Brad Setser

Follow the Money

Cross border flows, with a bit of macroeconomics

Print Print Cite Cite
Style: MLA APA Chicago Close


Should SAFE be considered a SWF?

by Brad Setser
June 11, 2008

SAFE is China’s State Administration of Foreign Exchange. It enforces China’s capital controls — and manages the foreign exchange reserves of China’s central bank.

Lately though it hasn’t necessarily been investing in classic central bank reserve assets. Participating in a TPG fund is something an aggressive sovereign fund might do, but not something a traditional central bank would even consider.

SAFE clearly wants to show that with enough flexibility, it can get the same kind of returns (or better returns) than the CIC.

If I had to bet, I would guess that SAFE now manages a larger equity portfolio (counting investment in private equity firms) than all but five or so of the big sovereign funds. ADIA, KIA, Norway’s government fund all likely have a bigger equity portfolio than SAFE. Temasek and the GIC likely to so too, though I am a bit less sure on that front. Temasek and the GIC combined likely have bigger equity market exposure than SAFE, but a lot depends on just how much equity SAFE has bought since June 2007 (the last good US data point). SAFE almost certainly has more equity market exposure than the CIC.

If China’s reserves continue to increase at $75-80 billion a month, SAFE could quickly become among the biggest sovereign equity investors.

The FT’s sovereign fund (and PE) beat reporter Henny Sender was struck by how large SAFE’s investment in the TPG fund was.

China’s State Administration of Foreign Exchange has agreed to invest more than $2.5bn in the latest TPG fund, in what could be the largest commitment ever made to a private equity firm, people familiar with the matter say … Investments in private equity firms are usually not made public, but industry executives believe the largest previous investment in a private equity firm came from pension funds in the US states of Oregon and Washington. The two funds both invested about $1bn to $1.5bn in Kohlberg Kravis Roberts.

I though am struck by the fact that $2.5 billion is less than a day’s reserve growth, and thus not really all that much money for China. $75 billion — really $80 billion after valuation changes are taken into account — translates into something like $4 billion to invest every business day.

The scale of China’s April reserve growth — and indeed China’s 2008 reserve growth, once the funds China is assumed to have shifted to the CIC in q1 are taken into account — is truly astonishing. China’s $75 billion in April reserve growth easily tops the United States $61 billion April trade deficit. And $60 billion a month still strikes me as a lot.


  • Posted by Dave Chiang

    Slightly off topic, but the NJ State Employees Pension Fund is directly financing the bailout of Lehman Brothers on Wall Street with an equity investment of several billion dollars. Former Wall Street executive Jon Corzine is the Governor of New Jersey that has oversight of the state-owned pension investment fund. Certainly the state employees of New Jersey didn’t vote in a shareholder proxy to invest their pension funds into Lehman Brothers that is laden with subprime mortgage bonds. Why is that any different than Singapore’s state-owned Temasek pension fund that also makes private equity and bond investments?

  • Posted by Twofish

    Personally I think it’s a bad thing that SAFE gets involved in currency investment since from a management point of view its better to split up currency regulation from currency management. You get all sorts of conflict of interest problems if you don’t split up those roles.

  • Posted by Twofish

    DC: Why is that any different than Singapore’s state-owned Temasek pension fund that also makes private equity and bond investments?

    1) It’s not different, and
    2) Do you think it is a good thing or a bad thing?

    This is one reason I don’t think that you will have the IMF or anyone else issue effective rules and limits on SWF’s. A lot of the political motivation and fear of SWF’s come from the fear of foreign interests buying US companies, but having limits on SWF’s is going to run into problems when it becomes very obvious that some of the larger SWF’s are those run by US states and local governments.

  • Posted by Dave Chiang


    The biggest sovereign wealth fund in the world is state-owned California CALPERS with $245.4 billion in current assets. Recently Calpers which previously avoided equity investments in Asia due to perceived market transparency issues changed it tune. A booming 10% GDP growth rate in China is simply too big to ignore. Although only a small percentage of total assets, California Calpers is currently invested in various Chinese equity shares including China Mobile, but PetroChina is too politically incorrect for the Hollywood crowd in California for Calphers to invest in. Oh my God! Can you imagine the political uproar among the Washington Consensus Elites if the Chinese government restricted US state-owned SWFs from investing in China?

  • Posted by MMcC

    I think there are some fairly good reasons not to think of SAFE as an SWF now or in the long run, even though it may, at times, appear to invest like one.

    1. SAFE is not particularly autonomous from its sovereign (it’s just one more arm of the PBoC) and it can be called on to meet short-term (if largely theoretical) liabilities. Those always seemed to me to be two of the more core aspects of SWFdom.

    2. SAFE is exploiting an absence of regulations to make investments such as the one with TGP – no-one has yet formulated guidance/legislation to prevent SAFE doing these kinds of deals, even though they’re not really part of SAFE’s raison d’etre. CIC, by contrast, is doing exactly what it was set up to do. Long run, these kinds of tensions in China are typically resolved in favour of the rule-abider.

    3. EM central banks (and their functional arms, such as SAFE) are holding and will likely continue to hold a greater proportion of equities than 1st world peers. They’re not playing by the same “make your currency a rock-hard, credible global reserve option” rules. That said, my count of CIC’s current equity holdings (listed plus mezzanine) is about USD$135bn; I’d be interested to know how Brad got to a higher number than that for SAFE. I’m not sure holding equities – even at 10-15% of the portfolio mix – turns a reserve into an SWF.

    4. If we think of SAFE as a form of insurance against speculative Rmb attacks, you can make a reasonable case that China is now over-insured. It’s that lack of credible demand for SAFE’s assets which gives them the freedom to make such illiquid and unlikely investments. SAFE’s TGP investment, among others, is, in my view, helping make the case in Chinese policymakers’ minds that the sterilisation policy needs to be revisited. In that sense, there’s a self-limiting dynamic being created that may limit SAFE’s future SWF-like investments.

    5. SAFE is to some extent closing doors to CIC by making these investments. They’re not yet competing for deals but they’ve made it less likely, in my view, that CIC can go to TGP for future investments. What’s worse, in my view, is that SAFE are a scarier investor than CIC: they really are a direct tool of state policy in ways CIC can claim not to be. SAFE’s actions may, in worst case, tempt legislators in destination countries to write anti-SWF rules which debar both CIC and SAFE. That would be very problematic for China, who are making every effort to ensure doors stay open for CIC, even when they are closing to other Chinese quasi-governmental investors.

    Mid-to-long term, I think SAFE are going to be put into a smaller box, one with investment restrictions more typical of other sterilised reserves and 1st world central banks. That iterative, problem-solving limitation approach is also consistent with the way SOEs and other quasi-state organisations have been treated in the past when they’ve encroached into territory that the government has reserved for another player.

  • Posted by Rachel

    Brad, what about the CIC/JC flowers fund? wasn’t that more than 2.5b? I thought it was 80% of a $4b fund.
    Though perhaps that is a different category given that the CIC fund was a one-off joint fund not a contribution to broader capital raising by a private equity fund?

  • Posted by Twofish

    The investments that CIC has made in private equity done well, and I suspect that the main argument that SAFE used to get permission to what it did was “we can do things better than CIC.”

    MMoC: SAFE’s actions may, in worst case, tempt legislators in destination countries to write anti-SWF rules which debar both CIC and SAFE.

    This assumes that they weren’t planning to do so in the first place. The question that I have asked continuously about SWF’s is that “what is the issue that needs to be fixed?”

    If the main issue is foreign control of domestic industries, then it really doesn’t matter if the foreign investor is transparent or not.

  • Posted by Twofish

    To DC: It isn’t. Japan has a much larger SWF in its Postal Savings Fund. It doesn’t do much foreign investment so no one notices it.

    Also, you didn’t answer my question. Given that Calipers, NJ Pension fund, and Temasek look very similar, do you think that they are good ideas or not?

    DC: Oh my God! Can you imagine the political uproar among the Washington Consensus Elites if the Chinese government restricted US state-owned SWFs from investing in China?

    No honestly I can’t. Most political uproars are window dressing for someone that stands to lose money on an issue, or else single issue groups that make a lot of noise but have very little power or influence once the noise dies down.

    Also the Washington Consensus has been dead for about a decade. I don’t think that any of the “Washington elites” really take it seriously any more, and I doubt that anyone on Wall Street ever did.

    One interesting thing is that Chinese investment law is that it more or less doesn’t care very much about the source of the money. Money is money. The main restrictions are on the destination of money.

  • Posted by dr.dan

    offtopic : india raises rates to 8%

    its getting out of hand , i am afraid

  • Posted by Dave Chiang

    Given that Calipers, NJ Pension fund, and Temasek look very similar, do you think that they are good ideas or not? -TF

    So long as there is adequate regulatory oversight of state-owned funds, I don’t see anything wrong with them? Without the proper regulation of financial markets, the inevitable result is Enron, Worldcom, Fannie Mae, Countrywide Mortgage, Tyco, Freddie Mac, Bear Stearns, etc. With the current “anything goes” regulatory environment, we are sure to have more “blow-outs” in the financial market. Capital is lazy too without regulatory discipline.

  • Posted by Dave Chiang

    The real issue here is developing nations are not run by puppets anymore and last seen, the US Treasury controlled IMF is not in any position to dictate anymore. If the US minds its own business in the world, that will solve 99.99% of current US foreign policy problems.

  • Posted by bsetser

    MMcM. I agree that SAFE is exploiting a lack of clarity about what it can and cannot do. It clearly got freedom to take more risk, but the exact boundaries don’t seem to be defined.

    My estimate is based on the fact that the US data showed $25-30b in Chinese holdings of US equity at the end of q2 07, with almost all of the increase coming from mid 06 to mid 07. I assumed a comparable or slightly lager flow in 07/08 (we are getting close to end june) — something that is consistent with anecdotes that SAFE has been buying equities in its own name/ SAFE’s activities in Australia and Europe. I also arbitrarily assumed that the US accounts for maybe 50% of SAFE’s equities, which generates ball park of around $50b in US equities/ $100b in global equities. That is roughly 5% of China’s total reserve assets (assuming ongoing growth) — a sum that “feels” reasonable.

    I am curious how you estimated $130 for the CIC — I accept that it is likely that they got $210b – $70b (huijin) – $20b (CDB) by the end of March. But some of that will be set aside for ABC methinks, tho less than initially forecast. That leaves roughly $100b. And maybe not all of that will be in equities and perhaps not all of that has been invested as of now — I was under the perception that the CIC hasn’t formally awarded a lot of mandates yet, but that is very third hand information. I suspect that others have better data than I.

    I disagree a bit tho with two of your arguments:

    a) that most SWFs are autonomous from the government (and more autonomous than central banks reserve managers)

    and b) the cIC has been set up to be something other than a tool of state policy.

    on a) ADIA, GIC and the CIC all report directly to the top level of their respective governments. there is no arms-length relationship as far as I can tell. ADIA’s board is basically drawn from a single family … that need not imply a decision to invest for non-commercial reasons (ADIA has tried to avoid doing that, Abu Dhabi’s big domestic investments have been done through other state vehicles) but there is a very close relationship. Here the CIC fits with a broader pattern.

    and on b) it is hard for the CIC to be seen as a purely commercial investor when it is the holding company for strategic stakes in state banks and it is reported (per the FT Huawei story) to have a mandate that includes supporting outward investment by SOEs.

    In that sense, SAFE — as a manager of external assets — is purer.

    I fully agree with your point that a central bank that feels free to invest in an illiquid PE fund with a long-tie up period clearly has more fx than it needs for conventional central bank purposes.

  • Posted by Twofish

    bsetser: CIC all report directly to the top level of their respective governments. there is no arms-length relationship as far as I can tell.

    In the case of CIC there is a board of directors that insulates the activities of CIC from the State Council. There isn’t this layer of insulation with SAFE.

    Wen Jiabao doesn’t have the legal authority to directly order CIC employees to do anything. He can withhold approvals, and there are other ways that he can get his way, but these are direct and information. He does have the legal authority to order SAFE to do something. CIC is owned by the state but not part of it, while SAFE is part of the state, and that makes the rules quite different.

    My reading of the SAFE action is that the central government has decided that they want more control over how reserves are invested.

  • Posted by Twofish

    Basically the difference between CIC and SAFE is that in the case of CIC, Hu Jintao can’t show up in someone’s office, point to them and then fire them on the spot and order that someone be that person’s replacement, whereas in the case of SAFE, he does have this authority.

  • Posted by bsetser

    SAFE reports to the PBOC which reports to the state council.

    The CIC reports to the state council. the CIC’s board is made up of government ministries. I fail to see the difference.

    It strikes me that the CIC is institutionally far more a creature of China’s government than an independent body making arms-length investments.

    And I would rather surprised if Hu and Wen cannot order someone at the CIC to be fired. the CIC’s investments — the big ones at least — all seem to have required state council approval. IF there is an arms length relationship there i just don’t see it.

  • Posted by MMcC

    bsetser: “I disagree a bit tho with two of your arguments:

    a) that most SWFs are autonomous from the government (and more autonomous than central banks reserve managers)

    and b) the cIC has been set up to be something other than a tool of state policy.”

    Really sloppy writing from me there and I apologise for it: what I should have said was that CIC, unlike SAFE, only has one policy goal to pursue – make money for itself – and that this unity of purpose makes it more immune to the kinds of “what’s the real agenda” criticisms that multi-tasking agencies like SAFE face from foreign scrutineers.

    I fully respect your view of the bone-headedness of folding ABC, CDB et al into CIC if immunity from such criticism was a design goal. I think MoF and others saw the anchor stakes NCSSF had in a number of banks as a model worth following to provide future funding to CIC and didn’t think through all of the implications. Neither of us are looking to China for evidence of policy genius. I think how they treat those banks – particularly what they refuse to do for them/allow them to do – is going to be closely watched and I’m at least reasonably confident that CIC aren’t going to deprive themselves of any opportunities by favouring their temporary wards.

    I’m not sure that the FT’s question about CIC’s role supporting SOEs is really all that valid (and, to be fair, the FT didn’t allege that CIC had that role, only that no-one at the time was absolutely sure if it did or didn’t.) Since that article was published last December, CIC officials have been asked 100 times in 100 venues what they planned to do to support SOEs. They’ve answered “nothing – it’s not our job” 100 times and no action that they’ve taken appears to contradict that statement. If there’s one Chinese governmental investor that appears determined to colour inside the lines at the moment, it’s CIC…

    bsetser: “SAFE almost certainly has more equity market exposure than the CIC.”

    I took this to mean total equity exposure, not foreign equity exposure. On foreign exposure, you may well be right. I would be surprised to learn that CIC’s foreign equity holdings – including the nearly-completed first mandates – exceed USD30bn. My USD135bn figure includes domestic equity.

  • Posted by bsetser

    MMcC —

    I actually think the CIC has more than one objective. It existts to support China’s fx regime by helping to sterilize reserve growth. That could imply constraints on its portfolio allocation across currencies that conflict with its return objective. It clearly means that it cannot invest domestically even if domestic investments would produce higher returns — ergo, it is a sovereign fund with an expected negative net worth in its own currency.

    Moreover, the moment the CIC got the SCBs, it clearly got an additional objective — namely protecting china’s strategic objectives in the financial sector.

    And I saw in this week’s FT a CIC official was quoted to the effect that supporting SOEs going forth is one of the CIC’s goals (it was in the context of Huawei’s bid for GE). It certainly has been a goal of the state banks (note CDBs financing of Chinalco) and i am not totally sure that their activities can be distinguished when:

    a) the CIC owns the state banks, so coordination is certainly possible (noting 2fish’s argument that such coordination is rare in practice in cHina)

    b) the fx the state banks have orginated from the recap and various government policy initiatives (swaps, required reserves that are held in $).

    I think a fair argument can be made that the CIC and SAFE have effectively hired the state banks to manage some of their fx, which would make them accountable for the actions of the state banks.

    And yes, I meant foreign equity market exposure. I don’t worry about what sovereign funds do inside their own countries; that is there business and don’t distort the global flow of funds.

  • Posted by Rien Huizer

    1.This discussion seems to assume quite a few things about the way in which the PRC is actually governed and how businesses are managed. Probably no one knows what messrs Hu and Wen can or cannot do to bureaucrats or even private sector people. probably a lot more than their formal authority when their prestige is high and very little when they would have to beg others for support. There is simply not much of a tradition of formal institutionalization and authority.No doubt the money cannot be used to help in- and external enemies of these gentlemen, even if the state would benefit financially. On the other hand, one should not expect the finance theory orthodoxy (and what about gambling against Icelandic banks?) of the Norwegians.And one should expect some agency problems..
    2. even accepting that there is no clear consensus of the defining characteristics of SWFs vs central banks or similar reserve agencies, the distinction with pension funds is pretty clear: pension funds invest money on which the retirees have first claim i.e. microeconomically there are identifiable principals who know their entitlements and have an incentive to monitor management (to what extent they do is of course another matter). SWFs are maybe like the windshield wipers of a car parked in soviet-era moscow: they belong to the people i.e. no-one.

    3. Quite a bit has been written about SWFs, much of it speculative because of lack of factual data and iffy estinates of the future (does anyone believe that current trade imbalances will go on unarrested for more than 10 years?) but it seems to me that people miss the following points, which apply to China in particular:

    – structured pools of money are merely a policy tool, and it is the circumstances and policy responses, not the tools that we should worry about
    – the obvious fact that the IMF-SWF dialogue is unlikely to produce anything that will constrain the policy response of certain countries to the circumstance that they have to invest abroad for FX management reasons (themselves highly debatable) whilst they are also still quite backward in terms of social and human capital, in fact unable to absorb the volume of physical capital that they are generating. The only effect might be that the outflow will get a different label.

    4 Ideally (assuming some sort of budget constraint growing out of the accumulation of external deficits in a small group of countries) there should be a multilateral regulatory regime that would allow complete freedom of international investment to state entities only for (a) pensionfunds (ie with identifiable claims and claimants) provided these conform to a minimum standard of disclosure and governance (b) Norwegian style funds with the same rationale (substitution of state mineral wealth by external financial wealth, but no identifiable claimants) and similar levels of governance and disclosure and (c) tribal funds (ADIA etc) where it is diffcicult to make a distiction between the state and the ruler’s tribe or alliance of tribes. However in the latter case investments should be transacted via licensed intermediaries conforming to a high level of counterparty risk management and regulatory compliance. None of these categories should have more freedom to manipulate markets that ordinary commercial investors. Anything else, and especially central banks, should restrict investment to government paper and qualifying (e.g. sponsored by the CB/bank regulator of the recipient country) bank deposits. Strategic style investment, a la Temasek and power trading a la ADIA should be regarded as potentially hostile state behaviour and dealt with bilaterally. Recipient countries should have the right to deny government stakeholders voting rights for reasons of national interest (whatever that may mean) and transactions resulting from market manipulation unwound at the expense of the initiating souvereign..This should punish those who want to sterilize but not accept the carrying cost of idle FX balances, those who want to play the lottery with state funds and spoil markets in the process, and repel those who want to undermine my national interest by exploiting temporary weakness of my country’s industrial crown jewels.

    Now, that would be a show of economic sovereingty on the part of the recipients.