Pension Retirement Mauritius

The National Pension Fund and its financial implications on the economy of Mauritius Chapter 1: Introduction The philosophy of the National Pension Fund (NPF) includes the idea that one ought to earn a reasonable proportion after pension age of what one earned during one’s working life. If you have contributed to the NPF and built up your pension points, you will get a pension which, when added to your old-age pension will be a reasonable. The National Pension Fund scheme is proposed as another mandatory saving for retirement. Once it is set up, the NPF will fit into Pillar 2 of the Multi-Pillar Model of the World Bank. The NPF nevertheless will not replace provident funds or retirement mutual funds, but rather improves saving channels for future retirees. Mauritius is found in the developing countries group where contractual savings, savings with insurance companies and pension funds exceed 40 percent of Gross Domestic Product and which represent a greater potential force in the domestic financial system.

Pension funds account for 75 percent of contractual savings. The pension system is a balanced and well-managed multi-pillar. In Mauritius there have not many authors that have write specifically on that subject, that is, financial implication of National Pension Fund on the Mauritian economy. I have mainly used the research made by other analysts in other countries and try to apply it on the Mauritian economy. Obviously the result will not be the same, but try to make an estimate of it. Objectives of that Project:

• Analyse the overall financial implication of NPF

• Testing the financial effect of NPF on national savings

• Estimating the relationship between fiscal balance of Mauritius non-retirement account and the net saving that occurs within the NPF

Chapter Outline Chapter one gives a brief overview of how the project is carry on. Chapter two makes an overview of the National pension fund, its evolution, structure and its financing source as well as government expenditure and the future of NPF. Chapter three is the literature review, that is, what writers around the globe have commented on the pension system. Chapter four is the research methodology. The research is carried out using regression equation to examine the financial implication NPF on our variables. Chapter five then come the analysis based on the results obtained, that is the financial effect of NPF on national savings and the relationship between fiscal balance of Mauritius non-retirement account and the net saving that occurs within the NPF. Then finally chapter seven will include suggestions and conclusions. Chapter 2: Literature Review Introduction Pension funds is be defined as forms of institutional investor, which collect, pool and invest funds contributed by sponsors and beneficiaries to provide for the future pension entitlements of beneficiaries (E PhilipDavis 1995). Pension fund offer individuals the mean to collect saving over their working life so as to finance their consumption needs in retirement, either by means of a lump sum or by provision of an annuity, while also supplying funds to corporations, households (via securitised loans) or governments for investment or consumption. Bodie(1990a) has formalized pension funds’ function as a form of retirement income insurance. E Philip Davis (1995) suggests that pension funds perform a number of the functions of the financial system more efficiently than banks or direct holdings.

Their growth complements that of capital markets and they have acted as major catalysts of change in the financial landscape. But this is not the only reason for growth. It is also a consequence of fiscal incentives and benefits to employers, as well as growing demand arising from the ageing of the population. Pension funds are typically sponsored by employers, such as companies, public corporations, industry or trade groups; accordingly, employers as well as employees typically contribute. Funds may be internally or externally managed. The pension system is commonly divided into three pillars. The first pillar is the pay-as-you-go system based on payments by public institutions which are mainly funded by tax revenues. The second pillar constitutes fully funded pension funds with mandatory membership and the third pillar is based on fully funded pension saving schemes with voluntary membership. In a pay-as-you-go system, each generation pays for the costs of the currently retired in return for a commitment for the same treatment during its own retirement.

Workers who spend their entire work and retirement life under a PAYGO system with constant tax rates will earn a real return on their contributions equal to the growth in the workforce plus the growth in the real wage (Samuelson, 1958, and Aaron, 1966). Pension funds provide millions of people in the world security and comfort in old age. Pension funds represent the savings of millions of people, and as Paul Myners says, the ability of funds to invest these assets effectively has a profound impact on their economic well being.

Because so many people depend on pension funds to provide for their futures, ensuring the funds serve the needs of their members is a priority for Government. The social security system on the other hand as stated by law, guarantees people covered by its provisions either because they perform an occupational activity or meet the requirements established for non-contributory type social security, as well as dependent members of the family or similar, adequate protection in the contingencies and circumstances. Social Security has been defined as “the protection which society provides for its members through a series of public measures against the economic and social distress that otherwise would be caused by the stoppage or substantial reduction of earnings resulting from sickness, maternity, employment injury, invalidity, old age and death; the provision of medical care; and the provision of subsidies for families with children.” In the Social Security system, the money you pay into the system gets immediately paid back out to the people who are currently getting Social Security checks. The Social Security tax has been raising more money than is needed to pay for current benefits, in order to build up a surplus to help finance the retirement of the Baby Boom generation. The money is used in a sense to finance the government deficit, just like any other money the government borrows, Dean Baker (1998). The Social Security system is primarily a pay-as-you-go system, meaning that payments to current retirees come from current payments into the system. So Social Security will be the foundation of your retirement income. That’s because:

• You won’t outlive your Social Security retirement benefit. It will be there for you for the rest of your life.

• Your Social Security benefit won’t lose its value. From time to time, Social Security benefits are adjusted so they always keep pace with inflation.

Why National Pension Fund? Worker myopia, or lack of foresight poor planning occurs because people give too little considerations to their future economic needs when making decisions about saving for retirement. Most people seem to have a natural inclination to live for today and avoid thinking about old age and death. Hence, they give very little systematic thought to the financial issues of old age until they come face to face with them. By the time they recognize they may have a problem when they retire, it is usually too late to fix.

Government intervention through NPF has help people set aside a portion of their earnings when they are working so that they have an adequate income when they retire. Without compulsory contributions for retirement, myopic workers would not save enough to ensure an adequate retirement income and poverty would result. Another rationale for the existence of the compulsory contribution to the NPF is to protect the prudent that saves for retirement against those who do not save. Under a purely voluntary system some will contribute, others will not. As Boulding (1958) puts it in his argument, those who do not insure will have to be supported anyway-perhaps at lower levels and in humiliating and respect-destroying ways when they are in their non-productive phase of their life, but that they will escape the burden of paying premiums when they are in their productive phase. In fairness to those who insure voluntarily and in order to maintain the self-respect of those who would not otherwise insure, contributions for retirement should be made compulsory. Hence, mandatory contributions are necessary to achieve the retirement savings results that people need to have so as to have an adequate standard of living in their retirement years. Pension funds are also an important source of capital accumulation that can be used for different purposes as the build up the basic of national infrastructure, power stations and electric networks, Olli E. Kangas (2006). The Finnish case demonstrates that it was possible to unify social policy goals with the economic goals of building up modern industrial market economies. The Finnish experience has serves as a good example of how social policy has been successfully used as a developmental strategy, Mkandawire (2001). Pension funds are not only vital to the pension holders they provide for.

They are also key players in the economy as a whole. Government Budget Pension funding issues have an important, but often hidden, impact on the finances of state governments, J. Fred Giertz (2003). In most countries, contributions to retirement funds are made by employers and employees each year. Yet, there is no requirement in the short run that these contributions be sufficient to fully fund the systems.

Governments always ensure that pension payments are actually made to retirees, regardless of the level of contributions, as they are generally the funders of last resort. If pension systems are under funded, governments must deal with this problem sooner or later through additional contributions to the systems. If systems are over funded, government resources can be redirected from pensions to other government programs,J. Fred Giertz (2003). It is seen that private pensions reduce public pension spending in the longer term, once private schemes are mature. Private pensions is likely to increase budgetary pressures in the short term: if workers’ contributions go into their individual pension accounts, they cannot be used to pay for the pensions of the older generation; thus, governments have to finance pensions for the transition generation through taxation or borrowing, Nicholas Barr (2001). This will in a way affect the government budget. Unsustainable pension systems can be a problem to fiscal stability, economic growth, and poverty reduction. The need for pension reform has become pressing as demographic aging has strained pension systems around the world, leading to large expenditures, large deficits, and high contribution rates. In many cases the pension system has become a source of fiscal and macroeconomic instability, a constraint to economic growth, and an ineffective and or inequitable source of retirement income. J. Fred Giertz (2003)suggests thatnot only are pension asset changes large in comparison with state budgets, they are also growing and becoming more volatile. This trend is likely to continue and the relative size of state pension obligations is increasing.

This suggests that pension funding is becoming an increasingly important aspect of state government. He also states that ‘state pension funding’ today is no sounder than in the early 1990s. This is not necessarily a cause for alarm, but it is a source of concern. Pension funding will be an increasingly important demand on state finances in the up coming years. In the G-10 (1998) report, it states that the ageing of populations could have dramatic effects on government finances. Under current policies, government spending in the G-10 countries is projected to rise sharply over the next several decades for several reasons. Per capita expenditure for the elderly is high in the areas of public retirement benefits and, in some countries, welfare support. Public expenditure on medical and health support for the elderly is also high and has been rising. If advances in medical technology come at ever increasing cost and if the incidence of health expenditure on the elderly continues to rise, the fiscal burden could become substantial in some countries. At the same time, government revenues will be adversely affected as the baby boom generation moves from its high income generating years to retirement.

Countries whose revenues are tied more to consumption or value added taxes will tend to experience less of a deterioration in revenues than those that depend more heavily on income or payroll taxes. This would create a severe drag on national saving at a time when saving will be crucial to fostering the growth of labour productivity. Impacts of ageing population Norman Vincent Peale quotes that: “Age-based retirement arbitrarily severs productive persons from their livelihood, squanders their talents, scars their health, strains an already overburdened Social Security system, and drives many elderly people into poverty and despair. Ageism is as odious as racism and sexism.” Barry Bosworth (2003) argued that slowing economic growth and population aging in the major industrial countries have placed increased financial strain on pay-as-you-go (PAYGO) public pension systems. Retirement pensions have become a serious fiscal concern in most industrialized countries.

Pensions are largely paid for from tax revenues and it is foreseen that contributions will need to be raised substantially during the coming decades. The World Bank (1994) states that high taxes are harmful to economic growth, since they reallocate resources to the informal sector, thereby reducing output in the more efficient formal market sector of the economy. The reasons are that many people are now nearing retirement age and that the populations nowadays live longer and have fewer children than in the past. Nicholas Barr (2001) argued that the effect on funded schemes is more restrained but equally unavoidable. When a large generation of workers retires, it liquidates its financial assets to pay for its pensions. If those assets are equities, sales of financial assets by the large pensioner generation will exceed purchases of assets by the smaller younger generation, leading to falling equity prices and, hence, to lower pensions. Alternatively, if those assets are bank accounts, high spending by the large pensioner generation will generate inflationary pressures and again reduce the value of pensions. Domestic savings The main views of the life-cycle theory stipulate that individuals try to smooth consumption over their lifetime, Brumberg and Modigliani (1954). Normally savings follow a hump shaped pattern, that is, income is relatively low when individuals are either very young or retired as during their working life savings rate is higher .Ageing Population increases the proportion of households with a relatively lower savings rate in the economy which leads to a decrease in private savings. Estimates of the impact of a change in the age structure of the population on private savings, shows that population ageing will be likely to reduce savings. As regard to public savings, population ageing is likely to exercise considerable pressure on public finances, Weil (2006). In the situation of the pension schemes of the current pay-as-you-go pension schemes that exist in many states, an ageing population implies that the number of beneficiaries increases while the number of contributors to the system decreases.

The ageing population will also adversely affect public finances through higher healthcare and long-term care costs, given that older populations are more likely to make use of healthcare facilities, which, to a large extent, are provided by the public sector. Both microeconomic and macroeconomic studies find that the observed age profile of saving generally conforms with the life-cycle model, which implies that saving rates rise over a worker’s active career and then decline in retirement. Compared with macroeconomic analyses, microeconomic studies tend to show smaller variation in saving rates over the life cycle, this may be of the highly skewed distribution of wealth and saving across households, Ralph C. Bryant (2004). At a micro level, company or other obligatory pension funds can implement enforced saving by deferring wages and salaries, thereby reducing risk of a low replacement ratio. At a macro level, the increase in saving is not usually one-to-one, as increased contractual saving via pension funds is typically partly or wholly offset by declining flexible saving, E Philip Davis (1995). The remaining effect most likely results from liquidity constraints on some individuals (especially the young), who are unable to borrow in order to offset obligatory saving via pension funds early in the life cycle. It can also be anticipated that, even in a liberalized financial system, credit constraints will affect lower income individuals particularly severely, as they have no assets to guarantee and also have less secure employment.

Therefore forced pensions saving will tend to increase their overall saving particularly markedly, Bernheim and Scholz (1992). On the other hand Samwick (2000) found a lower rate of saving in countries with extensive PAYG systems. Agosin (2002) extended their analysis and shows that the rise of saving was concentrated in the business sector, and that the net change in household saving was small. Implications for equilibrium real interest rates The forecasted declines in savings make the expected consequence of ageing on the equilibrium real interest rate ambiguous. If investment falls faster than domestic savings at each level of aggregate income, the real interest rate that clears the market for loanable funds is expected to fall, since it is difficult for savers to find profitable investment opportunities, J.C. Trichet (2007). On the other hand, if domestic savings were to fall faster than investment then the real interest rate would rise to reflect the relative scarcity of financial funds. This likely decline in interest rate that equalizes savings and investment could be identified developed financial markets. Even though the actual impact of the evolving demographic structure on the equilibrium real interest rate in the capital markets is something that is going to occur with a considerable lag, some economists have suggested that expectations of such developments may have already started to exert some influence on the pricing of bonds. Among other things, these analyses suggest that ageing could have contributed to the “flattening” of the yield curve that has been observed over the recent past, J.C. Trichet (2007). However as it is based on the assumption that capital market participants are perfectly forward looking, an assumption which is questionable, it should be treated with a great deal of caution: if it is true that financial markets tend to overreact to short term phenomena, the effects of ageing on the yield curve could be limited, DellaVigna and Pollet (2005). It has to be taken into consideration that these quantitative simulations require a number of qualifications. On one hand, some real world factors may make the true decline in the equilibrium real interest rate larger than estimated in macroeconomic models.

For instance, older people may save more than predicted by the life cycle theory as they may want to leave a bequest to their children, putting further downward pressure on the equilibrium rate. The degree of risk aversion may also change with age as if the older people were systematically more risk averse than the young one, the accumulation of precautionary savings would lead to a higher than predicted savings rate and a lower than predicted real rate, Bakshi and Chen (1994).Moreover private savings rates may be significantly affected by pension reforms, Miles (2002). Pressures on Prices Hans J Blommestein (1998) states that concerns have been expressed that the growing demand for high quality private securities like equity and corporate bonds, associated with the growth of advance funded pension systems in search of investment opportunities (thereby increasing the demand for financial assets) and falling public sector borrowing requirements (thereby reducing the supply of government securities), would put strong upward pressure on the prices of financial assets. Here, the combination of the widespread privatisation of state owned enterprises and reform of pension systems brings the opportunity of killing two birds with one stone. Pension reform, which would increase the demand for equity, and privatisation, which expands the supply, at the same time permits a more balanced growth in private securities markets, at least over the medium term. In a somewhat longer term perspective, population ageing may have an impact on the risk premium, that is, the difference between the returns on stocks and the yield on bonds. As asset preferences vary across age groups, the ageing of the baby boom generation could affect both absolute and relative positions of stock and bond prices. On average, middle age is the portion of the life cycle when saving rates are highest. Moreover, middle aged workers generally are more able and willing to hold a riskier portfolio; that is, one weighted more heavily towards stocks than bonds. This is a consequence of two factors: first, while still working, a stockholder is better able to make up for any bad equity returns; second, middle aged workers have a longer time horizon and thus are willing to accept more risk in exchange for the expectation of higher returns.

Moreover, higher demand for stocks relative to bonds should increase the price of stocks relative to bonds, thus decreasing the equity premium. Thus, some have hypothesized that an ageing population would cause the equity premium to increase. But if the age of the population is increasing at least in part because life span is increasing, and thus time horizons are lengthening, then the ageing of the population does not necessarily imply that average risk aversion should be increasing and risk premium on stocks should be rising. After the baby boomers begin to retire, saving rates would tend to fall, stock and bond prices to decline, and the equity premium to rise as baby boom retirees shift their portfolios away from stocks toward bonds, Hans J Blommestein (1998). Population age structure can influence the demand for different classes of financial market assets both because of its effect on saving and because young, middle aged, and elderly savers may seek to hold their assets in different forms. Empirical studies have uncovered evidence that population age structure affects stock market prices and the real returns of different classes of financial assets, but the consistency of this evidence is not overwhelming. It is unclear whether the effects of demographic influences on asset prices and returns are large relative to the effects of other and less predictable determinants of prices and returns, Ralph C. Bryant (2004). Implications of population ageing for the conduct of monetary policy The life-cycle theory stipulates that , individuals during their working lives accumulate financial wealth in order to finance their consumption during retirement. As a consequence, populations who are near to retirement age will tend to have higher wealth to income ratios. Simultaneously, expected imbalances in publicly financed pension schemes make it possible to consider that the increasing number of retirees would depend more on their own accumulated wealth, as opposed to public pension provisions, to maintain their consumption levels.

Consequently, the fraction of the population exposed to asset price fluctuations could increase with ageing, Young (2002). Bean (2004) argues that longer life expectancy would presumably strengthen this effect. Therefore, the transmission channel of monetary policy may be affected by ageing. In particular, the so called wealth channel, which links asset prices to consumption, may gain relative importance and play a vital role than in the past, G10 (2005). Miles (2002) points out that the monetary policy multiplier would probably rise with population ageing, mainly as a result of the increased wealth channel and greater price impact of monetary policy decisions. In spite of this, he also mentions that an older population is less likely to be credit constrained, especially when the pension system is reformed towards more funded systems. This might reduce the effectiveness of the credit channel. Depending on the relative importance of these channels, monetary policy could, in principle, become more or less effective with ageing. Miles suggests that the first effect is expected to dominate. A move towards demographic structure in which the population accounts for an increasing elderly population is expected to generate a gradual but persistent change in savings habits. This may results in an impact on the demand for all classes of assets even though certain sector of the capital market are likely to be affected more substantially than others. If, for example, older people are more risk averse and prefer to hold financial assets paying fixed income returns such as government securities, then the demand for government bonds would tend to increase relative to riskier investment options, such as equity, Bakshi and Chen (1994) and De Santis and Lührmann (2006). In this situation, where a larger part of households’ wealth is invested in nominal assets, price stability would be even more important for households, G10 (2005) and Bean (2004). Stable prices ensure that the real value of both pension entitlements and savings is maintained and prevent arbitrary redistributions of income and wealth to the detriment of the most vulnerable groups in society, in particular, pensioners. It is likely that, as a significant fraction of wealth is accumulated in real estate and financial assets, households’ exposure to asset price movements will tend to increase. This might coincide with a situation in which a large fraction of the population in their old age dis-saving phase are disposing assets in order to finance consumption during retirement. In this respect, some authors have warned that, when the baby-boom generation retires and starts to dissave, excess supply in financial markets could lead to a significant decline in asset prices, the consequences of which might be felt by the entire population, Siegel (1998), Abel (2001) and (2003). This view is known as the “asset meltdown” hypothesis.

Yoo (1994) estimated that asset prices may drop by as much as 15% as a result of demographic change alone. This is why a credible commitment to maintaining price stability and, as a reflection, an orderly financial environment is and will remain so important for maintaining the standard of living of people, particularly for the poorest and the most vulnerable. Investment of Pension Fund The rapid growth of pension funds in many countries, and the stimulus they are providing to the growth of capital markets, both suggest that their activities as financial intermediaries merit considerable attention, E Philip Davis (2000). Pension funds have an impact on the stability of financial markets in several ways, most significantly through their investment behavior. Since early withdrawal of funds is usually restricted or forbidden, pension funds have long term liabilities, allowing holding of high risk and high return instruments.

Accordingly, monies are intermediated by pension funds into a variety of financial assets, which include corporate equities, government bonds, real estate, corporate debt (in the form of loans or bonds), securitised loans, foreign holdings of the instruments mentioned above and money market instruments and deposits as forms of liquidity. Hellwig (1990) suggest that financial institutions can form long term relationships with borrowers, which reduce information asymmetry and hence moral hazard. Apart from economies of scale these considerations have arisen in the literature mainly for debt finance and for banks. Whereas the importance of information asymmetries and incomplete contracts is equally recognised for equity finance, the role of financial institutions as counterparts is less well developed. Equally, institutional investors such as pension funds may not rely on the same information and control mechanisms as banks. The role of pension funds is clearly not to facilitate exchange of goods, services and assets directly.

This is because, unlike banks, money market funds, and to a lesser extent long term mutual funds, they do not offer liquid liabilities. Nevertheless, pension funds have had an important indirect role in boosting the efficiency of the financial systems, by influencing the structure of securities markets. This effect on micro structure links to their demand for liquidity, i.e. to transact in large size without moving the price against them, anonymously, and at low transactions costs. Pension funds provide risk control directly to households via the forms of retirement income insurance they provide, an advantage which largely reflects the unusual (among financial intermediaries) link of pension funds to employers. To assist in undertaking this risk control function they diversify assets as noted above and also act in securities and derivatives markets to hedge and control risk. As institutional investors, pension funds are well placed to use derivatives and other means of risk control; many innovations have been introduced or developed specifically to cater for their demand (Bodie 1990b, 1999). E Philip Davis (1995a) suggests that as pension funds focus mainly on government bonds and high grade corporate bonds, while banks tend to monopolise small business financing. And Lorenzo Bini Smaghi (2006) states that investing wisely matters for long term economic wellbeing, and that the portfolio allocation decision is of paramount importance in order to maintain living standards in the old age. Pension funds are the fastest growing of all financial institutions. They now cover half the labor force and represent one-eighth the financial assets of the entire household sector, Vincent P. Apilado (1972). The size of pension funds has also had an impact on the structure of financial markets: countries with large funded pension schemes tend to have highly developed securities markets; in countries with small pension-fund sectors, capital markets are relatively underdeveloped (the equity market in particular, Hans. J. Blommestein (1998). Living Standard M. PONDS (2003) states that the raison d’être of wage indexed defined benefit pension funds is to provide insurance against standard of living risk after retirement, based on intergenerational risk sharing.

Pension funds necessarily have to accept mismatch risk in providing this kind of insurance. Mismatch risk taken by the pension fund is risk for the fund’s stakeholders. The material living standards of tomorrow’s working and retired people will depend on the goods and services produced by those who will be working at the time. Changes in retirement income financing might alter the relative living standards of workers compared with retirees, but only later retirement could have a large effect in increasing living standards for both, Peter Hicks (2004). On the other hand, a society with a higher old age dependency ratio and fewer young people around is also one in which young workers, who realistically can rely less on publicly funded pension schemes, have to accumulate a larger amount of financial and real wealth in order to maintain living standards after retirement. This implies that young workers should invest more, and not less, in higher-risk, higher return financial assets such as stocks and foreign assets, Lorenzo Bini Smaghi (2006). The ability of pension funds to maintain living standards during the retirement of the contributors is indeed crucially dependent on the performance of financial markets.

Consequently, the expansion of a funded pension sector requires a sophisticated and well regulated financial market. By the same indication, funded pension systems are affecting financial markets and, indeed, the world economy as a whole. The magnitude of this effect can be gauged by examining the accumulation of retirement funds in relation to GDP, Hans J Blommestein (1998). In the G-10 (1998) report it is stated that as ageing raises the number of consumers relative to producers, the growth of material living standards (i.e. consumption per capita) will fall unless relative declines in the workforce are offset by increases in labour productivity and the effective supply and utilisation of labour. Decreases in labour force participation rates associated with projected demographic trends alone would depress the growth of GDP by as much as ½ to 1 percentage point per year in many of the G-10 countries between 2010 and 2030 EMPERICAL EVIDENCE Government Budget From 1990 to 2002, the US state pension liabilities increased approximately three fold while tax receipts increased less than two fold, J. Fred Giertz (2003). In Irish, it is found that public spending on pensions may increase from roughly 5% of GDP today to 11% by 2050. An ageing population will also lead to increased public spending on other age related expenditure, namely health and long-term care, Brian Cowen (2007). Estimates by the OECD for the period 2005-2050, ageing in the euro area could increase public pension costs by 3% of GDP, public healthcare costs by 3.7%, and long-term care costs by 2.2%. Saving Many studies have looked into household data (surveys) to test whether individual saving behavior is consistent with the life cycle theory, and it is generally found that the savings rate of retirees does not fall as much as the life-cycle theory predicts. However, Miles (1999) clarifies that “the mistake here is to count all of pension receipts as income” as opposed to sales of their pension “rights”. Bosworth, Burtless and Sabelhaus (1991) compute the average savings rate for households over 64 years old in the United States, correcting for the pension adjustment suggested by Miles. While the unadjusted savings rate was 14.9%, the adjusted figure was only 1.8%. Private savings rate in the European economies could fall from 15.9% of GDP in 1990 to around 4.5% by 2060, based on models that assume that households behave as predicted in the life-cycle theory, Miles (1999). In a famous 1973 paper, economist Martin Feldstein claimed that the United States’ PAYG social security system reduced personal saving by about 50 percent, and the country’s capital stock by 38 percent Börsch-Supan, Ludwig and Winter (2004) state that the ageing process may lead to a considerable fall in both private and public savings, with the savings rate in the European economies most likely to fall by 5-6 percentage points over the period 2000-2050. Rate of interest Chinn and Frankel (2003) found that US and European long-term real interest rates on government debt depended significantly upon current and expected levels of debt over the period 1973-2003. Laubach (2003) estimates that a 1% increase in the expected US debt-to-GDP ratio led to an average rise in US long-term rates of 5.3 basis points over the period 1985-2002. A different result is reported by Barro and Sala-i-Martin (1990), who analyse short-term real interest rates in Belgium, Canada, France, Germany, Japan, the Netherlands, Sweden, the United Kingdom and the United States over the period 1959-1988, finding that fiscal variables are not significant either for the expected real interest rate or for the domestic investment/saving ratio. Prices Because asset preferences vary across age-groups, the ageing of the baby-boom generation could affect both absolute and relative positions of stock and bond prices. On average, middle age is the portion of the life cycle when saving rates are highest.

Moreover, middle aged workers generally are more able and willing to hold a riskier portfolio; that is, one weighted more heavily towards stocks than bonds. The real return on United States stocks, for example, averaged 9% over the period 1947-96 with a standard deviation of 17%. This implies that there is about a 30% probability of a decline bigger than minus 8% or a rise bigger than 26% in any given year.

The average real return on long-term United States government bonds over 1953-96, however, is much lower – 3% – but also less volatile – these returns have a standard deviation of 2%,Hans J Blommestein (1998). Total pension assets in the OECD area rose from almost 29% of GDP in 1987 to almost 38% (or around $8.7 trillion) in 1996. From 1990 to 1996 the average annual growth of asset holdings by pension funds was 10.9%.The magnitude of this effect is gauged by examining the accumulation of retirement funds in relation to GDP. These figures understate the financial importance of population ageing and pre-funded systems because assurance companies and mutual funds are also involved in the production and sale of retirement-income products. But these aggregates hide a deal of variation between individual countries: the figures range from 117% of GDP in Switzerland and nearly 90% in the Netherlands to 3-6% in France, Germany and Italy. Monetary policy Bean (2004) argues that longer life expectancy would likely reinforce this effect. Poterba (2004) suggests that families whose head is over 65 years old will hold almost half of all corporate stock and 64% of all annuity contracts in the United States in 2040, up from 33% and 50% respectively in 2001. Miles (2002) points out that the monetary policy multiplier would probably rise with population ageing, mainly as a result of the increased wealth channel and greater price impact of monetary policy decisions. However an older population is less likely to be credit-constrained, especially when the pension system is reformed towards more funded systems.

This might reduce the effectiveness of the credit channel. Depending on the relative importance of these channels, monetary policy could, in principle, become more or less effective with ageing. Miles suggests that the first effect is expected to dominate. Yoo (1994) estimated that asset prices may drop by as much as 15% as a result of demographic change alone. For a different viewpoint, relatively large swings in asset prices generated by changes in the demographic structure are very difficult to predict, because there has only been one baby boom shock in the United States in the post Second World War period. Hence, he argues that it seems risky to formulate predictions with only one data point in the sample. Investment of Pension Fund At end-2004, the largest international asset manager (UBS in Switzerland) had more than USD 1 975 billion in total assets.

The world’s largest pension fund is the Government Pension Investment Fund in Japan. This fund invests two-thirds of its assets in Japanese bonds (primarily government bonds). At end-March 2005, this fund had USD 817 billion in total assets. A number of central banks also invest substantial assets in international capital markets through their foreign exchange reserves. At end-2005, both the Japanese and Chinese central banks had foreign exchange reserves of more than USD 800 billion, (Annual Report for the Government Pension Fund – Global, 2006). Total pension assets in the OECD area rose from almost 29% of GDP in 1987 to almost 38% (or around $8.7 trillion) in 1996. From 1990 to 1996 the average annual growth of asset holdings by pension funds was 10.9%. US pension funds grew by 12.5% in that period and those in Canada and the United Kingdom by rather less – 6.5 and 6.8%, respectively. Sometimes the difference is striking: Italian pension funds expanded by 37.8%, although from a very low base, and those in Germany by only 7.9%. The growth of Japanese pension assets has been 9.5%, and French pension assets had a growth rate of 13.3%, Hans J Blommestein (1998). Pension fund assets represented 88 per cent of GDP in OECD countries (compared with 76 in 2002). However, the size of pension fund assets differs considerably from country to country. It is particularly high in the Netherlands, Switzerland, the United Kingdom and the United States (where it is close to 100 per cent of GDP), and very low in countries such as Germany, France and Italy (where it does not reach 3 per cent of GDP), Lorenzo Bini Smaghi (2006). Chapter Three: Overview The Origin of National Pension Fund (NPF) The National Pension Scheme (NPS) was introduced in April 1976. It provides for the payment of contributory pensions to those employees of the private sector who have contributed to the National Pensions Fund (NPF). All employees working for employers with more than 10 employees had to join the Scheme as from July 1978. Other employees joined the Scheme as from January 1980 except household workers for whom the operative date was July 1980. The self-employed were also eligible to join the Scheme as from July 1980 on a voluntary basis. Contributory is an important pillar that covers employees of private sector firms. It was introduced in 1978 and is a defined benefit scheme operating on the French point system. Private-sector employees are required to participate in the NPF. Exceptions include very low paid workers and those sugar industry workers who, when the NPF was introduced, elected to remain within the already established Sugar Industry Pension Funds (SIPF). Civil servants and employees of local-government and statutory bodies are also exempt. EVOLUTION OF NPF IN MAURITIUS Government’s long-awaited contributory pension scheme was approved by the National Pensions Act of 1976, but began to operate only in July of 1978. Participation is mandatory for workers older than 18 years of age, with the exception of employees with very low earnings, the self-employed, and public sector employees.

Self-employed workers and the unemployed are offered incentives (2/3 contribution for the same benefit) to participate voluntarily, but few do. Workers in the public sector have no need to participate because Government meets their retirement income needs with generous pensions that are income related and non-contributory. Contributors to the NPS benefit also from life and disability insurance during their working lives; this supplements basic widows’, orphans’ and disability pensions awarded to residents less than 60 years of age under the non-contributory system. The contributory pension scheme in Mauritius is an early example of what has come to be known as a “notional defined contribution” system. It is “defined contribution” because benefits depend strictly on contributions, not on final or even the average lifetime income of participants. It is “notional” because participants do not purchase portfolios of stocks and bonds, or shares in an investment fund; instead, they receive points, the values of which do not depend in any way on how, where, or whether their contributions are invested. People 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.

• In December 1976, the pathway to the Welfare State was consolidated by the implementation of the NPS. The scheme provided for the payment of a series of Non-Contributory Benefits on a universal basis that is without means test and financed by the Government.

• In July 1978, the contributory part of the NPS was introduced to cover all employees of private sectors with 10 employees or more and the Law provided for the payment of Contributory Benefits to insured persons or their dependents if contributions had been paid to the National Pensions Fund.

• On 1August 1979, the Industrial Injury Allowance of the NPS was introduced which provided compensation to insured persons injured at work.

• On 1January 1980, the National Pension Scheme extended the coverage of all other potential employees.

• By 1July 1980, provision for the coverage of all potential insured persons had been made by the National Pensions Scheme, extending self-employed, non-employed and domestic servants (Private Household).

• 1July 1991 – GN 126/91 – The Finance and Audit Act (Employees Welfare Fund). Employers had to pay 2.5% as Contributions.

• 1July 1994 – Act No. 14/95 – National Savings fund Act replaced the EWF Regulations 1991. This Act to provide for the establishment and management of a National Savings Fund for the benefit of employees on retirement or death.

• 1July 1995- Surcharges on NSF applicable.

The Structure of the NPF Mauritius has a well functioning three tier pension system, which provides non-contributory basic pensions to elderly, widows, invalids and orphans, requires mandatory participation in an earnings related system for private and public sector employees in separate schemes, and provides for supplementary pensions on a voluntary basis. The first tier, namely the basic retirement pension scheme, which is subject to residence test, has a strong poverty reduction and redistributive effect. The second tier, which is mandatory, is made up of two contributory schemes (one for the private sector and the other for parastatal bodies) and a non-contributory one for civil servants and local authorities (except for Service Family Protection Schemes for survivors, which are contributory). Contribution to the National Pensions Fund is mandatory for private sector employees as well as employers. The Civil Service Pension Scheme as well as the scheme for local authorities is non-contributory. Parastatal pension funds, to which contributions are made by employers only, offer benefits similar to those provided by the civil service scheme and the public corporations contribute to the appropriate fund.

All employees, that is, those of the private sector, civil service and statutory bodies also benefit from the National Savings Fund (NSF), which provides a lump sum at retirement. Contributions to the NSF are made by employers only. The mandatory schemes for private sector employees will also be subject to the pressure of the ageing population. It would mean lesser people contributing to the National Pension Scheme while a larger number of members would become eligible for pension benefit payments. These pensions will, moreover, have to be paid over a longer period of time in view of longer life expectancy of Mauritians in the coming decades. Assuming current level of employment, preliminary projections indicate that by 2010 the number of civil service pensioners will represent 75% of the number of employees and by 2025 there will almost be as many pension beneficiaries as employees in the civil service. As for the Civil Service Pension Scheme, payments will be equivalent to nearly 35% of the wage bill of the Civil Service by 2025 compared with 20-25% presently. The third tier comprises private voluntary occupational schemes and pension schemes provided by insurance companies to individuals. These schemes are promoted under the present tax system.

Most of the voluntary occupational schemes are Defined Benefit (DB) schemes. There are at present over 1,000 private occupational pension funds covering about 10,000 employees. Technically, in a pure defined-benefit pension plan, the contributions rate does not remain fixed.

The latter is varied and will depend on actuarial assumptions concerning the age composition of the workforce, future salary increases, and the rate of return to be earned on the plan’s asset. But in Mauritius, the size of the contributions is defined in the NPF Act and the rates have been constant since the inception of the set up of the scheme. The contributory rates are as follows according to the NPF Act, first Schedule

INSURED PERSONS *	CONTRIBUTIONS BY EMPLOYEE (%)	CONTRIBUTIONS BY EMPLOYER (%)
Sugar Industry	3	10.5
Others (higher rate)	5	8.5
Others (standard rate)	3	6

*Insured persons excludes

• Public officer / employee of a local authority;

• An employee who is a member of a pension fund or scheme of an approved body managed by the SIC;

• Employed exclusively on weekends and public holidays;

• Self employed, non employed or prescribed persons;

• Those who have already elected to receive pensions (normally after 60 years)

A defined-benefit pension system can be overfunded or underfunded. It is underfunded when the present value of the fund’s liabilities to employees exceeds the present value of the funds assets. It is overfunded when the present value of the pension liabilities is less than the present value of plan’s assets. Contribution made to the National Pension Fund and contributory benefits: The National Pensions Fund – (Annex 1,Table 1) The number of employers contributing to the NPF in respect of their employees was estimated at around 16,500 in 2004/2005 compared to 15,400 in 1999/2000. The contributions received, exclusive of surcharge increased by 34.1 percent, rising from Rs 979.0 million in 1999/2000 to Rs 1,312.9 million in 2004/2005. At the end of June 2005, total net assets of the National Pension Fund more than doubled, increasing from Rs 18,887.3 million at the end of June 2000 to Rs 38,157.0 million. Types of Contributory Pensions – (Tables 2(a) – (b)) (i) Contributory Retirement Pension (CRP) The Contributory Retirement Pension is payable to a person on reaching the age of 60 years if that person has contributed to the National Pension Fund. (ii) Contributory Widow’s Pension (CWP) The Contributory Widow’s Pension is payable to a widow whose late husband had contributed to the National Pensions Fund. (iii) Contributory Invalid’s Pension (CIP) A person is entitled to a Contributory Invalid’s Pension if he/she has previously contributed to the National Pension Fund and suffers from a permanent incapacity of at least 60 percent. (iv) Contributory Orphan’s pension (COP) The Contributory Orphan’s pension is payable to orphans under the age of 15 years (or 18 years if the child is at school) if any of the deceased parents had contributed to the National Pension Fund. (v) Industrial Injury Benefits The different types of allowances payable in respect of a work accident affecting an employee insured under the National Pension Scheme are:

• Industrial Injury allowance as a result of a total temporary incapacity, subject to medical evidence.

• Disablement pension due to a permanent incapacity (partial or total).

• Survivor’s Pension as a result of the death.

• Constant attendance allowance resulting in a total temporary incapacity or a 100 percent disability and requiring the constant attendance of another person.

• Dependent’s pension, orphan’s pension, child allowance, clinical expenses, clothing expenses, hospital expenses, artificial aid

Financing Of NPF Funds are received from government to finance basic/non-contributory pensions, food aid allowance, inmates allowances and a guarantee of a minimum level of contributory pension. These expenditures therefore do not have a net effect on the long-term solvency of the NPF since they are completely funded by the government and consequently their trend and implications for the government and the NPF have been ignored in this project. The NPF revenues emerge from (i) Contributions, Surcharges from employees and employers (ii) Interests/Dividends from Investments (iii) Redemption/maturity of loan stocks/debentures (iv) Maturity of deposits in Banks (v) Repayment of Loans (vi) Commissions as agents for collecting funds for IVTB. EWF, NSF (vii) Rent income for NPF buildings (viii) Any miscellaneous revenue. Apart from revenue received from government to pay basic pensions and allowances, NPF obtain its revenue primarily from contributions from employers, employees and also from investment income. These two sources of income form more than 85% of total income receivable.

Government contribution could be seen as the running cost of the NPF fund which is around 10%-15% and if it would had be run by itself, it would have been above 20% as in European countries. The NPF has accumulated substantial financial resources equivalent to 17 percent of GDP. These are invested in government bonds of about 58 percent and with relatively small shares in corporate securities and foreign assets. The performance of the NPF as a financial institution has been satisfactory. The performance of the NPF has been disfigured by the discovery of a fraud in Feb 2003. It had been ongoing for five years and involved a time deposit of MUR 500 million with the Mauritius Commercial Bank (MCB), the largest and oldest commercial bank in the country. However, its non-detection for many years indicates a major deficiency in internal audit and control systems at both of these nationally important institutions. Government Expenditure on Social Security and Welfare Government expenditure on Social Security and Welfare consists of expenditure and transfers made by government for the social welfare of the community, in line with the recommendations of the Government Finance Statistics Manual 1986 of the International Monetary Fund. It covers all expenses made by the Ministry of Social Security and the Ministry of Women’s Rights together with components of social welfare of all government organizations as well as Local Government. Expenditure on Social Security and Welfare increased by 62.6 percent from Rs 5,953.2 million in 1999/2000 to Rs 9,681.9 million in 2004/2005. The latter amount represents 22.1 percent of total government expenditure in 2004/2005 against 21.7 percent in 1999/2000. Figure 1 shows the share of each component of government expenditure on Social Security and Welfare. It is observed that the main components are the Basic Retirement Pension with a share of 36.0 percent followed by Public Service Pensions (29.5 percent). Future of NPF A key element of the projections of NPF is the increasing maturity of the contributory element. Under the Scheme provisions, the contributory pensions of members who were over age 40 in 1978 (and so retired before 1998) were doubled, thus partly offsetting the reduced period over which contributions were paid. The contributory pensions of members who were aged between 20 and 40 in 1978 (and so due to retire over the 20 years from 1998) are increased as if they have been contributing for 40 years.

The cost of contributory pensions is projected to increase for the period after 2020 because of the expected continuing increase in the number of Scheme members over pension age. The central projection indicates that the Fund balance will remain healthy over the 40 year period, although it is projected to fall from over 60 times the annual expenditure on contributory and industrial injury pensions in the year 2000 to about 20 times the annual expenditure by 2040. For the next 15 years, contribution income is projected to exceed expenditure on contributory and industrial injury pensions with the excess income available for new investment. Thereafter, expenditure on benefits is projected to exceed contribution income and it will be necessary to meet the shortfall from investment income and or the sale of assets.

The precise timing of this change will depend on whether the assumptions are borne out and on the actual investment returns achieved on the Fund’s assets. The pensioner support ratio is a particularly important indicator in assessing the likely future trend in the balance of the Fund. Whether pensions are financed on a pay-as-you-go basis (basic pensions) or on a funded basis (contributory pensions), the effective cost of supporting pensions at any point in time must fall on the economically active population at that time. The steep decline in the pensioner support ratio therefore implies that a greater part of the wealth produced by the working population will be transferred to the pensioner population. In order to assess the financial viability of the Fund over the long-term future, it is also necessary to consider the rate of investment return earned on the assets of the Fund. The Fund’s liabilities are, broadly speaking, related to price and/or earnings inflation, and so it is appropriate to consider the future rate of return in excess of price (or earnings) increases. KEY POINTS FROM THE ACTUARIAL REVIEW AS AT 30™ JUNE 2000: The Actuarial review is usually carried out every 5 years, in accordance with Section 38A of National Pensions Act. It is prepared by the Government’s Actuary’s Department, London. Expenditure on contributory pensions increased partly because of the increasing number of contributory pensions in payment and partly because of an increase in the average amount of contributory pension.

Currently, most pensioners receive relatively modest contributory pensions, based on incomplete contribution records, because entitlement has only been built up from contributions paid since 1978. Since the inception of the NPF, contributory pensions in payment have been increased, at the discretion of the Minister, broadly in line with prices (the CPI). For the central projections, it has been assumed that this policy will continue, and that contributory pensions will increase in line with prices after they come into payment. Currently, contribution income is significantly greater than the combined expenditure on contributory pensions, industrial injury pensions and administration costs. Over the next 15 years, it is estimated that substantial amounts of excess income will be available for investment although the amounts available will gradually decline. It is estimated that contribution income and benefit expenditure will be approximately in balance in 2015. Thereafter, as contribution income stabilises and expenditure on pensions continues to increase, benefit expenditure is projected to exceed contribution income by an increasing margin. This shortfall must be met from investment income and/or the proceeds from the sale of assets. Chapter 4: Methodology Financial effect of NPF on national savings The impact of the pension system on individual saving has been a concern since Feldstein’s (1974) pioneering paper, which centers around the funding status of social security and in particular on the degree to which an unfunded pension system reduces private saving. For Palley (1998), the way to finance social security that has the least effect on private savings is to cut payroll taxes and finance the public pension system from general tax revenues. In an accounting sense, savings equals disposable income less consumption, and we can write the savings function (S) as: S= ?0 + ?1X, Where ?0 and ?1are parameters and X is a vector of exogenous variables, including income. Savings can be break down into retirement savings (SR) and other savings (SO): S = SR + SO, and there may be substitution between SR and SO. Assuming that SR is exogenously determined, in other words, that retirement saving is determined solely by policy factors as is the case with state PAYG and private DB schemes, we can write SO= ?0 + ?1X – ?2SR Where ?2 indicates the extent of substitution between SO and SR. If ?2 = 1, then retirement savings fully offset other savings and total savings is unaffected by increases in pension wealth. Assuming ?2 < 1, i.e., that SO and SR are not perfect substitutes, the savings function S can be written as: S = ?0 + ?1X+ (l- ?2)SR The set of other explanatory variables (X) takes into account the different macroeconomic environments, habits, policy incentives and so on that could affect savings patterns.

People save when they feel insecure or when there is macroeconomic instability or an expected downturn. The systematic determinants of private savings, actual and expected growth rates and the terms of trade are beyond the direct control of policies. The impact of interest rates on household savings is uncertain because income and substitution effects work in opposite directions. In the standard neoclassical view, as higher interest rates increase the opportunity cost of consumption, households increase their savings (the substitution effect); on the other hand, with higher interest rates, positive savers increase consumption because of their rising wealth (the income effect). According to the life-cycle model (LCH), the higher the ratio of the elderly population to the working-age population, the lower will be the aggregate saving rate because the old are retired and do not save while the young work and do save. One of the main views of the life-cycle theory is that individuals try to smooth consumption over their lifetime, Brumberg and Modigliani (1954). Normally savings follow a hump shaped pattern, that is, income is relatively low when individuals are either very young or retired as during their working life savings rate is higher. The budget deficit negatively affects the overall saving rate when the savings of the private sector (mostly households) do not fully offset the deficit.People could react to government deficits (negative savings) by increasing their saving in anticipation of higher taxes or by reducing savings in expectation of the faster inflation that will eventually occur. Referring back to Equation

[1] we had the level of national savings as a function of the level of savings for retirement and a vector of exogenous variables, X. S = ?0 + ?1X+ (l- ?2)SR Those factors that will affect savings behavior, that is, X should include the effect of interest rates, credit market development, the government budget, the availability of retirement benefits from the public pension system, and the age composition of the population. Definitions of Variables and Time-Series Properties of Data Specifically, we analyzed data on savings and these other factors in Mauritius over the period 1990 to 2005. Data is compiled from the Central Statistics Office (CSO), Bank Of Mauritius (BOM) and National Pension Fund (NPF) to calculate annual time series for the following variables:

• SAVR is the aggregate saving rate based on the current account definition of savings as the residual difference between total income and consumption. Here we are going to use the gross national saving (GNS).

• PENSION is defined as the ratio of net investment of national pension funds over the working age population.

• RINT is the real interest rate : Real interest rate = Nominal ROI / CPI x Base year index

• PAYG is defined as the real expenditure on NPF divided by the population aged 60 (men)/60 (women) and over and is a measure of the level of pension benefits per elderly person.

• The dependency ratio, DEP, is calculated as the ratio of persons over 59 to persons aged 15-59. The dependency ratio tells us how many young people (under 15) and older people (over 59) depend on people of working age (15 to 59). The dependency ratio is worked out with this formula Dependency Ratio = Number of Dependents X 100 % Population (Ages 15 – 59) This indicator gives insight into the amount of people of non-working age compared to the number of those of working age. A high ratio means those of working age – and the overall economy – face a greater burden in supporting the aging population.

• FIS is the government’s fiscal balance to control for the effect of government deficits on private savings, i.e., excess of government revenues over expenditure.

• GROWTH is per capita GDP growth.

• CREDIT is domestic credit to the private sector as a percent of GDP to indicate the liquidity constraint on households. As we are looking at the relationship among these variables over time, we need to examine their properties as time-series and choose an appropriate estimation technique. We are going to use Augmented Dickey-Fuller (ADF) tests to check for stationary in variables. An augmented Dickey-Fuller test is a test for a unit root in a time series sample. An augmented Dickey-Fuller test is a version of the Dickey-Fuller test for a larger and more complicated set of time series models. The augmented Dickey-Fuller (ADF) statistic, used in the test, is a negative number. The more negative it is, the stronger the rejection of the hypothesis that there is a unit root at some level of confidence. In autoregressive time series models the presence of unit root causes a violation of the assumptions of classical linear regressions. A unit root means that the observed time series is not stationary.

When non-stationary time series are used in a regression model one may obtain apparently significant relationships from unrelated variables. This phenomenon is called spurious regression. As the ADF test is an asymptotic test, it may not be suitable for our small sample. However, because we found the variables in our sample to be of different orders of integration, the ARDL (Autoregressive-Distributed Lag) model, as explained below, is most suitable to carry out further investigation. This analysis led us to choose the following general specification of the savings relation for our sample: SAVRt = (PENSION t, RINT t, FISt, GROWTHt, CREDITt, PAYGt, DEPt) In the equation above, I will remove the coefficient FIS as when running the equation in Eviews, I’m getting result of insufficient data and while removing it, my equation respond to the analyse undertaken. Then my new equation will become: SAVRt = (PENSION t, RINT t, GROWTHt, CREDITt, PAYGt, DEPt) Cointegration Analysis We use the cointegration method which establishes the existence of a unique long-run equilibrium relationship between the variables.

Although their is strong theoretical reasons for believing that the flow of pension assets and the other variables influence savings, which in turn impacts investment, we cannot a priori exclude the existence of some reverse causality. Moreover, there could be important cross effects between pension assets and savings that could be captured by estimating the system as a vector auto-regression (VAR) model.

Given the limited number of observations(16) and the many variables (7) involved, we chose the Autoregressive-Distributed Lag (ARDL) approach of Pesaran and Shin (1999) rather than the Johansen-type VAR model (1988, 1996). The ARDL approach allows for inclusion of deterministic/exogenous regressors in the cointegrating relation. Estimation is carried out using Eviews 5.1. Vector autoregression (VAR) is an econometric model used to capture the evolution and the interdependencies between multiple time series, generalizing the univariate AR models. All the variables in a VAR are treated symmetrically by including for each variable an equation explaining its evolution based on its own lags and the lags of all the other variables in the model. Based on this feature, Christopher Sims advocates the use of VAR models as a theory-free method to estimate economic relationships, thus being an alternative to the “incredible identification restrictions” in structural models Lags are present in econometrics for several reasons.

Psychological inertia (habit), permanent vs transitory income, technical and technological reasons causing delay in implementing the changes in capital labor compositions, institutional reasons, labor contracts, etc. The autoregressive distributed lag model (ARDL) is the major workhorse in dynamic single equation regressions. One particularly attractive reparameterization is the error-correction model (EC). Its popularity in applied time series econometrics has even increased, since it turned out for non-stationary variables that cointegration is equivalent to an error-correction mechanism. By differencing and forming a linear combination of the non-stationary data, all variables are transformed equivalently into an EC model with stationary series only. The first step is to apply unit root test on all our variables then made an estimate of the ADRL model in order to derive the long run coefficient estimates. The order of the ADRL model is determined by means of model selection procedures such as Akaike, Schwarz, and the adjusted R-square statistic. The relationship between fiscal balance of Mauritius non-retirement account and the net saving that occurs within the NPF Here now we use another simple equation estimating the relationship between the fiscal balance of a state’s non retirement accounts, B, and the net saving that occurs within the National pension program, PEN. Annual changes in state personal income are included to adjust for cyclical fluctuations in the state economy: Bt = ? + ?PENt + ???t, Where B t = Balance in the non-retirement budget accounts in Mauritius in year t; PEN = Net accumulation within the National Pension account; ?Y = Percentage change in real personal income. The net accumulation of the pension fund is measured as total contributions, plus investment income, less total expenditures. The measure of investment income includes realized capital gains or losses on asset sales, interest, and dividends. The annual balance in a state’s non retirement account is computed by excluding investment income of the pension account from revenues, and redefining state expenditures to exclude pension benefits and include state contributions to the pension fund.

Since state contributions to the pension funds represent payments for accruing new liabilities, they reflect a component of current employee compensation (like wages or sickness pay) and should be treated as a current expense of government. The coefficient on pension accumulation in the equation measures the extent to which changes in the pension account are offset by associated changes in net state spending in other budget accounts. Chapter 5: Analysis For the regression equation to test the financial effect of National pension system on national savings are analyse below: Since the ADF suggest that all variables are non-stationary in levels and that it is not possible to reject the null hypothesis of order one at a high level of significance. Having performed unit root on all our variables at I(0) to check whether the variables are stationary, its only the SAVR and PAYG to be stationary at 5% significance level. Variable PENSION, DEP, RINT and CREDIT were found to be stationary at I(1) where as GROWTH is at I(2) that is , stationary in variables. Lags are taken to be automated on a level of three.

The results of stationary are found in Table 1. This test is carried out to assure that all our variables that are being used follow a trend so as to carry out the analysis with the ARDL equation.

Variables	Coefficient	Std Error	t-statistic
C	-46.838	21.529	-2.176
PENSION	0.244	0.079	3.080
RINT	0.378	0.177	-2.134
GROWTH	1.195	0.599	1.99
CREDIT	1.764	0.478	3.690
PAYG	0.545	0.498	-1.094
DEP	7.712	3.295	2.341

The long-run coefficients corresponding to this ARDL model are shown in the Equation below: SAVR = -46.838 + 0.244 PENSION – 0.378 RINT + 1.195 GROWTH + 1.764 CREDIT – 0.545 PAYG + 7.712 DEP The signs and magnitudes of the coefficients are all attuned with the model of pension savings. The long-run coefficient estimates of log(PENSION), log(RINT), log(GROWTH), log(CREDIT), log(DEP) are significance at 5% level of significance. The coefficient on PENSION is an estimate of the parameter (l-?2) that was used in equation [1]. The estimated long-run coefficient on PENSION stipulates that with a one unit (RS million) increase in pension savings per worker, aggregate savings in Mauritius will increases by 0.244 percent imply that other savings decline by 0.756 percent. In other words it is inferred that there is no perfect substitute between retirement savings and other forms of savings, that is, (l-?2) is not equal to zero in the sense that households compensate forced saving such as contribution to national pension schemes either by borrowing or by reducing their flexible saving. Due to illiquidity of pension savings that is such savings made cannot be withdrawn until retirement, there is possibility of an offsetting effect on other savings with the increase in pension savings. The additional savings that result from the National pension schemes should find its way to the capital market, increasing the equity market capitalization.

The variable PENSION is a substitute for that part of the market capitalization. Our finding suggests that, in the long run, higher pension savings relative to the working-age population has a significant positive effect on national savings in Mauritius. Simultaneously, it shows that any increase in anticipated pension benefits could also lead to a corresponding reduction in current non-pension savings. Based on the above results, national savings is negatively affected by an increase in the real interest rate (RINT), implying that income effects offset substitution effects. This means that with a 1 percent increased in interest rate, saving will be reduced by 0.378 percent while consumption will rise by 0.622 percent. But empirical evidence, as said by Edwards (1996) from cross-country studies shows that interest rates have little or no effect either on saving.

Dependency ratio (DEP) on the contrary, is found to have a positive effect on national saving that may be that there was not so much worker that has contributed to that fund that now are retired. Also as stated by the Actuarial Review as at 30 June 2000, state that the central projection indicates that the Fund balance will remain healthy over the 40 year period. The estimated impact of CREDIT on savings supports that there is liquidity constraints in Mauritius. In other words, credit constraints as a result of financial sector development have contributed to an increased in national savings in Mauritius.

The variable GROWTH also has a positive effect on national savings. It is found that per capita GDP increases by approximately 75%, which may be inferred that people have increased their saving rates. High economic growth is the key to boosting national savings by reducing unemployment related public spending. Finally, the estimated coefficient on PAYG suggests that an increase in government spending on pension benefits crowds out national saving, but the estimate is statistically insignificant. Diagnostic Tests:

Adjusted R-squared	0.971
S.E of regression	0.075
Durbin Watson statistic	1.941

The equation above passes these diagnostic tests stated above. It can be inferred that the Adjusted R-squared is quite high, reflecting the effectiveness of our model. Also the Durbin Watson test which shows that if the value obtained is near to two, then their neither positive nor negative autocorrelation in the model and it is approximately 1.9, which is good. The estimation that was used for the relationship between aggregate saving rates and changes in national pension wealth in Mauritius from 1990 to 2005, was based on a model which was applied in the UK, controlling for other factors that influence savings such as real interest rates, the dependency ratio and the other variables used in our model. It is found that higher pension assets are associated with a small net increase in national savings. We need to treat these results with caution because of the relatively short sample size. We also found that demographic pressures are likely to increased savings in the long run. The result is that other savings are largely offsets by increased private pension saving lends support to the three-pillar approach to pension systems recommended by the World Bank (1994). Both public and private pillars are essential for a well implemented pension system. Public pillars, that is, whether funded or unfunded, offer basic benefits that are independent of the performance of financial markets.

Also, from the point of view of maximizing social welfare, a PAYG system is needed to continue to supplement the fully funded system, as we found the impact of PAYG to be of no consequence in crowding out aggregate savings. A perfect pension system can include a basic minimum government benefit funded by some form of compulsory saving, increasing retirement age, and incentives for individuals to top up the package with savings of their own. The impact of National Pension Fund on savings may not be uniform across countries. It is found that the coefficient sign of some variables that where used are not the same as in UK, like CREDIT which is negative for UK and the DEP also is negative in UK. Also the coefficient intercept for Mauritius is negative which mean that it starts at a low level and goes up where for UK is positive. The flow of savings into private retirement accounts is likely to continue to increase because of concern over the aging of the population. Given the importance of domestic sources of savings to develop domestic capital markets, pension funds could act as the main source of long-term savings. ECM Here this analysis is about estimating the relationship between fiscal balance of Mauritius non-retirement account, B, and the net saving that occurs within the National Pension Fund, PEN. The results for this analysis are shown in the table below:

Variables	Coefficient	Std Error	t-statistic
C	-11.605	14.096	-0.823
Log(PEN)	1.076	1.601	0.927
?Y	3.228	1.660	1.945

In the table above, it can be inferred that all the coefficients are significant at 5% significance level. Also the test has an Adjusted R-square of 0.613 which reflect a fair coefficient of determination, that is, the entire coefficient in this model is very effective. The coefficients corresponding to that model is shown below: B = -11.605 + 1.076PEN + 3.228?Y The coefficient on pension accumulation in the above equation measures the extent to which changes in pension account offset by associated changes in net state spending in other budget account. The current percentage in real state aggregate income is included to control for business cycle effects on state level non-spending. The current changes in state level income are highly significance, indicating that there are strong cyclical influences on both state revenues and overall budget balances.

These coefficients imply a highly and statistically significant offset of about 107 percent of the accumulation. When the fund accumulate an additional Rs1000 in extra reserves, the deficit in state non pension budget account eventually decreases by Rs1070, the estimate thus indicate that the additional accumulation of funds in National pension account has in effect an impact on the net balance of the rest of state budget. From the above graph it can be inferred that there is a direct impact on the government budget in according the contribution to the pension. The contribution to the pension entails a lot of economic effects. Government must in each of its budget, be able to adjust the amount of pension to be disbursed in the following period. As shown in the graph above, this is plotted in order to follow the importance of pension on the government budget over the period 1990 to 2005. It can be seen that there have been a considerable increase in the amount disburse to the pension by the government during this period. This is due to the increasing ageing population.

Also, it can be inferred that with the increasing rate of the pension, there have been an increase in the government deficit. Chapter 6: Suggestions & Conclusion The contributory part of the National Pension Scheme, since its inception, had been badly designed financially. There are many errors which have been made in the past and which are irreparable.

Reform of the NPF is long over due. This is true all the more due to the demographic time bomb. This chapter deals with some suggestions to improve the NPF current financial standing. It must be pointed out that we need complex solutions to complex problems and the pension reform debate is one that is very open. Suggestions (i) Increase contribution income Contribution income could be increased if either the contribution rates were increased or if changes were made to the maximum and minimum limits for paying contributions so that either more people made contributions or the average amount was raised. This would increase contribution income and it causes virtually no immediate effect on pension cost.

The new money available for investment would rise in the short term. However, if pension point continues to be earned based on the present relationship between the cost of a pension point and its value, then the rise in income in the short term would lead to higher pension cost in the longer term. Thus, although improving the position in the short term, the fundamental problem would remain in the long term. If contribution rates were raised but the target pension kept unchanged (effectively increasing the cost of pension points), then contribution income would rise without any increase in long-term pension costs. In order for such a change on its own to stabilize the Fund, the contributions rates would need to be approximately doubled on the basis of the main assumptions of the Report. (ii) Reduce the target pension If contribution rates were held constant and the cost of a pension point increased then the long-term cost would reduce and the progress of the fund would move towards a more stable position. In order for such a change on its own to correct the position, the cost of a pension point would approximately have to double on the basis of the main assumption of the report. (iii) Raised pension age The expenditure on pensions could be reduced by raising the minimum pension age, which is now going to be sixty-five, thus lowering the number of years for which pensions are payable. Increasing the pension age to 65 is, however, insufficient to stabilize the Fund. (iv) Reduce increases in pensions payments Contributory pensions are assumed to increase with the same level of inflation. A lower rate of increase of pensions in payment would reduce expenditure and thus move the fund towards stability. (v) The National pension scheme is based on both a defined-benefit and a defined contribution system. This is because both the pension benefits (by a formula) and the contribution rates (in the NPF Act) are defined. NPF can no longer pretend to be consisting of both systems. It must opt to be one of them. The NPF can opt to be a defined-benefit system. In this case the contribution rate will be varied at regular intervals based on the returns on investments and on actuarial assumptions. The contributor will receive, at regular intervals, a summary of his/her pension entitlements.

However, the NPF Act must be amended to ensure that the government will pay the NPF commitments if it goes bankrupt. This will ensure that the government becomes more alert to the operations of the NPF. Alternatively, NPF can opt to be a defined-contribution system. In this case the contribution rate will remained fixed all through the working life of the contributor. Each member will be informed, at regular intervals of the accumulated amount of his/her contributions and pension entitlement. However, the latter will fluctuate depending on the performance of the investments. CONCLUSIONS The NPF is a very important cell in our society. It is called to become an even more important as the full effects of the changing demographics is felt. It is recommended to consider all dimensions of pension reforms including its social, political, cultural, economic, and financial aspects. If all these aspects are considered, the new consensus that will emerge will be strengthened and will help secure the future generation a decent and peaceful retirement income.

Cite this page