Retirement financing

Retirement schemes. When workers retire, they need a certain amount of money to tide over this economically non-productive period of their lives. Social unrest can result from not having enough to live on in retirement. Due to the fact that a significant fraction of workers don’t plan for retirement (ask yourself if you proactively do) and the high downsides of social unrest, government and private corporations stepped in to provide retirement schemes. They faced two options, a defined benefits scheme where workers contribute some amount of their salary, and they get a promised set of benefits when they retire (which may not be related to their contributions). The promised set of benefits would be funded from the next generation of workers, who would work to support an older generation. The other, a defined contribution scheme, where workers would contribute a set amount at a regular period, and what they would get back was a straightforward function of what they had contributed over the course of their working lives. When they retired, they would get back their money (either as a one-time payment or series of payments) but that money would be their principal + any investment income they got from it. A retirement scheme must provide some benefit over just preserving principal, otherwise it is better to keep money under your mattress until your retirement.

The rationale for a defined benefits scheme: low volatilty. Defined contributions scheme (e.g. capitalized pension plans) typically invested in the stock market, in order to generate high returns on capital. However, in 1920-1930, those in the Western world who invested in stock markets for retirement found themselves ruined, due to the high volatility of capital returns. Wage growth, though lower than the growth rate of capital, is about 5-10 times less volatile, and is much more reliable and predictable.

The two drivers of a defined benefits scheme – productivity and demography. A defined benefits scheme (AKA a PAYGO or Pay As You Go system) directly transfers wages from active workers to retirees. PAYGO creates “intergenerational solidarity” since the rate of return for the retirement scheme is drawn out of workers’ wages, which is limited by the growth rate of the economy. Therefore workers have a direct interest in ensuring wages rise as fast as possible in the future, by having more children, and investing in schools and universities to train those workers. (There is a minor distinction to be made between funded and unfunded defined benefits schemes – in an unfunded benefits schemes the wages are directly transferred, but in a funded benefits scheme the wages may be held in reserve to pay future retirees. The distinction is in the amount held in reserve, and in both cases the direction of money transfer from the active working to retirees is the same). In mid-20th century continental Europe when these programs were conceived, both productivity growth and demographic growth were high. Consequently economic growth was about 5%. In the 21st century, economic growth is projected to be 1.5%, driven by a falling birthrate. Additionally, older workers began to live much longer (increasing the retirement funding burden). Together, the flattening pyramid of an increasing number of elderly at the pinnacle and a decreasing number of young workers at base are called “the demographic transition” and they have made a PAYGO system unsustainable. Therefore many governments are breaking promises made to an older generation of workers and cutting benefits. Additionally it is unclear if productivity gains will continue – a metaphysical hypothesis that the 20th century was our “miracle” century is called a “Great Stagnation” hypothesis – where we picked low-hanging technological fruit. Another hypothesis is that the 20th century was powered by cheap oil and coal, dirty energy that cannot be continue to be used due to its role in warming the planet, potentially catastrophically.

The flip-side of the coin. The defined contribution scheme has two flavors, one run by government and the other by private corporations. It appeals to “individual responsibility”. The Provident Fund (a defined contributions scheme) was a fund to provide for subjects’ retirement adequacy, funded purely by their own contributions earlier in life. After retirement, subjects would receive a monthly payout from the Provident Fund. The Provident Fund was set up by the British in Malaya in 1955, and was consequently exported to their colonies in Sub-Saharan Africa. The goal of the Provident Fund was to reduce reliance on public assistance from the colonial government (for aging retirees), and it is probable that the British, setting up a welfare state in Britain, set up these Provident funds to minimize the public obligations on the British empire. The Provident fund was taken over by the newly independent Singapore government, and was called the Central Provident Fund (CPF).

Increasing complexity. A pure retirement account features a large amount of idle money which could be put to better use. Subsequently the Singapore government has allowed 5 major uses of funds during the contributor’s working life over the years – homeownership, family protection, healthcare, investment, and education. (Aspalter, p173). In addition, there was also a structural change to CPF made in 2009, where for the first time the possibility of unfunded liabilities arose – because the government introduced an annuity scheme where they took on the risk, not any third-party insurer. CPF Life was an annuity that is mandatory for every citizen meeting a low minimum bar to buy, with guaranteed income for each citizen in retirement. The CPF Board is not longer merely an administrator of funds, but also a shadow insurer with actuarial responsibilities to ensure that the annuities given for each citizen do not overly exceed their contributions, creating unfunded liabilities that require transfers from the Budget.

However a side effect of allowing these options has been a rapid increase in the minimum sum set aside in the CPF, which is the amount locked into the CPF system until the Retirement age, in order to preserve the original goal of retirement adequacy.

Combatting volatility through government guarantees. The value-add of a defined contribution scheme is in its returns, and therefore in the expertise of the fund managers. CPF funds are deducted from the taxpayer, and given to the CPF Board for administration. The CPF Board buys Special Government Bonds from the Monetary Authority of Singapore (MAS), which invests it in the Government of Singapore Investment Corporation (GIC), one of Singapore’s two sovereign wealth funds. However, as already mentioned, the key drawback of this scheme is in the volatility of its returns. The Government explicitly guarantees a 2.5% nominal interest on CPF balances. (Currently there are three sets of Government guarantees: 2.5% nominal interest on all CPF accounts, an additional 1% interest rate on first $60,000 of all combined balances, and SMRA (Special, Medisave and Retirement Accounts) are pegged to 12 month average yield of 10 year Government Securities (10YSGS) + 1%. The 10YSGS were pegged to 4% until 31st December 2013.)

4 questions occur to me as to the wisdom of the system:

  1. The current CPF is endlessly complex (as is every social security system in the world, to increase the options available to the individual citizen). Is it wise to keep it in its current form, or is it better to split into three? It seems that there should be a clear separation into three major categories for CPF – pure retirement adequacy (the bare minimum for the retirement account, and the compulsory purchase of CPF Life), insurance (family protection, and healthcare through Medishield), and investment (education, investment, and home ownership).
  2. Should people have the freedom to opt out of CPF? (with the caveat that the government will not provide for you should you fail to provide for your own retirement)
  3. How can we prevent an elderly retired underclass? Asher and Bali in a 2013 paper (Asher, Bali, p183) mention that CPF real rate of return (1.42%) trailed annual real wage growth (5%) and real GDP growth (7.9%) between 1987-2011. This implies that retirees will be in a relatively poorer position over time against the economy, and will not participate in the country’s future growth – possibly creating a disenfranchised class of old folks who relied purely on CPF savings. We see this in the “tissue aunties”, old ladies who go around hawker centres to sell tissue paper, and cardboard collecting elderly, who are the face of urban poverty in Singapore. It seems there are two options – increased government disbursal, or encouraging supplementary private retirement savings. One option considered by Asher and Bali for government disbursal is for GIC to return more of the differential between their inflation-adjusted return of 5.3% per annum and the minimum nominal return of 2.4%. Another option is to somehow encourage more saving in the working population besides their CPF. However, Singapore is one of the most financially unsustainable countries to live in, with housing and education taking up huge chunks of the working population’s capital. With such burdensome expense over the working populations head, where are the savings going to come from?
  4. The trade-off between the two systems in numbers. The government guarantees of a minimum of 2.5% in nominal returns reduces volatility to almost zero. How does the average CPF rate of return compare with PAYGO system (with similarly low volatility)around the world? Is it higher?


  1. Shamugaratnam, Tharman - Singapore Deputy Prime Minister and Minister of Finance's 2015 Budget Speech on CPF.
  2. Asher, Mukul G  and Bali, Azad Singh - Fairness and Sustainability of Pension Arrangements in Singapore: An Assessment
  3. Aspalter, Christian, Discovering the Welfare State in East Asia.