Housing deficit in Ghana is estimated as about 1.7 million affordable housing units. A lack of development finance has been identified as one of the limitations to affordable housing delivery. In an attempt to reduce the deficit, the Minister for Works and Housing, Lawyer Atta Akyea has proposed the channelling of a portion of the first tier defined benefit (DB) pension funds into a government-sponsored ‘affordable housing’ project. It is important to note that this proposal although radical is not new; some housing researchers in Ghana hold similar views. Ayitey et al (2010) for example made a similar proposition.

Such proposals follow the idea that pension funds are cheaper sources of investable funds and have long-term investment horizons, which is perfect for housing investment. Despite some level of truth in this claim, I will be quick to point out the inaccuracy in the view that pension funds are necessarily cheap funds; because it depends on who is borrowing it. Government is deemed to be risk-free (theoretically) and hence can borrow the pension funds at lower (risk-free) rates compared to corporations and individuals. Indeed, the term structure of interest rates indicates that the yield on a long-term instrument is often higher than short term instruments[i]. This may be due to the payment of an illiquidity premium to the investor for deferring their consumption to a later date[ii]. So, pension funds although may be patient funds, are not necessarily cheap funds as the Minister touts in the media.

His argument about maturity matching of pension fund investments and housing is also valid so far as the demographic structure of Ghana’s population favours the youth and our pension fund is immature – no pressing need for liquidity because there are a few pensioners to receive pension pay-outs. This demographic advantage may exist currently given that more than 50% of Ghanaians are considered youthful (57% are 24 years and below) but there is a need for caution.In this brief article, I discuss a few reasons why caution must be taken in our attempt to force the Social Security and National Insurance Trust (SSNIT) by law to allocate portions of the first-tier pension funds to the government so-called ‘affordable housing’ project.

First, it is important to note that government’s ‘affordable housing’ projects have often failed to deliver affordable housing to the masses who are usually targeted[iii]. It is very much like government scholarships for the poor or Cocobod scholarship for the kids of cocoa farmers – the rich and politicians often capture it! So, what is new about this new housing project? There is not a robust evidenced-based for Ghanaian government’s direct involvement in affordable housing development and financing. There is a need to clarify the exact role the government is playing in the delivery of this project.

Second, it is also important to state that a chunk of SSNIT funds are required by law -National Pensions Act, 2008 (Act 766) as amended - to be investment in government securities like Treasury bills and bonds (see sections 175 – 178 of Act 766). So, SSNIT is already substantially exposed to government’s risk-free instruments. Thus, investing another portion of the first-tier funds managed by SSNIT in the government’s affordable housing project will increase the funds exposure to government-linked projects. Two kinds of risks are envisaged here. The first is that pension contributors may suffer from low returns on government securities. Second, pension contributors are likely to be exposed to a high risk of loss depending on whether the associated default risk is guaranteed. This may be the case if the funds are utilised as development finance, as in a case of a loan to the private developers involved in the scheme. If default risk is not guaranteed, then SSNIT may also be overly exposed to risk than required based on portfolio requirement. This may jeopardize the underlying defined benefits to pensioners if the housing investments fail or yield lower than expected returns. The risk of failure is high given the frequent abandonment of government affordable housing projects with changes in government.

A further increment in investments in government securities via this housing scheme may also create diversification problems for the fund manager (SSNIT) and pensioners at large. DB pension schemes guarantee a certain level of pension pay-outs and as attract lower returns. The limitation imposed on SSNIT by law in terms of what assets and the percentage of fund is required to be invested in it (see section 175 – 178) serves as a constraint on portfolio diversification. Meanwhile, diversification is expensive and requires a lot of money over and above what may be left after deducting the statutory allocations to government securities. The capital-intensive nature of real estate assets may worsen the diversification constraint. Hence, SSNIT may end up with a poorly diversified portfolio with potential negative ramifications for pension contributors.

Even without the political risk of project abandonment and despite the diversification potential of real estate as an asset class in investment portfolios, real estate in general has similar risk-return profiles like ‘risky’ equities (stocks)[iv][v][vi]– buying a share of a company. Although some investment strategies like core (fully tenanted) and core-plus (low vacancy) real estate may have low risks, real estate development; i.e. developing a property from scratch as in this proposed scheme, attracts the highest risk and may expose pension contributors to unnecessary risk of real estate if not guaranteed somehow. This may be valid especially if an optimum allocation to real estate has already been reached. This is further worsened by the poor record of SSNIT’s management of its housing investments. The case of SSNIT flats is a typical example. A great deal of real estate investment management expertise, which may not exist currently, is needed by SSNIT to protect the investments of contributors. A rare combination of expertise in affordable housing investments, real estate finance and pension fund management may be needed!

The risk of excessive exposure to risk is even higher if we think about the new three-tier pension fund holistically. In view of the need to enhance pension pay-outs to pensioners, the second tier defined contribution scheme and the voluntary third tier was introduced to enable investment of portions of the pension contributions in higher yielding assets to diversify the low returns from the government-dominated first-tier scheme. Therefore, a certain diversification structure has already been imposed by law. Coupled with the lack of clarity about how the SSNIT funds would be used in this new proposal by the Minister, whether as development finance – loan to private developers for the development of the houses - or end-user finance to households for the purchase of the house, the investment of pension contributions may suffer from unwarranted high risk without a guarantee of high returns.

Perhaps, the Minister’s argument should be that a portion of the first-tier funds lent to the government, not the remainder after allocations to government securities, should be channelled into affordable housing. Many people are likely to support such a proposal because such a move represents a more productive use of scarce funds in securing the real needs of the people whiles they are alive than the wasteful investments of governments sometimes. The other option is to pursue a regulation for the implementation of section 103(2) of the Pension Act (Act 766), which enables a pension contributor to secure a mortgage for a primary residence with his/her accrued second-tier pension assets. This is technically called pension asset-backed housing finance, the practice of which is common in South Africa, Singapore, Brazil, Mexico, Kenya, and Zambia among other countries. Interested persons may read my articles on pension asset-backed housing finance for details[vii][viii].

The specifics of the Minister’s intentions are missing, but the few information available in the media about his proposed intention to force SSNIT to invest in affordable housing raises a red flag for comprehensive benefit cost analysis. We must therefore hasten slowly in action and invest some quality time and effort in understanding the implications of the Ministers seeming innovative proposal.

Dr. Kenneth Appiah Donkor-Hyiaman

The author is a Real Estate Consultant, Researcher, Lecturer, and Managing Partner at MeTis Brokers (a private equity real estate investment and management firm).

[i] Malkiel, B.G., 1989. Term structure of interest rates. In Finance (pp. 265-270). Palgrave Macmillan, London.

[ii] Modigliani, F., 1944. Liquidity preference and the theory of interest and money. Econometrica, Journal of the Econometric Society, pp.45-88.

[iii] Arku, G., 2009. Housing policy changes in Ghana in the 1990s: Policy review. Housing Studies24(2), pp.261-272.

[iv] Ross, S.A. and Zisler, R.C., 1991. Risk and return in real estate. The Journal of Real Estate Finance and Economics4(2), pp.175-190.

[v] Chan, K.C., Hendershott, P.H. and Sanders, A.B., 1990. Risk and return on real estate: evidence from equity REITs. Real Estate Economics18(4), pp.431-452.

[vi] Devaney, M., 2001. Time varying risk premia for real estate investment trusts: A GARCH-M model. The Quarterly Review of Economics and Finance41(3), pp.335-346.

[vii] Donkor-Hyiaman, K.A. and Owusu-Manu, D., 2016. Another look at housing finance in Africa: The anatomy of pension asset-backed housing financing. International Journal of Housing Markets and Analysis9(1), pp.20-46.

[viii] Afrane, S.K., Owusu-Manu, D., Donkor-Hyiaman, K.A. and Bondinuba, F.K., 2016. Towards Innovative housing financing in Ghana: an evidence-based from South Africa’s pension housing financing system. Public Policy and Administration Research, 4(4).