INTRODUCTION

In 2009, Ghana introduced a three-tier pension policy comprising a mixture of pay-as-you-go (PAYG) social insurance and private pension plans in the form of individual accounts.1, 2 Although the three-tier institutional arrangement of pension schemes has been a common feature of reforms in Latin America and Central and Eastern Europe,3, 4, 5, 6, 7, 8 in sub-Saharan Africa, Ghana remains the only country so far with this type of pension plan. The predominant transformation undertaken in the Ghanaian pension reform was the addition of two private pension pillars to the existing partially funded PAYG social insurance model. The addition of two private pension pillars is puzzling not only because Ghana – like most African countries – lacks the financial market to hold or invest pension funds, but also because entrusting old-age income security into the hands of private fund managers in a country with fragile private sector has implications for the protection of pensioners.

Against this background, the article examines the three-tier pension scheme introduced in Ghana focusing mainly on the extent to which risks associated with decentralized and competitive pension programs are addressed. It shows that although private pensions have been framed as instruments for giving workers and pensioners a choice and freedom over retirement decisions, as well as a mechanism for capital market development through individual savings, they also expose pensioners to threats of income insecurity at old age. As the major changes in the overall pension system in Ghana centered on the incorporation of private pillars, the first section of this article provides analysis of the nature and operations of pension programs heavily aligned to market solutions. The second section discusses the structure of Ghana's three-tier pension scheme. The third sections analyzes how specific risks associated with reliance on private sector instruments for retirement income security are regulated under each of the three tiers. Section four provides an analysis of the overall weaknesses of the new Ghanaian three-tier pension system in relations to income security of the aged. The article concludes by providing alternative policy options for further reforming the Ghanaian pension system in a way that would ensure income security for the aged without compromising the strategy of using retirement contributions in funded arrangements for capital market development.

UNDERSTANDING THE OPERATIONS OF PRIVATE PENSION SCHEMES

Private pensions are typically funded schemes that are neither redistributive across nor within generations. They are funded because benefits are paid from financial assets buildup over a period of time through membership contributions and investment returns.9, 10, 11 Under this arrangement, ‘an individual's pension is an annuity whose size – at any given life expectancy and rate of interest – is determined only by the size of his or her lifetime pension accumulation’ (p. iv).12 Financing of private pensions is usually through employer–employees contributions that are invested, and when the employee reaches retirement age, becomes permanently incapacitated or dies before retirement the total contributions and any accrued interest are paid as a lump sum to the employee or his/her dependants.9, 10, 11, 13

The distinguishing element in private pension schemes is that they are often encouraged through tax incentives, but they are not tax financed, and benefits are directly linked to contributions.14 In most cases, benefits are paid in the form of one-off lump sum payment,11 although the design of some of these schemes permits the purchase of annuities. The use of one-time lump sum benefit payment is premised on the idea that defined contribution schemes (a) are simple to operate and it is ‘easy to explain the scheme to all workers, even those who are illiterate or poorly educated’ (p. 573)15; (b) contributors can ‘identify their social security savings’ and ‘claim proprietary rights over’ in the future (p. 197)10; and (c) allow contributors to acquire ‘income generating assets so as to avoid subsequent recourse to any form of public assistance' (p. 199).10 They are therefore not based on intergenerational transfers, and there is little consideration for collectivism, risk pooling and solidarity. The choice between private and public pension plans therefore has implications for the income security of pensioners and the sustainability of government budgets.

Supporters of private pensions often highlight the merits of private pension plans in juxtaposition to what are often considered the weakness of public pensions in the form of PAYGO or social insurance plans. For instance, it is argued that public pension expenditures and the generosity of PAYGO programs are incentives for early labor market exit and thus a disincentive to work. Consequently, private pension advocates portray the traditional PAYGO social insurance systems as obsolete, unsustainable and detrimental to the economic progress by emphasizing economic efficiency and the growth advantages of private or individual retirement savings.16, 17 In the context of developing countries where financial systems are underdeveloped, it was argued that private pensions would not only contribute to economic development by enhancing access to capital for productive activities, but also help improve the overall financial market by making them more liquid, efficient, and transparent through the operations of institutional investors.16

In addition, advocates of private pension plans argue that the long-term returns on privatized pensions would be ‘higher than the annual percentage growth in total real wages, on which public pay-as-you-go systems depend’ (p. 25).18 As a result, private pension plans were projected as an effective strategy for restoring financial health of national economies in a way that simultaneously improves the ability of individuals to exercise control over their retirement income decisions. The thrust of this argument is that contributors will earn a significantly higher return if their contributions are invested in stock markets rather than transferred to current retirees as is the case in PAYGO schemes.16 Others argue that because mandatory private pension plans use actuarially fair accounting principles, they avoid the controversies surrounding redistribution, and thus serve as a means to compel younger people to accumulate or shift consumption to later stages of their lives. This argument is often based on the notion that because individual pension benefits are directly linked to contributions in private pension schemes, participants in such plans have less incentive not to save for old age.19 Like in other countries, these arguments were instrumental in the decision to graft two private pillars into the Ghanaian pension system.

STRUCTURE OF GHANA’S THREE-TIER PENSION SCHEME

Until the recent reforms, Ghana's pension was characterized by uncoordinated fragmentation since independence. For instance, although civil servants were mostly covered by a scheme known as Cap 30 inherited from the colonial era, policymakers in the early post-independence era introduce provident funds that was later converted into a social insurance scheme in 1992. Thus, until the recent reforms, the country operated parallel public pension schemes – a PAYG social insurance known as the Social Security and National Insurance Trust (SSNIT) and the Cap 30 scheme – covering workers in the formal sector but offering different eligibility conditions. The Cap 30 plan had flexible retirement conditions, attractive benefit computation formula, lower voluntary retirement age and other benefit structures lacking in the SSNIT social insurance program. In addition, although both were defined benefit plans with the same contribution rates, benefits for those covered by the CAP 30 plan were paid from general revenues, whereas those covered by the SSNIT social insurance plan received their benefits from contributions made by current workers and returns on investment by the scheme. One of the marked differenced between these two public schemes was that whereas the social insurance scheme was partially funded, the CAP 30 plan was unfunded and often factored into the annual national budgets. Thus, retirement conditions under CAP 30 were unquestionably better than those under the SSNIT social insurance scheme, ‘which is why those who can keep themselves in the plan do so, and others outside it are fighting to get on it’ (p. 14).20

It was against the challenges posed by institutional fragmentation and inequality that the three-tier pension plan was introduced in 2009 to provide an organized and well-coordinated framework for pension retirement income security arrangement in the country. Beyond minor parametric reforms, the SSNIT social insurance scheme was kept, the Cap 30 was closed to new labor market participants but replaced by two additional private pension pillars – one mandatory, the other voluntary – layered on top of the social insurance scheme. Thus, the three-tier plan consists of the SSNIT social insurance scheme as the first tier and known as the national basic social security. It is an earning-related scheme, mandatory for all formal sector workers but voluntary for informal sector workers. This tier is based on defined benefit to ensure payment of regular monthly benefits to retirees. Financing of the first tier is through employer–employees contributions of 11 per cent plus returns on investment.21 The most pronounced parametric reform carried out in the SSNIT social insurance was an increase in the overall total employer–employee contributory rates from its original of 17.5 to 18.5 per cent, and a reduction in the total contribution to the SSNIT social insurance scheme from 17.5 to 11 per cent to make 5 per cent available for investment in the second, whereas the remaining 2.5 per cent directed toward financing the country's national health insurance program.1 The Ghanaian government participates in this scheme as an employer, as well as the ultimate guarantor of the tier in the event of insolvency. The guarantor position of government is deduced from its fiduciary responsibility to its citizens.

The second tier is mandatory individual retirement savings account based on the principles of defined contribution. Under the second tier, benefits were directly linked to contributions and interest accrued on investment. In terms of financing, each employee has the right to decide where and how to invest 5 per cent of their total social insurance contributions within the framework of the second layer. This tier is designed to formalize and incorporate existing occupational pension plans into the broader national pension system.1 The third layer is a voluntary retirement saving scheme to be in conformity with existing legislations such as long-term retirement savings act. As participation in this scheme is voluntary, it is embedded with attractive tax incentives to serve as a mechanism to encourage individuals who intend to undertake additional savings toward retirement. Similar to the second tier, the third tier is also a defined contribution scheme under which contributions are directly linked to benefits and paid out in the form of lump sum. Although Ghana's three-tier pension program is composed of both defined benefit and defined contribution elements, all the three tiers rely on market solutions through investment of retirement contributions. Whereas the contributions made towards the first tier is managed solely by a statutory body – SSNIT –, contributions meant for the second and third tiers are invested by registered institutional pension service providers including custodians, trustees and fund managers. These institutional service providers must first be certified and registered by a statutory agency known as National Pension Regulatory Authority (NPRA). The NPRA has oversight responsibility for the operations of all pension or retirement funds in the country.2

One of the primary objectives for grafting two private pillars into the overall pension system was to accumulate funds through retirement contributions for domestic capital market development. As a result, under the three-tier model, institutional providers of pension-related services are required to invest pension contributors in the following categories of portfolios: (a) bonds, bills and other securities issued or guaranteed by the Bank of Ghana or the Government of Ghana; (b) bonds, debenture, redeemable preference shares and other debt; (c) instruments issued by corporate entities and listed on a stock exchange registered under the Securities Industry Act, 1993, (P.N.D.C.L. 333); (d) ordinary shares of limited companies listed on a Stock Exchange and registered under the Securities Investment Act 1993 (P.N.D.C.L. 333) with good records, declared and paid dividends in the preceding 5 years; (e) bank deposits and bank securities; investment certificates of closed-end investment fund or hybrid investment funds listed on a Stock Exchange registered under the Investments and Securities Industry Act, 1993 (P.N.D.C.L. 333) with a good record of earning; (f) units sold by open-end investment funds or specialist open-end investment funds listed on the stock exchange recognized by the Board; (g) bonds and other debt securities issued by listed companies; (h) real estate investment; (i) and other forms of investment that the Board may determine. Fund managers, custodians and Trustees cannot invest retirement contributions outside the shores of Ghana unless any such in investment goes directly to any of the categories above.2

RISK REGULATION IN GHANA’S THREE-TIER PENSION SCHEME

By relying heavily on private sector instruments, Ghanaian policymakers subordinated the primary social protection role of pension schemes to the capital market development. This raises a number of issues with implications for income security of pensioners. First, the idea of using private pensions to ensure individual freedom over retirement decisions raises the question of myopia. In other words, the exercise of personal freedom in matters of retirement include the possibility of individuals undervaluing future needs in comparison to present desires.22 Second, in an environment of competitive private pension arrangement, individual consumers face the challenge of information overload, which can lead to choosing inappropriate (choice risk) investment products.23, 24 Third, administrative charges associated with competitive private pensions funds are relatively higher than in centrally managed PAYGO programs. This is largely due to the fact that fund managers often charge their operation costs and profits on the retirement contributions, thereby reducing accumulated retirement savings of participants. As others have argued, in countries where private pension plans have been implemented, the competition for clients has resulted in substantial marketing costs, which are often grafted into the overall administrative charges of individual retirement accounts. In the end, such huge administrative costs offset the social gains in the form of higher retirement incomes expected from operating efficiencies among private pension service providers, and thus reduce the overall rates of returns to pensioners.25 Fourth, financial markets are complex and volatile in both the short term and over periods of retirement savings.26 Market volatility in particular can undermine the objectives of retirement savings and create ‘artificial inequalities in between different birth cohorts’ (pp. 44–45).27

Specifically, these challenges raise the question of whether (a) private pensions should be promoted by the state through subsidies or collectively enforced through mandatory requirements; (b) legal minimum guarantees can be to insulate contributors or some elements of intergenerational transfer could be introduced to ensure income security of retirees; (c) individual freedom of choice be restricted or managed through assets pooling in a common investment portfolio; (d) and administrative changes by service providers be capped or competition be eliminated in favor of centralized accumulation to exploit the merits of economies of scale. Table 1 provides a framework for understanding private pension-related risks and the extent to which they are regulated under each of the three tiers that constitute the Ghanaian pension plan.

Table 1 Risk and regulation in Ghana's three-tier pension plan

Under the new pension plan, in the possibility of individuals undervaluing their future income security needs and hence not setting resources aside for future consumption, is prevented for all formal labor market employees through mandatory participation in tier one. The risk posed to retirement income security by market volatility is also addressed through intergenerational transfers and risk sharing. To prevent risks associated with choice in a competitive market environment, the first tier is designed as a centralized fund (no provider competition) to ensure pooling of assets of all workers in a collective portfolio, managed by the state and the social partners on behalf of participants in the labor market. The strategy of collective pooling of assets in a centralized corporatist fund can result in reaping high economies of scale, and prevent high levels of administrative costs often associated with competitive pension funds. The first tier vests the responsibility for status maintenance in the state and the social partners, although contributions to the scheme are also invested in market. Thus, this tier displays a high degree of collective intervention, and a certain percentage of guaranteed income replacement is ensured through intergenerational transfers, as well as sharing of financial market risks.

The second and third tiers had no mechanisms for mitigating threats posed by market volatility to retirement income security. As defined contribution plans, both tiers are designed to pay out lump sum benefits consisting of total contributions plus accrued interest to retiring participants. They have neither minimum guarantees nor risk-pooling elements to protect consumers against the possibility of market failure and default by providers. Although there exists pension regulatory authority to certify institutional providers of private pension, there is no restriction on the number of providers allowed to operate in the private market. By implication, both the second and third tiers are by default designed to allow unrestricted provider competition, as well as unrestricted individual choice. This means these tiers have no mechanisms to mitigate the challenges posed by information overload and choice risk to consumers of private pension products.

Similarly, as both tiers are designed to foster provider competition, there are no restrictions or caps on how much providers can charge on a participant's contribution in relation to administration and operations cost. It is assumed that in a competitive private market, competition among providers will force them to offer low administrative changes to attract more customers. The second and third tiers are, however, differentiated in two ways. First, whereas in the second tier myopia is prevented through a mandatory provision compelling all formal labor market participants to enroll in a private or occupational pension plan, the third tier is voluntary and encouraged through tax incentives, and thus relies on consumer sovereignty to overcome the problem of myopia. Second, responsibility for status maintenance is borne by individual participants in the third tier; however, such a responsibility is shared between the individual and the state under the second tier, thereby ensuring an intermediate level of intervention.

INSTITUTIONAL WEAKNESSES AND THREATS TO RETIREMENT INCOME SECURITY

The sociopolitical and economic environment in Ghana makes the adoption of private sector solutions for retirement income security problematic for a number of reasons. First, similar to most African countries, Ghana lacks financial market and instruments to hold pension fund. Even if the financial market required to hold pension funds exists, the rate of inflation in Ghana makes such an arrangement undesirable. Second, consumer ignorance and lack of knowledge about the operations of the financial markets in relation to private pensions is a threat to the success of the scheme. Third, private pension scheme necessitates effective and efficient government regulation of financial markets (if they exist) to shelter consumers in areas too complicated for them to insulate themselves. For all these reasons, the design of the first tier offers a better framework for the protection of pensioners, especially because it is mandatory, centralized, monopolistic, based on defined benefit principles, and pools assets in collective portfolio while ensuring intergenerational transfers. Beyond this, the new pension system in Ghana suffers from two major weaknesses that have implications for income security of the aged.

First, similar to previous schemes, the current three-tier pension system is biased in favor of formal labor market participants and largely reflects the interest of urban working class. In all the three tiers under the new arrangement, participants in the informal labor market are expected to contribute towards their retirement income security on voluntary basis. There are no mechanisms in Ghana's three-tier model to deal with the problem of myopia among informal sector workers. Yet, workers in this sector are the most vulnerable, especially because they are usually on low incomes or are self-employed, work in very small unregistered or household companies, and often, but not always, on a part-time basis in industries such as agriculture, construction and services. In most cases, these informal sector workers are unorganized and come from lower income and educated groups, where their knowledge and understanding of pensions is not only limited, but also their resources for long-term savings scarce.29

For the most part, as workers in the informal sector in Ghana are among the poorest, wage-related pension programs30 such as those embedded in the country's three-tier model do not serve the retirement income security needs of the category of citizens who practice their trade in this sector. Even if the problem of myopia is attacked through compulsory requirement as done for formal sector workers – given the pressing demands for food, clothing, housing, education and health – informal sector workers might not be able to give saving for future consumption a priority owing to demands on their resources. Thus, asking informal sector workers to voluntarily save extra money for the distant future is practically not feasible. This issue is even more complicated given that life expectancy among informal sector workers is lower than those in the formal sector.31 Although it can be argued that the Ghanaian state lacked the administrative capacity12 to capture informal sector workers in an earnings-related pension programs,32 the newly introduced three-tier pension program reinforces existing patterns of inequality among the citizenry and makes income security for all elderly Ghanaians possible only under conditions of full formal sector employment for all cohorts at all times.

Second, although NPRA is empowered to resolve all pension-related conflicts, there are no mechanisms to protect participants in the second and third tiers against market failures and or default by fund managers despite the fact that Ghanaians are compelled by statutory regulations to invest portions of their retirement income in private market through these tiers. In the context of Ghana where market institutions are weak, rates of inflations are usually high and unpredictable, and probability of investor fraud is high, compelling citizens to invest retirement contributions in the private market without insulating them against the possibilities of adverse market conditions by way of guaranteed minimum benefits is a risky business. The lack of an effective mechanism to compel fund managers and other institutional intermediaries to provide benefits to participants in the second and third tiers irrespective of market conditions shifts the risks associated with market volatility to participants in these tiers. In addition, although this is a problem of poor institutional design, it is by default a disincentive to fund managers, custodians and trustees to ensure prudential management of workers retirement contributions.

As Ghanaian policymakers were obsessed with using workers retirement contributions as resource pool for capital market development, they glossed over the threats that the new pension scheme, especially the second and the third tiers, poses to retirement income security of contributors. Thus, in both the second and third tiers, it is the contributor who bears the risk that his or her retirement account will increase in value by an amount sufficient to provide adequate income during retirement. If the contributions made to the account by or on behalf of a worker are insufficient, or if the investment portfolios in which the contributions were invested decline in value, the participant risks having an insufficient income in retirement to maintain his or her desired standard of living. It is worth noting that a country that compels its citizens to voluntarily or compulsorily entrust their pension contributions in the private market without any mechanism to ensure guaranteed income replacement in the event of market failure or default by service providers not only exposes its senior citizens to serious retirement income security risks, but also stands the risk of future social upheavals.

Finally, there are too many institutional providers of pension-related services under the three-tier model. There is a great deal of overlap in the functions assigned to service providers such as the custodians, fund managers and trustees. In fact, the sheer number of intermediaries between the contributor and his/her contributions in the retirement savings accounts raises challenges about the institutional ambiguity, especially in areas where the functions or roles of institutional providers overlap. In addition, as service providers rely on pension contributions for their operations under the three-tier model, benefits accruing to pensioners under all (especially tiers two and three) would be adversely affected given the number of institutional providers that have to deal with each contributor's retirement saving accounts.

CONCLUSION: OPTIONS FOR REFORM

Notwithstanding the challenges that prompted the reforms, the major issue that Ghanaian policymakers sought to address in the recent reforms was how to design a pension program in a way that meets the social protection needs of the aged without undermining the pursuit of economic development. However, concerns over economic growth compelled policymakers to design the three-tier model in a way that subordinates income security needs of the aged to the quest for accumulating resources for capital market development through investment of retirement contributions. Thus, under the current arrangement, the income security needs of the aged can easily be compromised if steps are not taken to address the challenges discussed above. The first step in this regard is to activate Article 37 section 6 (b) of the 1992 Constitution by introducing social pensions on the basis of citizenships or residence for the elderly in Ghana. The above constitutional provision mandates the Ghanaian state to ‘provide social assistance to the aged such as will enable them to maintain a decent standard of living’.

Activating this constitutional provision would address the default exclusion of informal sector workers from the current pension scheme by ensuring that individuals who, for reasons pertaining to the nature of their life's circumstances, could not participate in the existing earnings-related programs are not left without any source of income. Such a scheme can be designed as a non-contributory, basic flat rate, but means-tested program directed mainly at poverty alleviation among the elderly through regular (monthly, bi-monthly and so on) income support, and financed from general revenues. As the social pension would be directed at poverty reduction, it is imperative to augment it by relying on institutional practices such as cooperatives and rotational savings clubs to enhance the opportunity structures for the emergence of similar practices for the purposes of protection of the aged.32 Within this framework, additional allowances could be given by the state, mutual and cooperative schemes or private households to the elderly with the physical and mental capabilities to help in raising children either under institutional daycare or within household, thereby freeing parents to actively participate in the labor and contribute to growth.

Second, as the second and third tiers in the current arrangement shift risks associated with private pension plans to individual employees or participants, it is imperative that the state, which mandates its citizens to invest their retirement contributions in such plans, establish a statutory pension benefit insurance agency to protect plan participants (citizens) against adverse market conditions and challenges associated with default by fund managers, custodians and trustees. Such an agency can be financed primarily through insurance premiums set by either parliament or the NPRA, and paid by fund managers, custodians and trustees. Other sources of financing the operations of such a pension benefit insurance agency should include returns on invested premiums, takeover of assets from defaulting retirement contributions investors like the fund mangers and other intermediaries. The maximum pension benefit to be paid by this statutory agency, should the need arise, must be set by law and reviewed annually to accommodate changes in the market.

Beyond the fact that aspects of the operations of this agency would also contribute to capital market development, it would ensure some measure of protection for contributors, insure the operations of institutions investing workers’ retirement contributions under the two tiers, and prevent the possibility of using monies from the country's general revenues for pension fund bailout and or bankruptcy protection. In order to insulate retirement contributions in the second and third tiers from mismanagement by investors, policymakers can cap or specify a threshold for administrative cost on retirement income contributions for the intermediaries between individual contributors and their funds, and or require a certain minimum percentage of returns on retirement contributions invested. This would foster a healthy competition by ensuring that only investors of retirement contributions that can offer benefits higher than this minimum would remain in the market.

Finally, it is important to consider a reduction in the number of pension service intermediaries from the current four, particularly for the second and third tiers. This implies that the functions performed by the custodians, trustees and pension fund managers must all be collapsed into one and vested in fund managers. The advantage of downsizing of the intermediaries is that, although they can be several fund managers operating in the market especially for the purposes of competition, they will reap economies of scale in a manner that will minimize the quantum of participant contributions spent by service providers on their operations. Reducing the number of institutional service providers would make it easier for contributors to understand who deals directly with their retirement income contributions and monitor the performance of their individual retirement accounts from a single source. In other words, the opportunity structures for transparency and accountability would be enhanced once the existing intermediaries are scaled down.