Skip to content

An Unappetising Picture

I thought a comment from someone calling themselves “Sotongmee” raised enough interesting points to merit reproduction in full in the blog together with my response. This person addresses me as “sir”. I would like to say once again that that is unnecessary and far too formal. I would much prefer if you would just call me Kenneth.

Their comment is as follows:

Sir, permit me to point out that both yourself and prof Balding have used the primary surplus as the sole inflow into the government’s asset under management (AUM) to derive the theoretical asset size (after factoring the annualized 20 year return which GIC claimed). There is however a big portion of reserves, the foreign exchange surplus, which is managed by the MAS, which might have been left out.

From Singstat’s database, in the last decade between 2001 and 2011, the government’s total financial portfolio assets flow amounted to SGD$96b (presumably all were placed with GIC), compared to foreign exchange reserves accumulated of SGD$262b managed by the MAS. The MAS’s website show the official exchange reserve position at $308b as of end 2011. Even at the current 30 year US Treasury yield of 3%, the annual investment income would exceed $10b. In reality the MAS probably invested in a combination of sovereign bonds over time and the yield is likely to be several % points higher.

Last, another interesting point I observed on 2 different sets of data of the official foreign reserves: MAS’s data showed that the year 2011 OFR was $308b compared to year 2000 OFR of $138b, or an increase of $170b. Singstat’s data however showed total OFR inflows over the same period of $167b. The difference cannot possibly be the yield the MAS has received on such a huge sum of foreign exchange reserves, which as estimated above, total at least $10b a year. Where is the missing investment income for the last 10 years?

Sources: MAS:

I hope this thread gets more attention, more qualified and curious people than myself should comment and point out any errors or omissions and hopefully shed more light on the subject. There is surely more than what we are being told and keeping quiet hoping for the best is like traveling on MRT hoping we won’t be late for work.

My response is below:

@Sotongmee, you paint an unappetising picture.

I will try to address your points here. However, firstly I would like to point out that at least initially Chris Balding did not use the Primary Surplus in his calculations. He used the General Government Surplus to start with until I pointed out to him that this was likely double-counting as investment and interest income and gains on land sales would already be in the General Government Surplus numbers. He then switched to using the IMF’s Operational Surplus numbers. However this would be inaccurate as well as it does not include development expenditures which should be subtracted to leave the net figure available for investment.

The OFR which show up in the MAS’s balance sheet as assets have their counterpart on the liabilities side of the balance sheet with the S$ deposits due to the government as well as deposits by local banks with the MAS. Since the government’s Statement of Assets and Liabilities shows cash deposits of around $125 billion on the asset side I do not think MAS’s assets are included in the Statement. If MAS’s assets were included the government would have to net out the cash deposits and other liabilities against the OFR. MAS’s net assets are only some $29 billion after netting off liabilities. As MAS is a Schedule 5 company wholly owned by the government the net assets should be included in the government’s balance sheet.

My objective is to measure what the theoretical level of net assets should be if the government’s assets had earned the returns they claim as well as including Temasek’s assets and estimated revenue from land sales. I have used only the Primary Surplus and the net incurrence of liabilities carried forward at GIC’s published rate of return and subtracted interest of 3.5% p.a. on government debt to arrive at a theoretical level of net assets before Temasek and estimated land sales.

You talk about MAS’s income from the OFR as probably exceeding $10 billion a year. However remember that MAS hold foreign currencies, mostly US$, and the S$ has been appreciating steadily over the last twenty years at a rate of about 2% p.a. MAS’s income statement for 2010/11 shows a hefty loss of S$10.8 billion. This was due to the value of its foreign currency assets growing much more slowly than its S$ liabilities. While MAS made S$8.9 billion in 2009/10 it lost a similar amount in 2008/09. In fact in that year the government injected S$16.9 billion into MAS raising its capital from $100 million to S$17 billion to cover the loss that year so MAS’s net assets should definitely be consolidated in the government’s balance sheet.

MAS’s foreign exchange losses are a possible indicator of what is going on with GIC and why the levels of assets and of net assets in the government’s balance sheet are so low. I pointed this out in “Where Have Our Reserves Gone” (

It is only by reducing the rate of return on assets to 5.2% that one gets to a theoretical total assets level of roughly $720 billion which is close to the figure for total assets shown in the government’s ALS. This is in line with what an unhedged return of 7% in USD would have been in SGD terms because of the appreciation of the SGD over this period.

However when one adds in Temasek’s assets and the likely revenue from land sales returns appear to have been much worse. I calculated what would be the theoretical rate of return on assets to equal the total assets shown in the government’s balance sheet at 31 March 2011 minus Temasek assets of $180 billion and estimated revenues from land sales of $100 billion. It is only when the return on assets is reduced to a shocking 2.5% in S$ terms while keeping the rate the government pays on its debt to CPF holders at 3.5% that we are able to reconcile our theoretical calculations with what is shown in the government’s balance sheet.

The likelihood that GIC’s and Temasek’s  foreign exchange losses are compounding their losses from poor investments in other areas  just make the government’s refusal to cut the employee CPF contribution rate or allow early withdrawals inexplicable. Of course they need to keep their irresponsible housing bubble going, into which they continue to pump more air with ever bigger illusory subsidies, (“A Bulge in the Pipeline”, . But with tax deductibility Singaporeans will continue to use CPF to service their housing loans.

As my writings make clear we do not need extreme scenarios to at least partially explain why the government’s returns from investment have been so poor. However even without these we need an explanation and a rethinking of the whole asset accumulation strategy. Singaporeans should own our SWFs, Temasek and GIC, directly, as I have advocated since 2009. As I said in “Sherlock Holmes and the Case of the Missing (or Merely Hidden) Reserves”, ( there are substantial hidden assets that Parliament, the President and Singaporeans do not know about or heads should roll for  incompetence.

Unfortunately the case of the IMF loan (which I will write about next) has demonstrated that without democracy there will never be accountability and transparency, and that small elite will continue to think that they have unfettered discretion over our money, however poorly they manage it.


  1. Dear Sir, can you dumb down your articles regarding the missing funds? Illustrate with tables, graphics, flowchart etc.
    I am afraid you lost me with the current articles, which are more suited to scholarly types.


    • Thank you. What a brilliant comment.

      First I must add that this blog is primarily an examination of Singapore’s economic model, as realised by the PAP government. So it is necessarily quite technical at times and deliberately not dumbed down into political catch phrases. In fact, “Rethinking the Rice Bowl” is not a political blog at all but an economic one. I’m writing here as an economist. Having said that I think your point is entirely valid and has a great deal of merit.

      I’m not sure about graphs and charts. Most of those I see in our local blogs are vague and .. well they look pretty but they’re usually not very scientific. The biggest obstacle for me (besides the time constraint) is that the data set for a valid chart graph or picture is simply not available. That annoying lack of transparency again! So I usually put forward several scenarios, point out the gaps, suggest possible answers but no-one is able to give a definitive picture. Even our first elected President was unable to get the required figures.

      The danger is that in dumbing down I will make the arguments invalid. If you were going into a duel against an opponent with a 9ft long razor sharp sword would you want to be armed with a small broken bottle? Dumbing down would be like taking away my fine weapon and replacing it with a broken bottle. But again I am with you and I want to make the material presentable and easily understood to the non economist.

      How about this? Why don’t you and all my other readers ask me questions. Just point out where my article starts to get bogged down or highlight technicalities that you are not commonly known or ask anything. I will answer the questions here in a new article and we will all be the richer for it because your questions will surely teach me more than I can teach you.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: