PensionReforms' summary and comments
This 2003 report, by one of PensionReforms' contributing editors, looks at the case of Mauritius - how a relatively poor country introduced a universal, Tier 1 pension and what happened next.

" ... every elderly resident of Mauritius receives income support from a system of non-contributory pensions that dates from 1950.  Subject only to minimum residence requirements (12 years from age 18 for citizens, 15 years from age 40 for non-citizens), every resident aged 60 or over is eligible for a monthly pension that amounts, in the current fiscal year (2002/2003) to the following:
. age 60-89: Rs 1,700 (USD 58)
. age 90-99: Rs 6,400 (USD 220)
. age 100+: Rs 7,300 (USD 252)."

The pensions are neither income nor retirement tested.  They are treated as ordinary income for tax purposes so, in the progressive regime, higher earners pay proportionately more tax on the pension.

In real terms, these pensions range (in 2003) from about 18% of per capita GDP for the smallest to about 92% for the most expensive (including a disability supplement).

The pension was introduced in 1950 as a stop-gap as the government had been sitting on recommendations to introduce a Defined Contribution, Tier 2 system.  In 1950, the pension was payable from age 65 and was fully offset by 'other income'.  A later attempt to start the contributory scheme led to the income test's abolition in 1958.  But again, the government thought that a temporary measure.  Between 1965 and 1978, a "mild" income test was re-introduced.

The contributory Tier 2, an early example of a Notional Defined Contribution (NDC) scheme, finally emerged in 1978 but seems to have reinforced, rather than undermined, Tier 1.  That's because the Tier 1 pension was not only increased but the income test was also abolished, probably helped by the surplus of 'contributions to benefits' in the Tier 2 scheme.  Between 1978 and 2000, the real value (after allowing for inflation) of the Tier 1 pension has increased by 2.6 times.

Further improvements were added to Tier 1 in 1983 and 1995 with extra pensions for the over 90s (at now 3.8 times the basic pension) and then the over 100s. (at 4.3 times the basic pension).

The ratio to the average Tier 1 pension to per capita GDP has moved less dramatically from 18.3% in 1978 to 21% in 2001 but has fluctuated somewhat - in 1975, it was as low as 11.1%.  The economic condition of Mauritius has improved somewhat over the years.

So, will all this become unaffordable as the age dependency ratio grows?  Not necessarily, the report argues.  Three numbers really matter when assessing affordability - per capita GDP, the amount of the pension and the proportion of pensioners in the population.

".the cost of pensions (as a proportion of GDP) is equal to the proportion of pensioners in the population times the ratio of the pension to per capita GDP.  GDP per capita is a crucial variable; with larger incomes, many things become affordable, including generous pensions for large numbers of retirees."

The report points out that Mauritians are likely to have a somewhat higher standard of living in 2040 than now (about 76% higher according to local projections) and a relatively modest 3% reduction in that through taxes is expected to pay for the pensions.  So, even though the cost of pensions will rise from 1.9% of GDP to 4.8% (2.5 times), the reduction in the relative standard of living of workers in 2040 is expected to be modest.  Pensioners will therefore share in the future prosperity of Mauritius (as they have in the past).

Other reforms are possible - increasing the earliest pension age from 60; linking pensions to prices rather than wages or re-introduce the means test abandoned in 1978.

The report looks in some detail at the income-testing option and concludes that a surcharge through the income tax system would be possible if it withdrew the Tier 1 pension gradually.  That would limit the damage an income test does to what have been described as the four major social costs of means tests, namely:
·   "Disutility and stigma
·    Informational distortion
·    Incentive distortion
·    Administrative and invasive losses"

The report also details the NDC Tier 2, designed to replace one third of covered earnings.  It is compulsory for covered workers - all those over 18.  The exceptions -  the very low paid, the self-employed and civil servants - mean that only about half participate and not all of those are consistent contributors.  "Pension points" are awarded for contributions and are periodically re-valued.  The accumulated value at retirement is exchanged for an annuity on the basis of 1 rupee for every 11 rupees in the scheme.  Although the points were originally supposed to increase in line with wages, in fact, they have been re-valued more in line with prices.

Most (80%) of the money in the National Pension Fund is invested in government bonds, 10% in house mortgages and 10% in company shares and bonds.  The main effect of the NPF:
".has been to ease the budget constraint of Government.  Government has been able to spend more, without the necessity of collecting additional taxes, because it has access to the contributions of workers who are forced to save for their own retirement."

Using Mauritius as an example, the report looks at the justifications for the reasons that governments:
"..are attracted to contributory pensions as a tool to advance any or all of the following goals:
. Increase national saving;
. Avoid redistribution of income and wealth;
. Ensure that living standards of workers do not fall in retirement;
. Build up a fund for government use."

The first (increasing national saving) is at best controversial; the second (avoiding redistribution) is arguable in the case of Mauritius because of the scheme's design and investment policy; the third (protecting living standards) is not working as designed because the value of the points awarded have not kept pace with wages.  However, the government has made substantial use of the National Pension Fund.  That has probably done more for the Tier 1 scheme than for Tier 2 benefits.

As the report notes, the case of Mauritius provides some valuable insights for students of public policies on pensions:
"In conclusion, the universal pensions of Mauritius are affordable, even with no change in the qualifying age for an old age pension, and even if pensions increase with per capita GDP to accompany changes in the standard of living of the country.  If the qualifying age were to increase from 60 years to 65, which could be justified because of increased life expectancy, the pensions would become even more affordable.  It comes down to political will, of collectively deciding whether universal pensions represent good value for taxpayers' money."

PensionReforms suggests a number of lessons may be learned from this report - first, being poor does not limit a country's capacity to introduce a universal Tier 1 pension.  Secondly, the real purpose of a Tier 2 scheme must be clearly established and governance structures put in place to increase the probability of those being achieved.  That seems not to have happened in Mauritius.  Thirdly, and probably most importantly, governments probably need to stick to things where they can add value; where they have a unique capacity to change things.  Mauritius shows how that worked for Tier 1 and how it has seemingly not worked for Tier 2. (File size 592 KB)  216