INTRODUCTION
Section 37C of the South African Pension Funds ActFootnote 1 (the Act) regulates the payment of any benefit payable upon the death of a member of a pension fund, and places a duty on the board to distribute the death benefit. In addition, the language of section 37C(1) provides that “regardless of the provisions of any other law, including the common law and notwithstanding the rules of a registered fund, any benefit payable upon the death of a pension member must be dealt with in terms of the scheme outlined in the Act”.Footnote 2 Section 37C therefore overrides the freedom of testation in relation to the benefits payable by a fund in the event of the death of a pension member and gives the board discretion to distribute such benefits.Footnote 3 The guiding principle is that the assets do not form part of the deceased's estate nor do the provisions of the intestate succession legislation apply;Footnote 4 instead the assets are required to be distributed in accordance with the statutory scheme which gives preference to need and dependency above the pension member's choice.Footnote 5 Effectively, under section 37C of the Act, the needs of dependants override the pension fund member's decisions on how their pension savings should be distributed.
The policy underlying this social security measure is to ensure that monies in respect of which the state allows major tax concessions should, in theory, be applied towards the benefit of the deceased member's surviving spouse, children and other dependants, thereby reducing the state's liability and promoting social protection.Footnote 6 Commenting on the purpose of section 37C, Hussain J in Mashazi v African Products Retirement Benefits Provident Fund has also explained that “[s]ection 37C is a social protection type measure, [because] it places the benefit payable on a member's death under the control of the retirement fund with the discretion to pay it to the member's dependents in such proportions as it deems equitable, thereby reducing the state's liability in this regard.”Footnote 7
While the legislature's worthy intentions in enacting section 37C are admirable, the pension funds adjudicator critically commented that section 37C:
“is a hazardous, technical minefield potentially extremely prejudicial to both those who are expected to apply it and to those intended to benefit from its provisions. It creates anomalies and uncertainties rendering it most difficult to apply. There can be no doubt about its noble and worthy policy intentions. The problem lies in the execution and the resultant legitimate anxiety felt by those who may fall victim to a claim of maladministration in trying to make sense of it. Any successful claim for maladministration will be borne ultimately by the other members, the participating employer, or perhaps even the members of the board of management.”Footnote 8
The difficulties brought about by section 37C arise out of the three duties the section imposes on the board of management: to identify a member's dependants; to effect an equitable distribution of the benefits amongst the beneficiaries; and to determine an appropriate mode of payment. This article will focus on the first duty imposed on the board, to identify potential dependants.Footnote 9
Section 1 of the Act adopted a broad definition of a dependant, namely:
“(a) a person in respect of whom the member is legally liable for maintenance;
(b) a person in respect of whom the member is not legally liable for maintenance, if such person -
(i) was, in the opinion of the board, upon the death of the member in fact dependent on the member for maintenance;
(ii) is the spouse of the member;
(iii) is a child of the member, including a posthumous child, an adopted child and a child born out of wedlock;
(c) a person in respect of whom the member would have become legally liable for maintenance, had the member not died”.
According to this definition, the legislature created three classes of dependants: a legal dependant in paragraph (a); a factual dependant, a spouse and a child in paragraph (b); and a future dependant in paragraph (c). This note focuses on cohabitants and whether they qualify as factual dependants in terms of the definition in paragraph (b). The issue in Hlathi v University of Fort Hare Retirement (Hlathi)Footnote 10 was whether a cohabiting partner, who had been in a relationship with the deceased member for 17 years, was a factual dependant under the Act. Brigitte Clark defines cohabitation as a stable, monogamous relationship where couples who do not wish to, or are not allowed to, get married and live together as spouses.Footnote 11 In Volks v Robinson (Volks), the Constitutional Court acknowledged lack of regulation in this area and called on Parliament to regulate it.Footnote 12
This note will compare early decisions in Van der Merwe v Southern Life Association Footnote 13 and Van der Merwe v Central Retirement Annuity Fund Footnote 14 with the decision in Hlathi. It is important to note that the decisions of the pension funds adjudicator create a precedent and therefore the adjudicator has to give reasons for departing from an earlier judgment. The adjudicator ruled in Beukes v Pepkor Retirement Fund Footnote 15 that, under section 30O of the Pension Funds Act, her decisions create legal precedent and pension funds must apply their minds to the rulings by the adjudicator. In the first two decisions the adjudicator applied two different tests to determine whether a cohabiting partner qualifies as a factual dependant under the Act. This note argues that the first test in Van der Merwe v Southern Life Association, which requires cohabitants to have a mutual dependence and share a common household (which has now been followed in Hlathi), should be welcomed. It rejects the dominant-servient test as laid down in Van der Merwe v Central Retirement Annuity Fund, and concludes that Hlathi contributes positively to the largely unregulated issue of cohabitation and therefore gives hope to cohabitants, especially women.
COHABITANTS VERSUS SPOUSES UNDER THE ACT
In order to understand the significance of the Hlathi decision, it is important to identify the two types of dependants under paragraphs (b)(i) and (ii) of the definition: cohabitants and spouses respectively. Under paragraph (b)(ii), a spouse is defined as a person who is the permanent life partner or spouse or civil union partner of a member in accordance with the Marriage Act,Footnote 16 the Recognition of Customary Marriages Act,Footnote 17 the Civil Union ActFootnote 18 or the tenets of a religion.Footnote 19 In other words, the Act recognizes three classes of spouses: civil marriage spouses, religious marriage spouses and civil union spouses. It is also evident that, for a spouse to be recognized for the purposes of the Act, it must be proven that she or he is a spouse of a deceased member in accordance with the Marriage Act, the Recognition of Customary Marriages Act or the tenets of a religion, or a civil union partner under the Civil Union Act. This means those who have participated in some form of a religious marriage such as a Muslim or Hindu marriage, but whose marriage is not recognized,Footnote 20 are included. Excluded from the definition are heterosexual cohabitants who have not gone through any form of religious or traditional marriage ceremony. This is in line with the Constitutional Court decision in Volks,Footnote 21 where the majority held that a person, who could have married the deceased but chose not to, should not be granted the rights of a spouse.
Nevertheless, even if a person is not a spouse under the legislation, he or she can still be considered a factual dependant under paragraph 1(b)(i). John Murphy, the first adjudicator, stated in Van der Merwe v Southern Life Association that section 1(b)(i) of the Act cannot be interpreted literally and in isolation; he therefore applied a purposive and contextual interpretation.Footnote 22 Murphy held that the purpose of the legislature in enacting the provision was to broaden the category of persons entitled to share in death benefits by including persons involved in relationships which the law does not traditionally accept as constituting legal dependency. He reasoned that the provision has the progressive aim of recognizing that modern society is tolerant of relationships besides the nuclear family arrangements sanctioned by the common law.Footnote 23 The dependants who fall under paragraph (b)(i) include persons who simply live together and for their own good reasons have not observed the formalities of entering into a formal marriage.
Accordingly, Murphy laid down a two prong test for determining whether a person is a dependant under paragraph (b)(i). The first prong of this test is concerned with determining whether the parties demonstrated a mutual dependence. According to the adjudicator, the inquiry is qualified by the requirement that mutual dependence must involve, among other things, an emotional and intimate or sexual bond between the parties.Footnote 24 The second prong of the test is concerned with determining whether the parties ran a shared and common household.Footnote 25 He also held that the true intention of the definition of a dependant in section 1 is to grant to the board discretion to accord same sex couples and unmarried cohabitants the same rights as legally married couples.Footnote 26 The reason is that such an interpretation would be in keeping with the spirit of the Bill of Rights.
However, in Van der Merwe v Central Retirement Annuity Fund,Footnote 27 a cohabitant partner was not recognized as a factual dependent under the Act. The second adjudicator, Vuyani Ngalwana, rejected Murphy's test in Van der Merwe v Southern Life Association and stated that something more is required than for cohabitants to be financially mutually dependant.Footnote 28 He held that to qualify as a factual dependant under paragraph (b)(i) required at the very least a dominant-servient relationship, in which one party is the substantive provider.Footnote 29 In this case, the adjudicator was faced with two sets of dependents: the deceased's two children (one of whom was still a minor) from a previous marriage, and his partner with whom he had been cohabiting for three years preceding his death. The pension fund had allocated 35 per cent to the children with the rest being given to the cohabiting partner. On an appeal brought on behalf of the children, the adjudicator found in favour of the children. Ngalwana found that, in the case before him, there was no evidence to suggest that the second respondent was factually dependent on the deceased for the purposes of section 1(b)(i) as alleged by the fund. This is because it appeared from the papers that she was the substantial creditor in the deceased's estate.Footnote 30
Ngalwana stated that this would indicate a pooling of resources, or mutual dependence, which does not suffice for the purposes of “factual dependence” as defined in the Act. It was also acknowledged by the fund that the cohabiting partner was employed and received maintenance from her previous husband. Ngalwana also found that the decision by the board to regard the cohabiting partner as a factual dependant was based on the fact that the deceased and the second respondent were mutually inter-dependent, having pooled their financial resources.Footnote 31 On this basis Ngalwana ruled factual dependency was not established.
The effect of this ruling was that boards of pension funds, in exercising their powers under section 37C, could only include a surviving cohabitant as a factual dependant in circumstances where he or she is able to prove that the deceased member was a dominant financial provider in their relationship. Ngalwana therefore held that the board had erred in considering the cohabiting partner to be a dependant of the deceased and the decision had to be set aside. Interestingly, this was not the first time Ngalwana used the dominant-servient test. He used this test for the first time in De Wizhem v South African Retirement Annuity Fund,Footnote 32 where he stated that “factual dependence” contained within paragraph (b)(i) must comprise a financially dominant/subservient relationship. If this were not the case, casual relationships would fall within the definition of a factual dependant, provided the parties were pooling financial resources.Footnote 33 This legal position prevailed until Hlathi was decided in 2009.
THE BACKGROUND TO HLATHI V UNIVERSITY OF FORT HARE
This was a determination by Mamodupi Mohlala, the third adjudicator, concerning the distribution of a death benefit under section 37C of the Act. The issue in this case was whether the board correctly identified the potential dependants and distributed the benefits pursuant to the Act. The facts of the case were as follows. The deceased was a member of the first respondent, a registered pension fund (Fort Hare Retirement Fund).Footnote 34 Following the deceased's death, a benefit in the amount of R400,000 before tax became available for distribution to his beneficiaries.Footnote 35 Investigations by the board of trustees of the Fort Hare Retirement Fund established that the deceased had two dependants. The first dependant was the deceased's mother, who was also the complainant in this matter. The second dependant was the deceased's surviving cohabiting partner, who was joined as a third respondent in this matter pursuant to section 30G(d) of the Act.Footnote 36 The board resolved to allocate one third of the benefit to the mother and two thirds to the cohabiting partner.
The complainant lodged a complaint with the Office of the Pension Funds Adjudicator pursuant to section 30M of the Act. She was unhappy with the trustees' decision to include the third respondent in the distribution of the death benefit and alleged that she was the sole beneficiary because the deceased was unmarried and did not have any children. The complainant also alleged that the cohabiting partner was not the deceased's dependant since she was employed in a well remunerated job (the facts do not disclose the cohabiting partner's job). Accordingly, the complainant requested the adjudicator to set aside the board's decision and replace it with an order allocating 100 per cent of the benefit in her favour.Footnote 37
As an interested party, the cohabiting partner submitted that she had a formal relationship with the deceased which lasted for a period of 17 years until his untimely death. She added that she and the deceased stayed together as husband and wife for a period of nine years until his death and furnished proof that indicated the deceased regarded her as his lifelong partner. The cohabiting partner also provided a detailed proof of shared expenses including a 50 per cent contribution towards bond repayments, household insurance, motor vehicle insurance and maintenance costs, telephone costs, family holidays and general entertainment costs. To further demonstrate that they lived as husband and wife, the third respondent stated that they could sign each other's cheques, and that they had a mutually supportive, loving and happy relationship.Footnote 38 She also disclosed that she contributed financially towards the deceased's tuition and maintenance whilst he was studying, as she had been the only source of financial support at the time. She also stated that they had plans to get married when the deceased had completed his studies.
THE ADJUDICATOR'S DECISION AND REASONING IN HLATHI V UNIVERSITY OF FORT HARE
In resolving this complaint, Mohlala first addressed the question of whether the third respondent qualified as a spouse within the meaning of the Act. In analysing the meaning of spouse under the Act, Mohlala found that the use of the word “permanent” in the definition of a spouse under the Act clearly depicts that the legislature intended that parties or persons who profess to be spouses in this context have an element of permanency to their life partnership. Mohlala then referred to the ruling by the majority of the Constitutional Court in Volks. In that case, the majority held that, for the purposes of maintenance claims against the deceased's estate (under the Maintenance of the Surviving Spouses Act), the different treatment of formally married spouses, on the one hand, and cohabitants in a permanent life partnership, on the other, is constitutional.Footnote 39 In light of this ruling, Mohlala held that the third respondent could not qualify as a dependant by virtue of being regarded as a “spouse” for the purposes of paragraph (b)(ii) for two reasons: that the Act does not apply retrospectively in this instance (the Act came into operation on 13 September 2007, while the deceased died on 27 March 2002); and that the third respondent's relationship with the deceased was not solemnized and registered by way of either a marriage or civil union or partnership.Footnote 40
Having found that the third respondent did not qualify as a spouse under paragraph (b)(ii), Mohlala proceeded to determine whether the third respondent qualified as a factual dependent under paragraph (b)(i). Under this analysis, Mohlala scrutinized Ngalwana's decision in Van der Merwe v Central Retirement Annuity Fund and rejected the view that factual dependency requires a dominant-servient relationship in which one party is the substantive provider.Footnote 41 She reasoned (therefore reaffirming Murphy's decision in Van der Merwe and Others v Southern Life Association and Another) that a purposive and contextual interpretation of paragraph (b)(i) reveals that the true intention of the legislature in enacting the provision was clearly to give effect to the purpose of section 37C of the Act, which is to protect proven dependency even for persons who are involved in relationships that the law does not accept as constituting legal dependency.Footnote 42 Furthermore, Mohlala noted that, since the legislature did not refer to “totally or wholly” dependent means, the test for dependency under paragraph (b)(i) was not exclusive. In other words, she found that “wholly” dependent is not the sole yardstick to determine or measure dependency for the purposes of paragraph (b)(i). Accordingly, it would be contrary to the intention of the legislature to exclude a party for the purposes of paragraph (b)(i) on the basis that he or she had an “inter-dependent” relationship with the deceased member, or alternatively that the parties had an equal relationship as opposed to a dominant-servient one.Footnote 43
According to Mohlala, factual dependency in such cases is proven by sufficient evidence that the parties in a relationship are inter-dependent, and as a consequence of the other party's death the surviving partner is left in a financial predicament or financially worse off.Footnote 44 In this matter, Mohlala noted that the evidence indicated that the third respondent and the deceased contributed equally towards bond repayments and other household expenses, and that the third respondent submitted that she and the deceased mutually supported each other in a relationship that lasted for a period of 17 years. Mohlala also noted that the financial position of the third respondent had changed dramatically since the death of the deceased.Footnote 45 As a result, Mohlala was satisfied that the third respondent and the deceased were inter-dependent and that, without the deceased, her lifestyle would no longer be the same. Therefore, according to Mohlala, the third respondent qualified as a factual dependent under paragraph (b)(i) of the Act, and she ruled that the third respondent was correctly identified as a dependant and awarded death benefits. In other words, Mohlala found that a cohabiting partner can qualify as a factual dependant under the Act as long as it is proven that the parties had an inter-dependent or equal and mutual relationship. The test by Mohlala is an expansion of Murphy's test in Van der Merwe and Others v Southern Life Association and Another, for Mohlala understands that a shared common household and mutual dependence between the parties can qualify as factual dependence under the Act. Added to that is that the parties can also be inter-dependent or equal and, as a result of the death of one party, the surviving cohabitant can be left in a financial predicament or be financially worse off. However, Mohlala omitted to discuss the requirement that mutual dependence must involve, among other things, an emotional and intimate or sexual bond between the parties as was decided by Murphy in Van der Merwe and Others v Southern Life Association and Another and it is not clear why she did so.
THE IMPLICATION OF HLATHI V UNIVERSITY OF FORT HARE RETIREMENT FUND
This determination is relevant today because it pertains to the very important issue of cohabitation. Since Parliament has not yet acted on this matter, the Hlathi determination is an important case that seeks to regulate this area in the context of pension fund organizations. It is the author's view that this determination gives some form of relief to the cohabitants in a cohabitation relationship who are left in a predicament by the death of their partner who is a member of a pension fund.
The other significance of the ruling in Hlathi is that it rejected the dominant-servient test to determine factual dependency in cases of cohabitation. Mohlala's decision should be welcomed for a number of reasons. First, there is nothing in the Act that refers to a dominant-servient or wholly dependent relationship as a standard for factual dependants. Secondly, the determination by Ngalwana in Van der Merwe v Central Retirement Annuity Fund is rejected and criticized because it is based on a patriarchal notion that women are servants and men are masters, because the reality is that in most cases it is women who are in need of maintenance after the termination of a relationship. It has been reported that women earn less than men, it is women who are caught in a cohabitating relationship where men do not wish to get married but women do, it is women who are caught in a predicament whether to leave a non-committing man in a cohabitant relationship or to leave a relationship and face poverty.Footnote 46 Ngalwana's attitude in Van der Merwe v Central Retirement Annuity Fund reflects the operation of a colonial ideology of gender where women are characterized as subordinates to men and he seems happy to keep the status quo.Footnote 47
It is surprising to find a determination concluded in 2005 which still used the term “dominant-servient”, which has colonial, apartheid and patriarchal connotations. This language employed by Ngalwana was previously used to oppress and indignify African people during the colonial and apartheid eras, and has been rejected by the constitution. There is no doubt that the constitution was adopted to improve the quality of life of all citizens of (and those who live in) South Africa.Footnote 48 The constitutional guarantee of equality entails an historical understanding of the type of society that South Africa once was and against which the new constitution has set itself, because the past political and legal system was based on inequality and unfair discrimination.Footnote 49 The term dominant-servient has no relevance in the interpretation of the Act.
CONCLUSION
There is no doubt that cohabitants still have a long way to go before they can be afforded full legal protection as spouses in South Africa. Hlathi is the beginning of such a journey in affording cohabitants some legal protection.
