| Individual retirement accounts are less efficient than state pensions - Government needs to reassess its strategy |
Back |
| The Pension Board’s latest report on mandatory pensions recommends an increase to the current system of state pensions plus individual retirement accounts. Shane Whelan, however, writes that some of the figures in the report are ‘blasphemous’, and argues that the higher administration charges associated with moving to individual retirement accounts will ultimately reduce a pension by about 20 per cent. Instead, he recommends that the Government uses the current state pension system as a starting point, which ‘enjoys near optimal economies of scale and the efficiencies of simplicity’ and restructures it along the lines of a financial contract. |
If mammon is the new god, then economists are the new theologians. Economists attribute mammon with omnipresence, directing actions through Adam Smith’s ‘invisible hand’. They claim mammon in purest form – the stock market – is omniscient, or, equivalently, in their more formal language it is ‘informationally efficient’ and achieves ‘Pareto optimality’. Naturally, some of the more zealous sects within the economist community seek to make markets omnipotent by extending their reach into ever more spheres of human activity.
The latest sphere they wish to convert is pensions. How a society treats its dependents, including its aged, has long been a familial contract, but had evolved over the last couple of centuries into a social contract, as family ties weakened with industrialisation. The next phase, starting in Chile in 1981 and spreading around the developing world by that missionary sect, the World Bank, is now replacing the social contact with the financial contract. Specifically, they want every worker to have their own individual retirement account investing in stock market securities, which will provide for them when unable to work.
The UK is almost converted. Ireland has before it the Pensions Board’s 500 page tome, Special Savings for Retirement: Report on Mandatory Pension System. The report ‘recommends that the most appropriate and practical approach to improving the position of pensioners in Ireland would be a combination of an increase in the State pension with a mandatory supplementary system…and individuals would hold Special Savings for Retirement Accounts’ (p. 94). So before us is the Irish compromise: the current system of state pensions plus individual retirement accounts. The recommendation is backed up by a debate-stifling appendix stretching to 355 pages of figures that estimates and contrasts the effects of the proposed system on the economy with alternatives. Such figures are, of course, as accessible as Latin to most.
On deciphering, the figures presented in these appendices are blasphemous - or so it is claimed in a paper of mine just published in the June 2007 Quarterly Economic Commentary of the Economic and Social Research Institute. The figures presented are inconsistent with market values. They assume that individuals will fill their retirement accounts with risky assets and, na?vely, do not model the consequences of the risks assumed. The perverse logic running through all the costings is that the more investment risk that is taken on, the higher the ultimate pension. Indeed, it can be demonstrated, reworking with their figures, that there is no pension crisis at all because the state can work financial miracles by achieving a real return of 4.6 per cent per annum, through investments in the National Pensions Reserve Fund, on bond borrowings that cost a mere 1.75 per cent real return per annum to service. Ignoring investment risk and its consequences will inevitably lead to such inconsistencies.
But the stock market does not ignore risk. In fact, the whole point of the market is to price and transfer risk in an economy. In his preface to A Tract on Monetary Reform (1923), J.M. Keynes remarked, ‘It is often supposed that the costs of production are threefold, corresponding to the rewards of labour, enterprise, and accumulation. But there is a fourth cost, namely risk; and the reward of risk-bearing is one of the heaviest, and perhaps the most avoidable, burden on production.’
All the figures change dramatically when proper allowance is made for the market price of risk. We can illustrate the impact in a simple, stylised way. This shows the present value of a pension, payable for 20 years from age 65, as a function of the investment return assumed. For simplicity, we plot the assumed investment return above wage escalation, implicitly assuming that pensions will increase in line with wages.
Assuming investment returns is in line with wage increases the present value of a pension is 20, irrespective of the current age of the pension saver. This is because the pension of one unit is payable for 20 years, so at a 0 per cent rate of discount it has a present value of 20 units. As the assumed investment return increases relative to wage increases the present value of the pension reduces, and reduces faster the younger the pension saver as the discount period increases.
However, it can be shown that low risk investments (e.g., index-linked bonds) deliver returns of the order of 0 per cent above wage escalation, while more risky equity investments have historically delivered returns of the order of about 3 per cent per annum above wage increases. Indeed, low risk investments typically have a large guaranteed component (hence they are low risk), so we can say with a high degree of confidence that the ultimate return will be close to the expected. We cannot have anything like that confidence for the expected return from risky investments – paradoxically, we can anticipate that we will not get what we expect from such investments.
Going back to the figures, we can make a crude, but insightful, identification between market values and present values. The market value of a pension backed with low risk investment is of the order of 20 (as we discount at 0 per cent) while for equity investments (as we now discount at a rate of 3 per cent) it is of the order of is significantly lower – for a 40 year old, it has a value of just 7.2 (as highlighted by an arrow). The market cost of the pension to a 40 year old is 20 assuming low risk but falls to 7.2 when assuming equity risk. We conclude that the market cost of the investment risk is a whopping 12.8 (that is 20 less 7.2) or two-thirds of the market value of the pension (12.8 divided by 20).
The market cost of the pension is 20, not the 7.2 produced by the methodology employed by the Pensions Board. Their figures are not market consistent, a cardinal sin in the new creed. The proposed new pension system is fundamentally flawed, built on unreliable returns from risky investments. Such a foundation will give way with another 1929, 1974, or a run of over two decades with a negative real return as delivered by the world equity market in the first two decades of the twentieth century, or a run of over 50 years with a negative real return as delivered by the French and German stock markets or, in short, the materialisation of some of the risks that are factored into the current price of risky assets.
The paper, Valuing Ireland’s Pension System, comes to a simple but significant insight when all the figures are put on a market consistent basis. It shows that the higher administration charges of moving to individual retirement accounts of about 1 per cent per annum in the saving period will reduce the ultimate pension by about 20 per cent. That is, the key difference between pension systems is in their administration efficiencies – the size of pensions out for contributions in. The current state pension system enjoys near optimal economies of scale and the efficiencies of simplicity that produce pensions of the order of one-fifth higher than the proposed individualised accounts for the same level of contributions. A sustainable pay-as-you go system, reassuringly like our current system, can easily be developed from our current starting point. It requires that the current state pension be structured more as a financial contract, with the associated financial discipline, than the ill-defined, politicised social contract it has been to date. This requires the removal of political discretion from state pensions – a challenge not to be underestimated – but surely justified by a pension 20 per cent higher for all. So refined, our current pension system will be engineered to last another 100 years. |
Dr Shane Whelan is an actuary at the UCD School of Mathematical Sciences.
|
| Article appeared in the August 2007 issue.
|
|
