Thoughts on response to CPF speech

Since my speech on the proposed CPF changes last week, and after thinking hard about the Government's responses (and in particular, Minister Tharman's statement which I thought was particularly pertinent to the points I was raising), I have some additional thoughts about the points I made. I had 3 main thrusts:

• The rate of return for CPF members can, and should be, higher.
• The compulsory annuity represented an unprecedented inroad on CPF members' rights over their CPF funds.
• The adequacy of the proposed changes in addressing Singaporeans' long-term needs.

The first point has attracted the most attention and the most response from the Government. Indeed, 2 PAP MPs (Mr Ong Kian Min and Mr Sin Boon Ann) who spoke the day after me largely concentrated on that point. Mr Low Thia Khiang also engaged Minister Ng on the issue of the CPF funds. So this post will primarily focus on the first point, and I will only touch briefly on the other two points.

Rate of return for CPF members

I am going to keep things simple, by breaking them (including my further thoughts) down into bite-sized bullet points. To recap, my argument was that:

• The CPF Board takes CPF members' balances and purchases bonds issued by the Government (specifically, the Ministry of Finance (MOF)), at coupon rates equal to the interest that the CPF Board has to pay to its members for those balances.
• MOF apparently takes the funds raised and invests them with the Government of Singapore Investment Corporation (GIC).
• GIC manages the funds invested by MOF. It presumably makes a profit on those funds (and gets paid management fees). In the first 25 years of GIC's existence (1981-2006), GIC's annual returns on funds under management were 8.2% in Singapore dollar terms. It has estimated that its returns for the next 25 years should be 6-8%.
• GIC pays the profits to MOF. MOF pays the CPF Board when the bonds mature, and keeps the excess.
• Presumably, if GIC had made a loss, MOF would still pay the CPF Board as per the coupon rates of the bonds, and would then make a loss.

Taking Minister Tharman's and Minister Ng's statements together, I think the Government's position can be summarised thus:

• CPF members' returns are 100% guaranteed. They are completely risk-free, and the capital will never be affected.
• Any sort of rate of return tied to equity investments as per GIC or Temasek Holdings (although I would exclude Temasek from consideration -- it is a private equity fund, and its historical annualised returns of 18% mask the tremendous amount of risk which is, to my mind, wholly inappropriate for CPF members) would necessarily include exposure to market risks, including risk of capital loss. That is not acceptable.
• The rates offered by the Government are very competitive (possibly better than market) for such risk-free investments.
• Commercial fund managers have declined to operate any funds for the CPF Board to compete with the Government.
• Any profits made by MOF from investing the funds go back into the Government's budget. The Government will use the budget for Singaporeans, and will return any budget surpluses to Singaporeans, particularly the lower-income ones, e.g. Progress Package, etc.

And here are my responses:

• The re-distributive objective stated by Minister Tharman (i.e. to put the gains from investing with GIC into the budget and then funding programmes to help and/or return money to the people) is praiseworthy. It is a Very, Very Good Thing. In fact I cannot overstate how much I admire, agree with and support that. But I have to caveat my support for the re-distributive objective.
• Firstly, any re-distribution back to lower-income Singaporeans will be in a form, and at a time, determined by the Government. It is not surprising that budget surpluses are returned to the people just before election time.
• Secondly, there is no guarantee that any budget surpluses resulting from the gains from GIC will be re-distributed back to the people. They can be spent on other expenditure items (e.g. defence), or even used to fund tax cuts. Take the US, for instance. Clinton left the government with a budget surplus, which has since turned into a massive budget deficit under Bush, thanks primarily to tax-cuts that benefited the rich, and massive military spending to fund Bush's wars. So there is no guarantee that the money will be returned to people. They can be spent in many, many other ways. Until and unless there is an institutionalised system around how the gains from investing CPF funds can be used, I am disinclined to place that much weight on the re-distributive factor.
• Thirdly, there are some unstated assumptions there, such as that better-off CPF members have not withdrawn their CPF funds for investment purposes (which is something I believe the richer are far more likely to do than the poor). (Note to o: see this link for details on what can investments can be made with CPF funds.) Another related, unstated assumption is that the aggregate amount of CPF balances belonging to better-off CPF members is sufficiently larger than the aggregate amount of CPF balances belonging to lower-income CPF members. To the extent that either assumption is incorrect, the re-distributive effect is reduced, since it then increases the proportion of the GIC gains that are generated from CPF balances belonging to the lower-income anyway. These assumptions have neither been validated nor debunked, but since it is the Government that is asserting a re-distributive effect, the Government should confirm this.
• As I pointed out in my speech, this entire mechanism is effectively a large, regressive tax on CPF wealth. With the CPF cap now fixed at a relatively low $4500 per month, the rich will have a much smaller proportion of their wealth entering the CPF system in the first place. On the other hand, lower-income CPF members are much more likely to have a bigger proportion of their wealth in CPF.
• It should also be remembered that when the Government crafts its "re-distributive" budgets, it typically weaves in its socio-economic goals into the policies. For instance, the Government wants to encourage families and discourage singlehood, so some/many tax breaks are not available to singles especially single mothers. That essentially results in a re-distribution of wealth from certain "undesirable" groups (such as singles, single mothers, etc.) to other groups, regardless of wealth. And indeed, it is quite arguable that singles are precisely the ones who need most desperately to ensure financial self-sufficiency in old age.
• The Government has pointed out that GIC investments are not risk-free. And it is correct. But if MOF is bearing the risk in investing with GIC, then it must follow that Singaporeans, as tax-payers, are indirectly bearing the risk. So the risk is always borne by us, it is just a question of whether it is directly as CPF members, or indirectly as tax-payers. However, if it is the former, then at least we have the opportunity to benefit directly from the upside. If it is the latter, our ability to benefit is left entirely at the discretion of the Government.
• Taking the re-distributive objective further, it is actually the lower-income who effectively bear most of the risk of investing with GIC. Since gains and surpluses will be spent on them, therefore losses will mean programmes meant for them will be cut (since there are no longer any gains and surpluses to fund these programmes). In other words, the lower-income will be the most badly affected by poor investment returns. Better-off Singaporeans will not be affected either way, since they are likely to receive far less regardless of whether GIC returns positive returns.

Because of all of the above points, and bearing in mind the fundamental distastefulness of using my money to invest and then not permitting me to partake of the fruits of the investment (arguments about fungibility and different transactions aside), I continue to disagree with the Government.

I also want to mention a rebuttal from Mr Zaqy Mohamed who spoke after me. He talked about the relationship between risk and returns (which is true), and then said:

"So I am not too sure if Mr Siew were in the Minister's shoes, whether he is willing to sign off on such a risky policy just to accept higher returns. And I am not so sure if he is willing to be accountable to explain to CPF members if their 30-40 years of CPF savings were to be compromised years from now."

By that logic, who should be accountable to explain to CPF members why past and current rates of return may insufficient for their retirement, and why is it that the Government can, by the stroke of a pen as some have mentioned, simply increase the returns as it has just done? So I don't think that comment was fair, and in any case I do not agree with Zaqy's comment at all.

Indeed, prior to my speech, I had thought of an idea that will allow CPF members to enjoy at least part of the upside from having GIC invest their funds, while guaranteeing a minimum rate of return without making the CPF scheme an "interest rate subsidy scheme" (as Minister Tharman called it) or imposing open-ended financial burdens on the Government. But I had omitted from my speech to avoid confusion and misunderstanding.

So, this is my idea:

• This applies only to SMRA funds, which Minister Ng has said normally stay within CPF for 30 years. I chose this because this idea requires a long investment timeframe to work. Also, these funds are not (or rather, less) susceptible to usage or early withdrawals than Ordinary Account funds.
• The CPF Board again purchases bonds from MOF. But this time, the bonds are structured so that MOF will invest the funds raised with GIC, and MOF will pay to the CPF Board the higher of (i) a minimum rate of return (e.g. what the Government is currently offering, or maybe 0.5% or 1% less), and (ii) the actual nett returns obtained by GIC (i.e. after GIC takes its management fee which is presumably low and reasonable).
• I will refer to the CPF-guaranteed minimum rates as "CPF-Rate". I will refer to the actual returns earned by GIC as "GIC-Rate".
• There is an additional condition in the MOF-issued bonds. If at any time, GIC-Rate is less than CPF-Rate (which means that MOF has to top up the difference so that the CPF Board receives the actual CPF-Rate), then MOF has first priority on the differential between GIC-Rate and CPF-Rate when GIC-Rate becomes more than CPF-Rate, until such time as MOF has recouped all amounts paid by it in topping up the difference between GIC-Rate and CPF-Rate (which can include a reasonable interest rate on those amounts, e.g. CPF-Rate). Once MOF has fully recouped all amounts paid by it to top up the difference, it ceases to have any claim to the differential between GIC-Rate and CPF-Rate, which will then continue to be paid to CPF members as an enhanced return on their SMRA balances.
• This scheme has the following effects: (i) CPF members have the opportunity to enjoy the potentially superior returns from the GIC-Rate (6-8% per year on average), while still having a guaranteed minimum rate of return and without the risk of capital loss, and (ii) the Government will, over the long term, not be out-of-pocket on its guarantee to the CPF Board.
• There may be short-term risks due to short-term fluctuations, which the Government will bear. Indeed, the Government is much better-placed to bear those risks. But over the long term, those short-term risks will be evened out. The assumption underlying the MOF's decision to invest the funds raised from issuing bonds to the CPF Board must be that over the long term, its returns will exceed the CPF Board's guaranteed returns. That must be so, otherwise MOF would be expecting a loss and CPF would then indeed be an interest rate subsidy scheme.

I would be happy to see if anyone has any comments (either posted here or e-mailed to me privately) on this scheme.

Compulsory annuity and CPF members' ownership rights

I will be brief here. Nobody really responded to this point that I made. But a Straits Times article on Saturday questioned if I had forgotten about Medisave when I made this point.

No, I had not. But Medisave is fundamentally different. Yes, it channels funds into a specific account where those funds can be used only for a stipulated purpose. But if I never fall sick, then I will never have to use those funds. I don't have to spend them. They are there, I can only use them for a certain purpose, but I don't have to use them for that purpose. That is the crux: I am not compelled to use those funds to purchase anything. Which is quite unlike the compulsory annuity, for self-explanatory reasons.

I was a little annoyed at the Straits Times article. I had spent quite a bit of time with one of the journalists who put that spread together, and would have expected the courtesy of an opportunity to respond if they were going to make that sort of unfounded assertion which basically misunderstood my point. Well, so be it.

Adequacy of proposed changes

In all the policy and research papers I've read in the course of preparing my speech (and I've had to read quite a few published by different organisations and authors), they always went back to a term that, a little surprisingly, does not seem to have made its appearance in last week's debate: replacement rate.

That term simply refers to how much of a person's regular income will be "replaced" by his retirement income. Retirement experts typically recommend two-thirds to three-quarters, to maintain an appropriate lifestyle. Statistics for CPF are not available, but a 2001 study estimated Singapore's replacement rate to be 28% based on the then-prevailing rules. The rules have changed, but it is an open question whether the replacement rate, even with the new changes, would have increased by much.

Indeed, that was something I had alluded to. How much can the compulsory annuity payouts get us? If it is so little ($149 to $178 in today's dollars by 2042, figures courtesy of Leong Sze Hian by way of The Online Citizen), then surely we will need other arrows in our quiver of policy responses for a satisfactory outcome?

One of the things that the Government mentioned was how so many people in Singapore owned their homes, which can be monetised. Well, firstly it is difficult for many people to monetise their homes (either by renting or selling) -- where do they stay after that? Not everybody has children or second homes.

Secondly, if everyone chooses to sell at the same time to monetise the single biggest asset they have, then can you imagine the effect it would have on the property market?

Finally, what about those who live in rental flats and do not own their own homes? What do they monetise then?

I noted that the Government went to great pains to illustrate how much more money people would have under the changes. Yet, there were no references to replacement rates or similar concepts.

The hour is late (3am!) and I'm tired. But I remain unconvinced that the proposed changes are adequate, or fair to the people.