On September 16, Alberta’s Minster of Finance proposed significant changes to the province’s public sector pensions. That same day, the Province launched a Public Sector Pension Sustainability website and released a policy document titled Charting a New Course: a Vision for Public Sector Pension Reform, both of which I encourage you to review.
Some of the key proposed changes include:
- Members whose age plus service equals 85 or more have so far been allowed to retire without a reduction to their pension anytime from age 55. For service after 2015, the pension will be actuarially reduced for those who retire before age 65, to reflect the true cost of paying the pension longer.
Guaranteed COLA, currently at 60 per cent of Alberta inflation, will be changed for service after 2015 to “target COLA” at 50 per cent of Alberta inflation.
The policy document lays out the Government’s justification for reform with a core tenet that the current pension framework is not sustainable:
Many defined benefit pension plans are facing significant funding deficits due to lower returns on investment, early retirements, the increasing ratio of pensioners to contributing members, and increased life expectancies. Many plans will not be sustainable in the long term if changes are not made to address these challenges. […] The combination of these factors has resulted in the plans being underfunded. Total contribution rates for employees and employers are 25 per cent of salaries or higher – among the highest in Canada.
As an active member of the Local Authorities Pension Plan (LAPP), I’ve taken it upon myself to review some of the Government’s claims and examine whether Alberta’s public sector pensions – and LAPP in particular – are sustainable. To be clear, the following are my personal views and do not necessarily reflect the views of LAPP or my employer.
Issue #1: Funding deficits exist due to lower returns on investment
The Government claims that lower and more volatile investment returns risk the sustainability of defined benefit pensions, and recent experience might indicate that this is indeed the case. In 2008, financial markets crashed around the world, several countries defaulted or nearly defaulted on their debts, and central bankers resorted to drastic and experimental monetary policy responses to avoid further catastrophes. For its part, LAPP’s investment return was a staggering -15.1 per cent in 2008, representing billions of dollars of losses on paper.
But LAPP has been in existence for over 50 years and is expected to continue operating for many generations to come. Unlike an individual investor, the investment horizon for a pension plan is many decades, and the long-term investment return is what really matters – one bad year should not be the sole focus. On a longer-term basis, LAPP has done relatively well:
As of December 31, 2012 the fund had a 4-year annualized return of 9.2 per cent, a 10-year annualized return of 7.6 per cent, and a 19-year annualized return of 7.3 per cent. Despite the negative events of 2008, these historical rates of return all compare very favourably against the plan’s long-term investment return target of 5.75% per annum. If anything, the Government should be communicating that investment returns have been higher than expected, not lower.
Although official performance numbers are not yet available for 2013, it has been a very good year for equities, especially from a Canadian investor’s perspective (a weakening Canadian dollar increases the return on non-Canadian assets). In fact, given its current asset mix, I wouldn’t be surprised if LAPP earned a rate of return in excess of 15 per cent in 2013, which could go a long way to improving its current funding deficit.
If investment returns have been relatively good, then why are Alberta’s public pension plans still in a funding deficit? One of the biggest factors relates to the way pension liabilities are calculated. Pension liabilities represent the benefits that will one day be paid out to retirees. Because most plan members will continue to work for several decades before retirement, calculating the size of their retirement benefits today is not an exact science. Periodically, an external actuary is hired to estimate the total dollars that will eventually be paid out to retirees. To come up with this estimate, many assumptions are made including how long retirees will live on average, what their salaries will be in the years leading up to their retirement, and what the rate of inflation will be over time. Once the future pension benefits are estimated, the actuary estimates a discount rate that is used to translate the future pension benefits into a single amount valued in today’s dollars.
The Impact of the Discount Rate Assumption
The discount rate allows the actuary to estimate the present value of a pension plan’s liabilities, an amount that can then be compared to the current value of the plan’s assets. If the assets exceed the estimated liabilities, the plan is fully funded; if however, the estimated liabilities exceed the assets, there is a funding deficit. In estimating the present value of its liabilities, the discount rate for LAPP is set equal to the expected long-term investment return on the fund assets – currently 5.75 per cent. Discounting future benefit payments with a higher discount rate leads to a lower estimate of the plan’s liabilities, while discounting future benefit payments with a lower discount rate leads to a higher estimate of the plan’s liabilities.
Over the past 20 years, the discount rate used to estimate LAPP’s pension liabilities has been steadily decreasing in response to decreasing bond yields and lower expected investment returns going forward:
The discount rate has declined from about 8.25 per cent in the early 1990s to 5.75 per cent today. Because employees start contributing to pension plans decades before they receive any benefits, estimating plan liabilities involves assumptions and calculations for many years into the future. The effects of compounding over many years of estimated cash flows, combined with the sheer size of a pension plan, mean that small changes in the discount rate are magnified and can lead to very large changes in the estimated liabilities. LAPP’s 2012 Actuarial Valuation Report – a document prepared by the plan’s actuary – provides context for the sensitivity of the plan’s liabilities to changes in the discount rate. As of December 31, 2012 a one per cent decrease in the discount rate would increase the estimated plan liabilities by $4.9 billion (up to $33.3 billion from $28.4 billion):
If a one per cent decrease in the discount rate adds an additional $4.9 billion to LAPP’s estimated liabilities today, it follows that the 2.5 per cent decrease in the discount rate experienced from 1990 to today has added billions of dollars to the estimated pension liabilities. But this increase in liabilities is not as a result of owing pensioners more money in the future (i.e. it is not as a result of higher wages or higher inflation), it is simply the result of using a progressively more conservative value for a key (and highly sensitive) assumption.
Assuming the actuary’s sensitivity analysis can be applied symmetrically to both decreases and increases in the discount rate, if the LAPP discount rate returned to the 8.25 per cent level used previously, the estimated plan liabilities would go down by $12.2 billion dollars ([8.25% – 5.75%] * $4.9 billion). In this scenario, the plan’s estimated liabilities would fall to just $16.2 billion, leading to a very healthy plan surplus of more than $6 billion ($23.0 billion assets – $16.2 billion adjusted estimated liabilities). I’m not arguing that LAPP should be using a discount rate of 8.25 per cent; the point is to highlight the sensitivity of the funded status calculation to the discount rate assumption. Given a 19-year annualized return of 7.3 per cent, there is however an argument to be made that the current discount rate of 5.75 per cent is overly conservative, and therefore artificially inflating LAPP’s estimated pension liabilities.
Issue #2: Early retirements, increasing ratio of pensioners to contributing members, and increased life expectancies
On the Public Sector Pension Sustainability website, the Government addresses a number of frequently asked questions. In response to the question, “Why does my plan have a deficit?,” the Government responds in part:
Many defined benefit pension plans are facing significant funding deficits due to lower returns on investment, early retirements, the increasing proportion of plan members who are retirees or deferred retirees compared with contributing members, and increased life expectancies. The current pension plans were designed decades ago, and much has changed. On average, today’s workers will retire three years earlier and live four years longer than those who retired a generation ago. Many will spend more years in retirement than they spent working.
Increased life expectancies
Unanticipated improvements in life expectancy over the past 50 years have certainly contributed to the current funding issues of Alberta’s public pension plans. The contribution rates required of members and their employers in the past were essentially calculated based on flawed assumptions. If the average pensioner lives for 25 years after retirement but their pension contributions were calculated based on the assumption that they would live for 15 years, the cost of the average pensioner’s pension benefit payments will be significantly higher than the contributions they made during their working years. Early LAPP plan members got a very good deal – paying in less than they will ultimately withdraw – and that has contributed to the current funding deficit of the plan.
But does increased life expectancy mean the public pension framework is unsustainable in its current form? Perhaps, but only if the life expectancy assumptions in place today are just as flawed as they used to be. To the contrary, current member contributions are calculated using assumptions that adjust for the improvements in life expectancies over the past half century. And they go a step further, also forecasting further improvement into the further, as described in LAPP’s 2011 Actuarial Valuation Report:
The rates of mortality assumed in a valuation serve two purposes: firstly, to determine what portion of the current membership will survive to retirement age, and secondly, to forecast the remaining lifetime of members once they reach retirement.
There is no reason to expect the mortality to differ from the 1994 Uninsured Pensioners mortality table. Furthermore, there is strong evidence of continuing improvement in mortality since 1994 and it has become an industry standard to assume this trend continues into the future. We have used the AA projection scale to allow for improvements in mortality since 1994 and indefinitely in the future.
To the extent that life expectancy assumptions now more accurately reflect the true cost of a pensioner’s future benefit payments, it is misleading to cite changes in life expectancy as a reason for proposing pension reform. The necessary changes have already been implemented by adjusting the pension plan assumptions.
Increasing ratio of pensioners to contributing members
The Government cites the increasing ratio of pensioners collecting benefits relative to active contributing members as another reason for Alberta’s public pensions being unsustainable. Like increasing life expectancy, this argument is only reasonable if the current pension assumptions are flawed. To the extent that today’s active members are contributing amounts that will on average pay for their future benefits (plus an amount to gradually reduce the current funding deficiency), the number of active or retired members is irrelevant. The purpose of a plan’s actuarial valuation is to establish appropriate assumptions and determine the required contributions today in order to sustain the plan into the future.
Again, early LAPP plan members got a very good deal – paying in less than they will ultimately withdraw – and that contributed to the current funding deficit of the plan. But going forward, the sustainability of Alberta’s public sector pension plans is ensured as active members now make contribution that more accurately reflect the true cost of their future pension benefit payments.
To date, plan members whose age plus years of service equals 85 or more have been allowed to retire without a reduction to their pension anytime from age 55. This subsidized early retirement is cited by the Government as another reason for Alberta’s unsustainable public sector pensions.
I have not had time to work through detailed scenarios to determine the full impact of the proposed changes in this specific area, but requiring members to work an additional ten years relative to the current pension promise in order to avoid a reduced pension seems like a very drastic one-time change.
For a member that would have previously reached their 85 factor at age 55, working until age 65 actually results in a 20-year swing: not only do they have to delay their receipt of pension benefit payments by ten years, they must also continue contributing into the pension plan for an additional ten years. Assuming average pension contributions of $8,000 per year, and an average pension benefit of $40,000 per year, the elimination of the 85 factor provision costs each member $480,000 ([$8,000 + $40,000] * 10 years).
Given that active plan members are already making higher contributions to the plan in order to eliminate the funding deficit that was created as a result of giving earlier plan members a very good deal, reducing future benefits for active members by forcing delayed retirement seems excessively unfair and unbalanced. I would also argue that it is contrary to one of the Government’s stated goals for pension reform:
There is intergenerational fairness for members and taxpayers.
The notion that the 85 factor provision equates to a subsidized early retirement is questionable. As part of the plan’s actuarial valuation, the current contribution rate already includes an amount to pay for the additional costs related to unreduced pensions for members that have reached the 85 factor (LAPP’s actuary estimated that the provision required contributions of 0.75 per cent of members’ salaries at the end of 2012).
Rather than describing this provision as an unfair subsidy that jeopardizes the sustainability of Alberta’s public sector pension plans, the Government should acknowledge that it is an additional benefit that is accounted for in the plan actuarial assumptions and that is being paid for by active members and their employers.
Issue #3: Total contribution rates are 25 per cent of salaries or higher
The following chart plots the combined contributions (as a per cent of salary) made by active LAPP members and their employers over the past 16 years (data prior to 1998 was requested from LAPP in September but as of the time of writing, the information had not been provided):
There are two series in the chart because active members are subject to two contribution rates depending on their annual earnings. Earnings up to the YMPE are subject to the lower contribution rate, while earnings above the YMPE are subject to the higher contribution rate. For context, the YMPE for 2013 was $51,100 and it typically increases several thousand dollars each year to keep pace with inflation. A typical LAPP member currently earns $70,000 per year and will be subject to a blended contribution rate of about 24 per cent of salary in 2014.
As the chart demonstrates, contribution rates have been rising steadily and have more than doubled since 1998. Increasing contributions necessitate larger deductions from active members’ paycheques (leading to lower disposable income) and increased payrolls for employers. Based on this single chart, one might argue that the current pension framework is indeed unsustainable: the trend of contribution rates is clearly up and the absolute value of contributions, at almost 25 per cent of salaries, could be reaching a tipping point where excessively high pension contribution rates become a deterrent, preventing talented, qualified employees from accepting jobs from public sector employers.
The elevated contribution rates of Alberta’s public sector pensions are cited by the Government as a key justification for reform. If contribution rates were expected to be 25 per cent (or higher) indefinitely, I would agree that the Government’s point is valid. But what isn’t being made clear in the Government’s claim is that the current contribution rates are temporarily elevated in order to systematically eliminate the plan’s funding deficit. By law, any funding deficit of an Alberta public sector pension plan must be eliminated within 15 years. As a result, the current LAPP average total contribution rate for 2014 is set at about 24 per cent, but almost one third of that amount is due to the special payments being made to offset the existing funding deficit:
Over time, as the funding deficit is eliminated – either through better than expected investment returns, increases to the discount rate assumption, or simply from the special payments that are being made each year – the contribution rate will decline. If there was no funding deficit at the beginning of 2014, the average LAPP total contribution rate would be just 17 per cent, which is significantly lower than the 25 per cent figure highlighted by the Government.
Costs versus benefits
While researching Alberta’s public service pension plans, spending hours reviewing annual reports, actuarial valuations, and the pension sustainability website, I’ve come to several conclusions:
- The assumptions used today to calculate the contributions required to fund future pension benefits are much more conservative than the assumptions in place a half-century ago. Today’s assumptions incorporate lower expected investment returns, longer life expectancies, smaller discount rates for estimating the present value of pension liabilities, and adjust for early retirement provisions.
Actuaries have learned a lot from their mistakes over the past 50 years, and I believe the assumptions underlying today’s pension plans accurately project the cost of providing future pension benefits.
With more accurate assumptions, today’s active plan members are contributing the true cost of their future pension benefits, making future funding deficits less likely.
The debate around Alberta’s public sector pensions boils down to a tradeoff between desired benefits and acceptable costs. The Government claims that public service pensions are too costly and that those costs are set to rise further, making pension plans unsustainable unless benefits are reduced. But the factors they blame for increasing costs have already been factored into plan assumptions, and contribution rates will actually decline as funding deficits are resolved.
The question should not be “Are Alberta’s public sector pensions sustainable?” – the answer to that is a resounding yes in my opinion – but rather “Are we comfortable paying the true cost of defined-benefit pensions in Alberta?” For LAPP, the long-term cost of maintaining the current plan provisions is about 17 per cent of member salaries, split between employees and employers. As an active plan member, I am comfortable with that cost given the benefits that are provided. If that is not a satisfactory cost/benefit dynamic for the Government, they should say so directly rather than manufacturing a pension crisis and attempting to justify reform based on inaccurate sustainability claims.
What do you think?
Now that you’ve reviewed my thoughts on some of the Government’s claims regarding the sustainability of Alberta’s public sector pensions, I encourage you to watch the following Government-produced video that summarizes their stance on pension reform. While watching, ask yourself if you still agree with their reasoning: