset allocation – though since everything is either listed equities or direct stakes prior to listing, there’s perhaps not much point. Temasek goes into more detail about individual investments than its peers. For example, its most recent review documents that it increased its stake in AIA to over 3.5%, and in ICBC to 8.9% of its outstanding H-shares, while it acquired 1.1% of Lloyds Banking Group.

GIC’s level of disclosure is arguably the weakest of the four, since it does not include total assets, nor one-year returns (though it does disclose five, 10 and 20 year returns, from which one-year returns can be roughly surmised). It is not clear how much is given to external managers, and how much of the portfolio is passively managed.

KIC discloses detailed asset allocation, asset and performance numbers, but typically not the biggest holdings. Curiously, it has stopped disclosing the split between internal and external managers.

Use of third party mandates

Three of the four are lucrative sources of mandates for third party managers, albeit at often razor-thin fees (except for some alternative asset classes). Temasek offers little.

As of December 31 2013, 67.2% of CIC’s funds were outsourced to external managers, a figure that has risen steadily over the last few years and is unquestionably the highest of the four. There was, however, a great deal of churn in individual mandates through 2013 as funds were reallocated towards public equities. The research group Z-Ben painted a bleak picture for fund managers in July after calculating that the average fee for external managers was around 32 basis points, a fee that would seem tight even if a large part of it were in passive allocations, but Z-Ben believes use of passives is still fairly small at CIC. It now has a reputation for very tight pricing, and for being willing to cut off non-performing funds swiftly. It does, however, outsource a greater proportion of its funds than any of its immediate peers.

Very little of Temasek’s funds are outsourced. Why would they be? When one is buying a stake in a company, one needs a broker or an M&A advisor, not a fund manager. The latest Temasek strategic review said that “less than 10% of our portfolio was invested in third party managed funds.” In fact, it would be a surprise, if the total was anywhere near 10%.

Temasek does, however, sometimes work with funds as co-investors, which likely accounts for the bulk of that figure. In 2013-14, for example, it structured a pool of private equity funds into an investment vehicle called Astra II for investors to participate in. Temasek also owns an asset management company of its own, Fullerton Asset Management, which started out handling in-house investment before going on to manage third party assets as well; it is possible that Fullerton is included in Temasek’s comment about outsourcing to third party managed funds.

GIC is known as a tough place to get a mandate, in large part because the in-house expertise is extremely capable and sophisticated. Generally, the best opportunities are thought to be in alternatives like private equity and hedge funds, and then often in a co-investment model rather than a mandate. The following sentence from the latest GIC Report is instructive: “GIC partners top-tier fund management institutions that offer access to opportunities, specialised capabilities, in-depth analysis and experience which complement our internal management capability.” The key word here is “partners”.

That said, GIC does also invest in real estate funds, private equity funds, bond funds, index funds and hedge funds, and gives external fund managers discretionary mandates in asset classes including both global fixed income and equities.

KIC, like many sovereign funds, started by outsourcing widely, and then developed expertise in-house and gradually took the simpler asset classes under its own management, using external managers for value-add. So, for example, it launched in-house global fixed income investment in August 2007, and equity in March 2008, then formed a special investment team in March 2010.

KIC formalised this arrangement in its 2013 annual report, saying: “Based on the alpha-beta separation program, we pursue returns slightly in excess of the benchmarks (enhanced beta) with a low level of tracking error for internal investments. We seek to capture greater excess returns (alpha) through allocations to external managers who focus on specific regions, sectors and asset classes.”

Although KIC has stopped disclosing the proportion of assets that are outsourced, it is believed to be about one third.

It will be useful for external asset managers to understand the new research centre KIC launched with some pride in January 2014. It describes this as “the cornerstone for long-term value investment”, and creates in-house projections as a basis for asset allocation strategies and model portfolios. This should be expected to have consequences on KIC’s asset mix and, consequently, manager selection.

Source of funds

CIC’s source of funds is a subject of some debate. For several years it typically received between $20 billion and $50 billions per year of injections from the state, but it is believed to be at least two years since one was awarded and no more are expected in the immediate future.

It had long been assumed that the reason Central Huijin, which generally has nothing at all in common with CIC International, had been put together in the same institution, was that as Central Huijin’s companies went to the public markets, the funds would be passed down to CIC. However, it is understood that this is not taking place, which raises the further question of what Central Huijin is going to do with those funds. There are hopes that this will become clearer in the year ahead.

Temasek started out in 1974 as a method of separating the ownership responsibilities of commercial companies from the day-to-day business of running the country. Initially, it was given stakes in big national companies as they were privatised, starting with the blue chips like DBS and Singapore Airlines, and then as Singapore privatised more assets in the mid 1990s, SingTel and Singapore Power. Since then, Temasek generally does not receive asset injections from the state, and must instead be a buyer and seller in order to diversify. It is not given foreign exchange reserves, which is the role served by GIC. It is expected for the foreseeable future that it will simply continue to manage the portfolio it already has.

KIC is distinctive in having two masters. It receives money both from the government and from the central bank, the Bank of Korea, and has faced a consistent challenge in the fact that the two institutions are believed to have quite different expectations around risk and reward.

It is expected to continue to receive new assets. In July 2014, Hongchul (Hank) Ahn, the chairman and CEO, wrote: “On the premise that the Korean government and the Bank of Korea entrusts additional assets as planned and that investment performance remains at past levels, assets under management at KIC should reach US$100 billion in the near future.” He also said that KIC was targeting US$300 billion in assets “before long”, which clearly would require further asset injections.

Conclusions

For external asset managers, CIC, KIC and GIC all represent enormous potential, but there is a clear pattern that managers are going to have to offer something different, with a proven track record of doing it well, and likely to have to compromise on fees as well. Alternatives, particularly private equity, continue to be a good way for an external manager to attract the attention of sovereign funds. That said, for eye-watering fees, there are believed to be good passive mandates available from all three as well. Temasek is less of an opportunity.