Analysis of South African Pension Fund Conversions: 1980–2006; Developing a Model for Dealing with Environmental Change¶
George, Dion Travers (2006). Analysis of South African Pension Fund Conversions: 1980–2006; Developing a Model for Dealing with Environmental Change. Doctor of Business Leadership thesis, University of South Africa (Promoter: Prof. R.J.O. Joubert). University of South Africa.
| Type | thesis |
| Scope | Wider apartheid literature |
| Method | survey fieldwork |
| Period studied | 1980–2006 |
| Regions | South Africa, Johannesburg, Cape Town, Durban |
| Themes | labour markets wages, unions strikes, business and state, resistance social movements, transition post apartheid, inequality poverty unemployment |
| Citation key | george2006 |
Argument¶
The thesis explains why thousands of South African pension funds converted members from defined benefit to defined contribution structures between 1980 and 2006. It identifies two waves: a 1980s wave driven by emerging black trade unions, who treated control of pension assets as part of the liberation struggle, and a 1990s wave driven by employers on the advice of consultants, who wanted to cap open-ended liabilities and pre-empt member-elected trustees. The author argues that the transition from apartheid was an 'environmental shock' that changed the rules of the game. Under that shock the conversions were too fast for sufficient thought, and uncertainty, power shifts and ethical failings followed. These generated unintended consequences such as stranded surplus and risk shifted to members, which later required redress through the surplus legislation. The evidence is 55 interviews with 44 experts plus eight fund case studies; from these the author builds a general managerial model, applied to pension reform and B-BEE.
Evidence and method¶
Semi-structured interviews with 44 industry experts (55 interviews, 2005) across government, regulator, adjudicator, ombudsman, unions, employers, trustees and service providers; eight fund case studies (Funds A–H) using trustee minutes, court judgments and interviews; secondary literature from management, finance and welfare economics; SEIFSA and South Africa Foundation background publications; National Treasury (2004) discussion paper.
Place in the debates¶
- Who drove the pension fund conversions: unions, employers or consultants? (see Post-apartheid economic policy, governance and corruption). Two waves: unions drove the 1980s transfers to provident funds, employers and consultants drove the 1990s conversions. Consultants (one 'emerging brokerage' in particular) were drivers, not merely facilitators.
- Did trustees make the conversion decisions (literature) versus company boards? (see Post-apartheid economic policy, governance and corruption). Finds the literature wrong: in all cases company boards decided and trustees merely implemented.
- Whether business and employers acted in members' interests during the transition (see Post-apartheid economic policy, governance and corruption). Argues business acted to restore certainty for itself, and members' needs were not the primary motivation of any stakeholder.
Findings¶
-
Conversions came in two waves with different drivers, and a third was beginning. (p. 4)
the conversion phenomenon occurred in two waves – one initiated in the 1980s and driven by the trade unions and a second in the 1990s, driven by employers
-
The first wave was union-led, politically motivated, and concentrated in the unionised metal industries. (p. 232)
In the 1980s the first wave was politically motivated and driven by the trade unions.
-
Early (union-wave) transferees got minimal benefits and none of the surplus. (p. 232)
When members transferred, they received minimal transfer benefits and no part of any “surplus” that may have existed in the fund at the time.
-
Second-wave employer conversions offered actuarial liability plus a sweetener, with the surplus largely retained. (p. 211)
in all instances of conversion (refer to cases A, B, C, D, E and G) members received an amount equal to their actuarial liability, plus a “sweetener” ranging from 10–30 per cent.
-
Employers conceded to unions in the 1980s to end strikes and work stoppages. (p. 223)
they initially resisted, but then relented because they wanted to end the strikes and work stoppages to get production back on track.
-
Unions acted as proxies for the banned liberation movements, so pension control was a political issue. (p. 236)
trade unions became the proxy of the government in exile and agitated for control over black members’ assets in pension funds.
-
Employers' second-wave motive was fear of the 50 per cent member-elected trustee requirement and rising costs including AIDS. (p. 211)
This entailed reducing the risk that was perceived in members electing at least 50 per cent of Boards of Management.
-
Actuaries redefined investment reserves as 'surplus', which was left behind in transfers (up to 40 per cent of the actuarial reserve). (p. 52)
Depending on the stockmarket, this difference could be up to 40 per cent of the value of the actuarial reserve.
-
Sanctions and the 1985–86 downturn raised employer sensitivity to pension liabilities. (p. 114)
In 1985–86 the economy faltered as the Rand dropped and sanctions started to impact. Pension fund assets took a knock.
-
Stakeholder power shifted from unions to service providers over 1980–2006. (p. 225)
Service providers have largely replaced employers and control the industry – often by having a persuasive influence over trustees.
-
Surplus legislation (2001) was the redress, reaching back to 1980 leavers. (p. 220)
The amendment requires funds to consider former members who left the fund from 1980 onwards.
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Semi-structured interviews with industry experts (44 respondents, 55 interviews), macro-level | interviews oral history | March–November 2005 | Main evidence for the account of two conversion waves, stakeholder motives, speed, consequences, power and ethics; presented as verbatim thematic responses. |
| Eight fund case studies (Funds A–H) with interviews and fund documentation (Appendices 8–15, 17) | author survey or fieldwork | 1980–2006 (data collected 2005) | Sub-cases used to confirm the expert interviews and to compare conversion method, driver, transfer values, surplus and later waves across funds. |
| Fund documentation: minutes of Boards of Trustees, court judgments, consultant reports, communication packs, valuator memos | company records or reports | 1980s–2005 | Documentary triangulation of case chronologies, e.g. Fund E conversion timeline, sweetener sizes and take-up. |
| Expert interview list and interview instruments (interview guide, contact summary, document summary, case analysis forms) | interviews oral history | March–November 2005 | Documents the sampled expert stakeholders (government, regulator, unions, employers, service providers, trustees) and the protocol used. |
| Table 1: respondent ratings of relative stakeholder power (1980s, mid-1990s, 2006) | author survey or fieldwork | 1980s–2006 | Respondents' scaled power scores for stakeholders converted into tables and graphs to show the shifting power dynamic. |
| National Treasury, 'Retirement fund reform' discussion paper, December 2004 | commission or official report | 2004 | Supplies the claim that 1980s–90s saw a dramatic transfer from defined benefit to defined contribution, plus umbrella fund growth and reform proposals. |
| Pension Funds Act 24 of 1956 and the Pension Funds Second Amendment Act (surplus apportionment; effective 7 December 2001) | legislation gazette | 1956–2001 | Framework for conversions (Section 14 transfers, 50% member-elected trustees) and the retrospective 'redress' of surplus apportionment back to 1980. |
| Official commissions and committees on retirement funding (Mouton 1992, Katz 1995, Smith 1995, NRCF 1997, Taylor 2002) | commission or official report | 1992–2002 | Cited as background assessments of retirement funding and the regulatory gap; Mouton referred to by respondents as prompting the adjudicator. |
| Wiehahn Commission (1977) recommendations on labour legislation | commission or official report | 1977–1979 | Context for emergence of registered black trade unions, via the SEIFSA brochure account. |
| SEIFSA 'Strength in Diversity' brochure (60th anniversary, 2003) | secondary literature compilation | 1930s–2003 | Source of historical background on the metal industries, industrial council system, strikes and the industry pension funds. |
| Business in South Africa in 2003 (South Africa Foundation) | secondary literature compilation | 1994–2003 | Supplies macro-economic and business context (GDP growth, savings, JSE ownership shares, BEE) for the post-1994 'environmental shock'. |
| Wellsted (2000), The Actuary; Andrew (2000) FSB Chief Actuary statement; Murphy (2001) on investment reserve/surplus | secondary literature compilation | 2000–2001 | Establishes how the investment reserve was redefined as surplus and the early transfer terms members received. |
| Metal Industries pension fund records (EIPF, MIPF, MIBFA) and related Registrar/Adjudicator correspondence | company records or reports | 1957–2004 | Documents fund establishment dates, membership (30 434 and 208 017), surplus of about R9 billion and the dispute over whether the PFA applies. |
| Adjudicator rulings and court cases (Sentrachem, Pepkor Retirement Fund v FSB, CAF) | commission or official report | 1998–2002 | Illustrative cases showing surplus manipulation and FSB/regulatory responses. |
| FSB data on unclaimed benefits; Registrar of Pension Funds annual report 2004 | official statistical series | 2002–2005 | Figures on unclaimed benefits (R970 million, 967 funds reporting) and pension fund assets over R1 trillion. |
| COSATU (1992) proposals on pensions; Cameron and Heystek (1998); Stafford (1993) and other South African pension studies | secondary literature compilation | 1981–2001 | Used for the union-side reasons for defined contribution funds and the 1980s politicisation of pension control. |
| DTI Codes of Good Practice (Nov 2005), BEE Act 2004, Financial Sector Charter (Oct 2003) | legislation gazette | 2003–2005 | Basis for applying the proposed model to the pension fund perspective on B-B BEE. |
| AC116 (1986) accounting standard on retirement benefit disclosure | legislation gazette | 1986 | Cited by respondents as requiring disclosure of pension contingent liabilities, a driver of employer conversions. |
New data¶
Qualitative dataset built by the author: 55 interviews with 44 industry experts (2005) across government, regulator, unions, employers, trustees and service providers, plus eight fund case studies (A–H) with interviews and documents; coded into themes and cross-case tables. Respondents were confidential and sampling was purposive and snowball.
Inference¶
Quantification: Illustrative figures.
Qualitative interpretivist multiple-case study (Eisenhardt/Miles & Huberman/Yin). Expert interview themes (horizontal and vertical analysis) are triangulated with eight purposively sampled fund cases, with within-case and cross-case comparison. Generalisation is analytic to a theory-building 'environmental shock' model, not statistical.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Conversions occurred in two waves: a union-driven wave in the 1980s and an employer-driven wave in the 1990s advised by consultants. (p. 232) | Expert interviews and cross-case table (Appendix 17) of Funds A–E and G; fund-level conversion dates and drivers. | Partial data |
| Employers moved to defined contribution to cap open-ended liability and to pre-empt 50% member-elected trustees, and consultants used surplus access as an inducement. (p. 187) | Expert interviews and case analyses (Funds A, C, E, G); interview statements on sweeteners and 50/50 trustee legislation. | Partial data |
| In first-wave transfers members received only their own contributions or actuarial liability and none of the surplus; second-wave members got actuarial liability plus a sweetener of 10–30% from surplus, leaving surplus behind. (p. 197) | Cross-case analysis of Funds A–E and G (transfer values), plus interview accounts and Andrew's statement. | Partial data |
| Stakeholder power shifted from unions and employers (1980s) toward the regulator, service providers and trustees by 2006. (p. 208) | Respondent ratings of relative power (Table 1, Graphs 1–4) combined with interview narratives. | Partial data |
| Members who converted and stayed are generally worse off, having taken on investment and interest-rate risk, while those who converted and withdrew are better off. (p. 210) | Interview respondents' views and case member perspectives; reference to the 1998 market correction; no quantitative outcome data. | Illustrative or anecdotal |
| Conversion decisions were made by company boards, not trustees, which contradicts the literature. (p. 224) | All eight case studies plus confirmation by expert interviews. | Partial data |
| South African actuaries' redefinition of the investment reserve as surplus enabled employers to access large surpluses on transfer, with some deliberately unethical manipulation. (p. 190) | Wellsted/Andrew/Murphy statements, Pepkor and Sentrachem cases, and interviews. | Partial data |
The author's caveats¶
The author acknowledges that qualitative samples are purposive and snowball rather than random, so results are generalisable to theory rather than populations. Researcher bias, small case numbers and confidentiality (respondents and company publications unlisted) are noted. Respondents sometimes contradicted each other, which the footnotes flag, and the author notes the sensitivity of the topic, though disclosure concerns proved unfounded.
Gaps in the evidence¶
The central claims about member outcomes, surplus size, transfer-value fairness and speed rest on recollection and perceptions from a small number of confidential interviewees and eight cases, with no systematic data. Fund-level actuarial valuations, transfer values, take-up rates and member benefit outcomes across many converted funds, plus FSB Section 14 certificates and registrar records, would test the waves and distributional claims. Such records exist at the FSB and in fund archives, and some became accessible under the surplus apportionment process. Archival union, employer-federation and consultancy records would test the contested attribution of agency to unions, employers and the 'emerging brokerage'.
Replicability: Data poorly documented.