A Pension Fund Problem

On July 13, 2020, Decree-Law 21/2020 was issued regarding reforming several pension funds laws and regulations. These were specifically related to 4 of the 10 recommendations made regarding the Social Insurance Organization (SIO), and were highlighted as urgently needed to prevent the funds from becoming insolvent.

The issue with the pensions funds is not new, nor can it be attributed to any single cause. The current debate (as I see it on Twitter, at least) seems to have coalesced into two main points of view: one group claiming that the looming insolvency is purely due to demographics and generous benefits to the public, and the other accusing the government and public officials of mismanaging assets, paying inflated bonuses and perhaps even engaging in corrupt practices. I seem to find myself somewhere in the middle. I think that a complete analysis would require a broader context regarding perceptions of how government operates, where priorities lie etc. Here, however, we’ll take a narrower approach, only looking at the proposed reforms and what they mean in terms of cost-saving, equity and sustainability.

The Problems

First, to understand the issue, we need to consider how the pension system actually functions. As an employee, you and your employer contribute a fixed percentage of your salary towards your pension (24% in the public sector, 18% in the private sector, not including unemployment deductions). Unlike with regular investments, these do not get saved away in a personal account until you retire, which you then draw from. Instead, the contributions the pension fund receives from employees today are pooled together and used to pay retiree benefits today. If there is a surplus, this is added to the funds’s assets and invested for a return to grow the pool. If there is a deficit, this must be taken from the fund’s assets.

Over the past several decades, the fund consistently saw a surplus in pension contributions vs. benefits. The number of retirees was relatively small compared to the number of workers, so inflows exceeded outflows, and the SIO was able to grow its assets. However, by 2016, this had changed. The labor market was fairly weak, so employment and wages hadn’t seen much growth to support the growing numbers of retirees, who were retiring fairly early (the average age of retirement is 49) and living longer than ever-before. In 2010, there 5.4 Bahraini employees for every retiree; by 2018, that number was only 2.6. Benefit outflows were now exceeding inflows from contributions. As our population continues to age and retire, we will see larger deficits, which will eat away at the BD 3.8 billion the fund had now amassed, until the the fund becomes insolvent.

Additionally, the fund’s investments were generating relatively low returns. Between 2008–18, the SIO’s investments generated an average annual return of 3.1%. For comparison, in 2018, Abu Dhabi Investment Authority’s 20-year annualized return was 5.4% — if we achieved similar results, the fund would have netted an additional BD 75 million per year. This is a substantial sum of money and would have given us an extra couple of years (2–3, by my estimate), although by no means could have solved the root problem.

On top of all of that, as part of the Fiscal Balance Program, the one-time Voluntary Retirement Scheme was launched in 2018. This resulted in approximately 10,000 public sector employees retiring with generous packages and benefits, resulting in a double-shock of both cutting inflows while increasing outflows. Between Q3 2018 — Q2 2019, the number of pension beneficiaries grew by 14%, which may have added almost BD 80-90 million in pension benefits paid out every year, accelerating the depletion of the pension fund.

One last note here — when I say SIO or “pension fund”, this is actually comprised of three funds: the public sector fund, private sector fund, and military. Each caters to its own employees and retirees. All in all, the actuarial expert, which I believe is the consulting firm Aon Hewitt, has determined that the public sector fund will go run out of funds by 2024 (even though as recently as December 2019 this was reported to be 2028) and the private sector fund will also become insolvent by 2033.

The Reforms

The initial four reforms were expected to increase the life of the pension fund by six years to 2030:

1. Merging the government employees’ pension fund and the social insurance (i.e. private sector) fund, creating a combined “Pension and Social Insurance Fund” managed by the SIO. As mentioned earlier, the public sector fund was expected to be depleted much earlier than the private sector fund. This would allow the assets to be pooled together to essentially allow for cross-subsidization between sectors, so that private sector contributions can be used to finance public sector retirees, bringing the remaining life of the combined fund to 2028. I can’t see any other major reasoning here, although economies of scale in terms of investing and returns is theoretically possible.
2. Suspension of the annual (3%) increase for all pensions prescribed under any law, pension or insurance scheme. In the event of an annual surplus (which I believe is highly unlikely anytime in the foreseeable future), this can be brought back. This seems to have attracted the most attention, because it will likely have the most visible effects. Inflation in Bahrain was 2.1% between 2014–18, so without any corresponding increases, the real value of pensioners’ wages will fall by 11% in 5 years, and 23% in 10 years. Some language was included to suggest that lower-income Bahrainis could be accommodated for, but to-date, the suspension applies to all. This would save BD 20 million in its first year, reaching over BD 200 million in year 10, adding two years of life to the fund (thereby taking it to 2030). It’s worth noting that the savings are largely from higher-income earners — 24% of pensioners receive BD 1,000+ and would make up 50% of the savings from this policy, compared to a similar number (25%) of pensioners who receive less than BD 400, yet only make up 10% of the savings.
3. Retirees cannot receive more than one pension under different retirement and insurance systems (except those entitled to pensions because of disability, work accidents or kinship). It is/was apparently possible to work a number of jobs, paying different contributions and ultimately collecting multiple pensions.
4. Pensioners who are also still working can opt to combine their previous service period with their new service period. Alternatively, they can continue receiving their pensions for their previous work period, as long as the current employment does not pay into the pension fund and would not be entitled to pension benefits. Items 3 & 4 seem to be more of corrections to existing irregularities, although could be bigger issues than I’m aware of.

There remain 6 reforms under discussion, which are on a collective take-it-or-leave-it basis and would prolong the life of the pension funds to 2086. These include much more tangible changes, from setting the minimum retirement age at 55 and imposing penalties for early retirement, to modifying the basis for the benefit calculation and gradually increasing employee contributions by 1% annually. One notable absence, in my opinion, is any consideration of progressive limitations or caps on benefits. As we’ve already seen, the costs and potential savings are disproportionately related to higher income-earners. Blanket policies distribute the burden equally across all income groups, and can especially hurt the poorest the hardest. Time will tell how things will turn out.