MICA’s Response Fails to Reassure
A few days back there was a comment posted on my blog by @Lengyiren drawing my attention to a posting on www.gov.sg saying that the government had rebutted my claims about the reserves. The link to the so-called rebuttal is http://www.gov.sg/government/web/content/govsg/classic/factually/factually-041012-istheresomethingwrongwithourreserves.
This portal is maintained by the brave people at MICA who defamed me in the WSJ by claiming that I had misrepresented basic facts about my father’s bankruptcy and ensured through their control of the media that their version was printed while my right of reply was denied. One can see the MICA trademarks of sloppy editing and elementary grammatical mistakes such as saying “are flowed to” instead of “flow to”.
Actually they do not mention me directly referring instead to “some online postings.” And in fact since it was Christopher Balding and not me who made the mistakes to which they refer, perhaps the post was solely directed at him. Nevertheless I felt it important that I respond lest readers think that I am guilty of the same errors.
As my response is rather long I have divided it into two parts. Part 1 refutes any implication that the errors they allude to in their posting apply to my calculations. Part 2, which will follow shortly, will deal with the claim that there are sufficient checks and oversight over our investment entities. In particular, the claim that with the Elected President and the Auditor-General to oversee them as well as their Boards we do not need transparency or accountability to Parliament or the people.
I reproduce their posting in italics below with my response underneath it in bold.
Some online postings attempt to estimate the size of GIC by using data on budget surpluses and issuance of Government securities to estimate fund flows into GIC over the years. When their estimate of GIC’s size exceeds other market estimates, they conclude that funds have therefore gone missing. Where do they go wrong?
Our reserves cannot ‘go missing’ (see item 4).
Some of the confusion created by these recent “estimates” of GIC’s assets arises from the following errors:
First, they assume that all the Government’s available funds are flowed [sic]to GIC alone. The Government has significant deposits placed in MAS. As of 31 March 2012, the Government has $147 billion deposited with MAS, compared to MAS’ Official Foreign Reserves (OFR) valued at S$305 billion. MAS has a significant proportion of its portfolio invested in liquid financial market instruments and hence earns a lower rate of return than GIC.
This cannot refer to me since I have assumed nothing of the kind. Of course I am aware that the Singapore government’s balance sheet includes a considerable amount of cash parked at MAS and that this should yield less than the assets managed at GIC. That is why when I calculated the theoretical level of reserves in “Where Have Our Reserves Gone”, (http://sonofadud.com/2012/09/07/where-have-our-reserves-gone/) I used a rate of return of 6.2%p.a. rather than the 7.2%p.a. in US$ GIC claims to have earned over the last twenty years. I further reduced this to 5.2% to take account of the appreciation of the S$ and the effect this would have had on returns in US$ when translated back into S$. My calculations showed that the level of net assets should still have been nearly twice what the Singapore government’s Statement of Assets and Liabilities showed them to be. I reproduce the table below: Summarizing these results in the form of a table:
|31/3/2011$ billion||Assuming 7% Return and 3% Cost of Debt||Assuming 7% Return and 3.5% Cost of Debt||Assuming 6.2% Return and 3% Cost of Debt||Assuming 5.2% Return and 3.5% Cost of Debt|
|Theoretical Net Assets||884||834||746||604|
In the face of the government’s refusal to provide the necessary information to its citizens, I made assumptions that the present value of the revenue from land sales over the past thirty years was of the order of $100 billion and also that the Total Shareholder Returns including dividends from Temasek have been $180 billion. Both of these estimates may be conservative. I later calculated that the rate of return earned by GIC would have to be as low as 2.5% in S$ to square the reported level of total assets with the cumulative Primary Surplus number, a reasonable estimate of revenues from land sales and Temasek’s TSR (“An Unappetising Picture”, http://sonofadud.com/2012/09/25/an-unappetising-picture/). If we make allowance for a low return on cash deposited with MAS (assume that the MAS pays the government 1%) then the rate of return rises slightly to 2.875%.
Second, debt servicing costs are sometimes ignored in these estimates. This results in an over-estimate of the assets accumulated through investing the proceeds from the issuance of Government securities, especially over long periods of time. For more information on the Government’s debt position, please refer to the MOF website [http://app.mof.gov.sg/sg_borrowings.aspx].
This is certainly not true of my analysis as anyone who has read my postings will know. I have assumed that the cost of debt (mainly securities issued to CPF) was 3.5% over this period which is somewhat conservative considering that Ordinary Accounts only receive 2.5%. Presumably, MICA are referring to Chris Balding’s analysis. I pointed this out to him when I met him in Hong Kong and went over his analysis with him. However I am not sure he has taken this on board yet.
Third, they overestimate the funds flow into GIC by including the interest and dividend income that the Government gets on its investments. These estimates incorrectly assume the full amount of government budget surpluses as fresh fund injections, without first removing the interest and dividend income portion.
Again, if it is meant to refer to me, then mea non culpa. As far as I could I used the figures for the government’s Primary Surplus which is Operating Revenue minus Operating and Development Expenditure. However I was hampered by the fact the Department of Statistics kept changing its definition of the Primary Surplus. Up until 1998 Operating Revenue included interest income which should properly be part of the General Government Surplus. I have tried to adjust for that as much as possible given the limited and often contradictory data available.
As can be seen from the fact that the IMF have restated their data for Singapore’s General Government Surplus twice this year I am not alone in this. The government have still not explained why the IMF made such a large error and why, in adjusting their numbers towards the Department of Statistics figures, they created further discrepancies (“Sherlock Holmes and the Case of the Missing Reserves”, http://sonofadud.com/2012/09/23/sherlock-holmes-and-the-case-of-the-missing-reserves/)
Before he met me, Chris Balding was using the general government surplus numbers and then compounding these surpluses at a carry rate of 7%. I explained to him that to use these numbers as a basis for calculating the level of theoretical assets would be double-counting as these numbers would already include dividends, interest income and capital receipts. After that I believe he used the Operational Surplus numbers from the IMF which were also too high because they did not subtract Development Expenditure.
I assume therefore that MICA are referring to Chris Balding. My analysis did not “go wrong” in the manner they describe. It includes the cost of government debt and tries as far as possible to exclude the possibility of double-counting by only using the Primary Surplus numbers. Also I have made assumptions about the value of revenue from land sales and about Temasek’s TSR which I believe are conservative in the absence of anything like full disclosure.
The government has failed to explain why returns appear to have been unacceptably low. Part 2 will refute its claims that without transparency and accountability there can be adequate oversight of our investment entities.