-A A +A

The Study in Brief

  • Simple changes to tax rules can improve retirement security for Canadians, as well as make the retirement system more equitable among different classes of savers, and more efficient at managing longevity risks for capital decumulation. This E-Brief provides a discussion of needed retirement-related tax changes impacting members of capital accumulation plans, such as RRSPs and defined-contribution (DC) plans, divided into the accumulation and decumulation phases.

  • Among the key tax changes recommended for the accumulation phase: increasing retirement savings room; for example, by discarding the annual income-based tax limits and replacing them with a uniform inflation-indexed lifetime accumulation limit; letting sponsors of group RRSPs deduct administrative expenses and payroll taxes like sponsors of defined-contribution plans; and creating a Tax-Free Pension Account (TFPA) that would primarily cater to the requirements of low- to mid-income earners.

  • For the decumulation phase, the authors’ major recommendations include: adding annuities to the list of investment products that can be held within a Tax Free Savings Account (TFSA); increasing the age thresholds for individuals to start receiving their public pensions; extending the age limits for when individuals must stop contributing to, or begin withdrawing from, their registered plans; equitable eligibility for the pension income tax credit and pension income splitting; and changes to mandatory minimum withdrawals from registered savings.

  •  Savers in capital accumulation plans need fairer tax rules and more options.

The authors thank Jeremy Kronick, Peter Bowen and members of C.D. Howe Institute’s Pension Policy Council, as well as members of the Fiscal and Tax Competitiveness Council.

Introduction and Overview

Canada’s retirement security system is composed of the incomes Canadians receive from publicly funded sources (Old Age Security and the Quebec/Canada Pension Plan) and privately funded sources (registered pension and savings plans, and unregistered savings). Most Canadians will draw on both sources to maintain their standard of living in retirement.

Workers accumulate private retirement wealth mostly through tax-deferred pension plans.

The rise of capital accumulation plans in the private sector
The vast majority of public-sector employees participate in a defined-benefit pension plan, which pays a life annuity upon retirement. However, in the private sector, defined-benefit pension coverage melted away by two-thirds in the last 30 years, from 26 percent of employees in 1989 to only 9 percent in 2019. At the same time, private-sector participation in capital accumulation plans (CAPs) increased at a rapid pace. These include defined-contribution plans, group Registered Retirement Savings Plans, and Deferred Profit-Sharing Plans, as well as hybrid plans that have been converted to defined-contribution with a legacy defined-benefit component, or shared-risk arrangements. In 1989, only about 4.5 percent of private-sector employees were covered by a CAP/hybrid, jumping to 28 percent in 2019 (Figure 1).

CAPs (and hybrids to a lesser extent) are fast becoming the preferred retirement wealth accumulation vehicle in the private sector. In addition to these tax-deferred registered plans, in which contributions are made out of tax-exempted income but withdrawals are taxed, workers since 2009 have had access to Tax Free Savings Accounts (TFSAs), in which contributions are made out of taxed income, but withdrawals are untaxed. TFSAs are becoming an important component of retirement planning, across the income spectrum (Laurin 2019).

The growing importance of capital accumulation plans as opposed to defined-benefit (DB) plans makes fresh policy attention to tax laws and regulations all the more pertinent for workers in their capital accumulation phase. But it also means that a growing number of new retirees and pre-retirement workers are planning and managing the decumulation phase of their retirement capital on their own. And demographic aging will compound the challenge. In the next ten years, the population older than 65 is set to increase by almost 2.5 million – which represents 50 percent of the overall projected population growth. We need tax-rule changes to improve the retirement security of a fast-growing number of CAP plan holders.

This E-Brief proposes retirement-related tax changes impacting members of capital accumulation plans, divided into the accumulation and decumulation phases of retirement planning.

With respect to the accumulation phase, we propose:

• more equitable tax-deferred registered wealth accumulation limits;

• changes to the tax recognition of administrative expenses in group RRSPs; and

• the creation of a new tax-prepaid option for long-term retirement capital accumulation.

With respect to the decumulation phase, we propose:

• measures required for two significant recent retirement saving products to gain widespread acceptance;

• adding annuities to the list of investment products that can be held within a TFSA;

• increasing the age to which individuals can defer public pensions;

• extending the age limits for when individuals must stop contributing to, or begin withdrawing from, their registered plans;

• changes to mandatory minimum withdrawals from registered savings; and

• more equitable eligibility for two tax breaks: the pension income tax credit and pension income splitting.

These simple changes to tax rules would improve retirement security, as well as make the retirement system more equitable among different classes of savers and more efficient in terms of managing decumulation risks such as longevity, price inflation, and market volatility.

The Accumulation Phase

Private-sector workers do most of their retirement saving through DC plans and RRSPs. Group RRSPs, in particular, are a growing source of retirement wealth (Baldwin 2022). In 2022, 52,549 employers supported such saving by organizing group RRSPs, with many matching at least part of their employees’ contributions. This is about five times the number of employers sponsoring DC plans (Benefits Canada 2022).The prevalence of larger employers offering DC plans and smaller employers offering group RRSPs results in a much higher number of plans for the latter but more similar number of members and assets under management.

Approximately 3.8 million Canadians participated in an employer-sponsored group RRSP in 2021, more than the 2.4 million private-sector participants in DC pension plans. In that year, members of group RRSPs held $136 billion in assets, slightly more than the $132 billion of private-sector assets held by DC plan members (Investor Economics 2021).

In addition, TFSAs have grown in popularity since their inception in 2009, and are clearly playing a role in retirement wealth accumulation (Laurin 2019). For low-income retirees, effective tax rates on their taxable retirement income are very high: the combined effect of income taxes and clawbacks of federal and provincial income-tested benefits, such as the federal GIS and associated provincial supplements, means that new retirees can expect effective tax rates greater than 50 percent on up to about $15,000 of pension income received in excess of average Q/CPP benefits (Laurin 2019). TFSA withdrawals, on the other hand, do not affect entitlement to income-tested benefits, which means that lower-income workers can use TFSAs for retirement income purposes without reducing their entitlements to government benefits.

It is crucial for private-sector workers to prioritize retirement savings, as they often struggle to save enough. To address this issue, we propose several tax-related policy changes intended to enhance the accessibility and adequacy of retirement savings. We are confident that these changes will greatly benefit the majority of private-sector workers.

Provide More Retirement Savings Room

Every year, mainly around tax filing season, there are reports that the stock of unused RRSP contribution room is staggeringly high, and growing. This is often used as evidence to overlook the need to increase contribution room. But participation is notably high among those with the greatest need to save in RRSPs to maintain their work-life standards of living in retirement, which belies what a simple look at unused room for all Canadians would suggest.

Considering that the primary goal of retirement savings is to provide an income stream in retirement sufficient to maintain pre-retirement living standards, lower-income earners appear to be well covered by public pension and other benefits. For them, tax-prepaid TFSAs appear to be a more tax-efficient vehicle to top up their retirement income from public sources. Other groups like students or young families may be in a stage of life when saving for a home makes more sense (Hamilton 2015). Those participating in well-run pension plans still accumulate RRSP room but may not need it as much.

Laurin (2014) found that RRSP participation is the highest among mid- to high-income earners not covered by a workplace pension plan. Nearly one-in-two with the greatest need to participate in an RRSP made a contribution in 2013, and they contributed more than 10 percent of earnings on average.

But many people may need more retirement savings room. Consider individuals in the latter stages of their career who have an inconsistent earning history. This could apply to recent immigrants, or mothers who only begin to earn higher wages later in life. Despite needing to save enough to sustain their current lifestyles in retirement, they may not have accumulated sufficient RRSP room to do so.

Furthermore, there are millions of Canadians without the benefit of career participation in defined-benefit pension plans. These individuals are unable to accumulate even half the amount of tax-deferred retirement wealth that some DB pension plan members enjoy, even if they desired to do so (Pierlot and Siddiqui 2011). The tax regulations include a formula to convert the maximum annual defined-benefit pension amount into an equivalent lump sum contribution.DB pension plan members can earn pension benefits no greater than 2 percent of pay per year, up to a yearly maximum of $3,506.67 in 2023. This is the basis for the yearly maximum contributions to an RRSP/DC plan. The DB maximums are converted into RRSP/DC contribution maximums using a conversion factor of 9. For example, 2024 maximum RRSP contributions equal 2 percent of pay (DB limit) x 9 = 18 percent of pay, up to a maximum of $3,506.67 x 9 = $31,560.However, this formula does not consider benefit enhancements, such as price-level indexation, nor does it reflect up-to-date assumptions regarding longevity and investment returns. Consequently, members of defined-benefit plans enjoy greater tax-deferred contribution room compared to members of capital accumulation plans.

The regulations governing transfers of retirement wealth from defined-benefit plans to capital accumulation plans provide a clear example of this disparity (Robson 2017).Commonly referred to as the “8517 limits” in relation to the relevant Income Tax Act regulations. Pantaleo (2019) provides an illustration where a friend, due to his employer’s dire financial situation, chose to receive the transfer value of his pension benefit as a lump-sum payment. The transfer value calculations are based on the full earned benefit and are up to date, unlike the maximum RRSP contributions explained earlier. As a result, Pantaleo’s friend was only able to transfer approximately 55 percent of the commuted value of his defined-benefit pension to a locked-in registered account on a tax-deferred basis.He faced substantially higher immediate income tax rates on the remaining amount than what he had paid in the past or would pay in the future. The reason: he surpassed the contribution room allowed for RRSPs. At the very least, the discrepancies between the accumulation room granted to defined-benefit plan members and that given to RRSP/DC contributors should be rectified.

Two options are available to increase retirement savings room. The first would be to discard the annual income-based tax limits and replace them with a uniform inflation-indexed lifetime accumulation limit calibrated to the maximum lifetime value of pensions now enjoyed by high-income members of public-sector DB plans (Pierlot 2008; Pierlot and Siddiqui 2011).Pierlot and Siddiqui (2011) propose a $2 million lifetime contribution limit. Alternatively, the second option would be to update the tax equivalency factor calculations (factor of nine) to reflect current longer life expectancy and lower yields on safe investments, such that the maximum annual contribution limit to RRSPs approaches the value of benefits accruing in a defined-benefit pension plan (Robson 2017).Robson (2017) calculates that updating the factor for current longevity and investment return conditions would require raising the factor from its current value of 9 to at least 15, with a corresponding increase in the annual RRSP contribution limit from 18 percent of income to 30 percent.

Increasing retirement savings room is not fiscally costly: tax concessions on the contributions will be paid back in the future with interest corresponding to the rate of returns on investments. In fact, the aggregate value of the deferred taxes on registered savings should be estimated annually, and included in the value of government assets such that the perceived annual fiscal cost of tax deferrals would be small (Robbins and Veall 2002, Robson and Laurin 2016).It is worth noting that Bill C-47, currently in the second reading stage at the time of writing, puts forth the requirement for financial institutions to annually report to the federal government the fair market value of clients’ RRSPs and RRIFs.

The true cost of tax deferrals hinges on tax arbitrage opportunities, which involve withdrawing funds at a lower effective tax rate than the rate charged on contributions. However, these opportunities rely on individual predictions about the future and may not come to fruition. Furthermore, even if some or most of these predictions do become a reality, what is financially relevant for the government is the collective sum of all of the individual outcomes. In terms of aggregate figures, the effective tax-deferral rate for RRSP contributions is actually comparable to the effective tax rate paid on withdrawals, as noted by Laurin (2019). As a result, the fiscal cost of expanding tax-deferred retirement savings opportunities is small on a net present-value basis.

Letting Sponsors of Group RRSPs Deduct Administrative Expenses and Payroll Taxes

DC pension plans and pooled registered pension plans (PRPPs) allow sponsoring employers to deduct some administrative expenses from external income and exempt payroll taxes on employer contributions. On the other hand, individuals enrolled in group RRSPs are responsible for covering administrative costs using their assets within the plan, and contributing employers must pay payroll taxes on their contributions. Fair treatment would necessitate enabling group RRSP sponsors and/or participants to deduct the same administrative expenses, while also exempting employer contributions to group RRSPs from payroll tax, as proposed by Robson (2010).Employer contributions to these plans are frequently locked in, much like pension plan contributions. However, to address concerns regarding potential early withdrawals of employer contributions by employees before retirement, the government could consider imposing an additional tax penalty on early withdrawals of employer contributions.

A New Tax-Free Pension Account (TFPA) to Complement the Use of TFSAs

To achieve optimal tax reduction over the course of one’s lifetime, it is advantageous to save for retirement in the tax-preferred form that is expected to produce the lowest total taxes. Depending on their income levels and various personal circumstances, it is evident that many low-income individuals would be better served by using TFSAs instead of RRSPs for their tax-assisted retirement savings, as highlighted by Laurin (2019).

Pierlot (2012) and Pierlot and Laurin (2012) proposed the establishment of a dedicated Tax-Free Pension Account (TFPA) that would complement the use of TFSAs. While TFPAs would be accessible to all individuals, they would primarily cater to the requirements of low- to mid-income earners.

Similar to the TFSA, a TFPA would enable tax-free accumulation and withdrawals, but, due to its pension nature, it would be wholly distinct from TFSAs. Existing defined-contribution plans could include a locked-in TFPA option, which would provide a tax-prepaid alternative for long-term retirement capital accumulation and receive the same creditor protection as pensions and RRSPs. Governments concerned about potential abuse could implement penalties for early redemption of funds to safeguard their pension-related rationale.Ottawa may also have concerns about low-income individuals who would be part of a TFPA potentially avoiding the 50 percent clawback rate on GIS in retirement. But we must consider that low-income savers who benefit from tax-prepaid savings are utilizing their after-tax income to save. If they have dependent children, their Marginal Effective Tax Rate (METR) during the saving phase is not significantly different from the rate they would encounter in retirement while receiving GIS income (Laurin and Dahir 2022). Even in the absence of dependent children, their METR is still likely to range from 25 to 35 percent. Although this is lower than the 50 percent clawback rate on the GIS, without the potential tax inducement provided by TFPAs, most low-income individuals would likely continue to not save anything for retirement at all.

In addition, the TFPA contribution limit would be integrated with that of RRSPs and registered pension plans (as originally proposed in Kesselman and Poschmann 2001).Laurin and Poschmann (2010) and Kesselman (2015) make a similar proposal for an integrated TFSA/RRSP limit. In particular, Kesselman (2015) proposed to grant additional TFSA room to individuals with unused RRSP room.Thus, there would be only one contribution limit for all tax-assisted retirement saving accounts (Laurin 2019). Unused tax-deferred contribution room would become available for TFPA contributions.In addition, perhaps for a limited period of time after the TFPA implementation, individuals wishing to cash out all or part of their RRSPs and invest the after-tax proceeds in a TFPA could be allowed to do so without losing the RRSP room on the withdrawals. While TFSAs differ in this regard, withdrawals from TFPAs would result in lost contribution room, similar to RRSPs and pension plans.

The Decumulation Phase

With the decline of private-sector defined-benefit (DB) pension plans, more and more retirees rely on accumulated savings in other types of deferred-income plans to provide for retirement. But no one can know how long they will live. Those whose principal source of retirement income is not guaranteed for life need to protect against outliving their savings, which may translate into excess precautionary savings and a lower retirement lifestyle than may have been possible.

Currently, retirees from CAP programs have essentially two options to utilize their funds to support income for life.

1) Payout Annuity. This is an annuity issued by an insurance company providing a guaranteed income for life. No other financial product available on the market today can provide a more secure guaranteed lifetime income. However, most retirees do not elect to purchase an annuity as they are reluctant to lock into an irreversible decision. Also, annuities purchased by individuals can be expensive since insurance companies must account for selection risk and for the cost of capital required to underwrite a promise to provide guaranteed monthly payments for life. Returns on “safe” assets have been low for a long time, resulting in low annuity discount rates and lower monthly payouts than might be expected from continued exposure post-retirement to riskier investments such as equities, which tend to produce higher returns.

2) RRIF/Life Income Fund (or Variable BenefitsVariable benefits are different from Variable Payment Life Annuities despite the similarity in name. Variable benefits follow essentially the same tax rules as a LIF however are paid from within a pension plan as opposed to transferring the pension assets to a LIF. from within a pension plan). RRSP and PRPP funds can be transferred to a RRIF and DC-plan funds can be transferred to a Life Income Fund. In a RRIF or LIF the plan member has control of the assets and determines which type of investment strategy is most suitable for them. Both RRIF and LIF accounts are subject to an age-related minimum annual withdrawal schedule and a LIF is subject to a maximum annual withdrawal schedule as well. Neither a RRIF nor a LIF provide any form of longevity pooling and therefore a plan member needs to estimate how long their retirement income stream will need to last. As it is not possible to know exactly how long a retirement will last or how financial needs will evolve, some members may “underspend,” leaving significant unused assets on death, or overspend and deplete their savings before death.

A number of changes to federal tax rules show promise to improve retirement planning and security for CAP members and help address the risks they face.


In recent years the federal government established two new products to help CAP plan members manage longevity risk in retirement.

The first of these products is the Variable Payment Life Annuity, or VPLA. These annuities can be offered by large DC plans and PRPPs and will allow members who elect this option to pool their longevity risk. A VPLA shares many characteristics of the payout annuity described above; however, the size of monthly payment amounts is not guaranteed: it can rise or fall with gains (better investment returns and higher mortality than actuarial assumptions used to determine the initial payment amount) or losses (worse investment returns and lower mortality). While the lack of guarantee may seem like a disadvantage, it also means that the cost of purchasing a monthly income stream for life can be lower. The product does not need to include the cost of guaranteeing longevity protection or investment returns, and members can elect (if available) a VPLA that invests in a broader spectrum of assets than risk-free investment bonds, potentially benefiting from exposure to higher-yielding assets. The level of investment risk that a member is subjected to in a VPLA can be essentially similar to the investment risk exposure in a RRIF or LIF. The crucial difference rests in the addition of longevity pooling.

Currently CAP members at age 65 may have an expected future lifetime of almost 21 years, but a 1-in-5 chance of living for almost 28 years.2018-2020 Life Tables for both sexes. Outside of a VPLA, if they want to increase their likelihood of not running out of money they may choose to plan for income over 28 years to age 93, or perhaps even to age 96 when they have a 1-in-10 chance of still being alive.

Inside a VPLA, with longevity pooling, they combine their assets with that of other plan members such that their income stream can be based on an average lifespan of 21 years. This will allow the members’ annual income to be 20 to 30 percent higher than it would have otherwise been outside the VPLA and will not run out if members live beyond their life expectancy.

However, if members pass away prior to reaching their average life expectancy, their remaining assets will not be available to their estates, but rather distributed among other members of the pool who outlive them, thus creating a tradeoff. A VPLA can be constructed with certain guarantees such as members may receive at least 10 or 15 years of payments even if they die sooner than expected to help reduce the risk of the estate receiving no benefit. If there is a guaranteed number of payments, the annual payments will be lower.

The VPLA holds tremendous promise in assisting the mounting population of retirees in CAPs to navigate longevity risk at an affordable price.

Presently, however, the efficacy of the VPLA as a large-scale decumulation solution is considerably curtailed by existing tax legislation. Current tax rules enable the establishment of the VPLA exclusively within specific types of pension plans. As VPLAs require significant scale to viably distribute the risks of longevity and investment, it follows that only participants of the most substantial DC pension plans (current PRPPs are too small to achieve this level of participants) will be in a practical position to avail themselves of such an arrangement.

The current rules also make it difficult for members from group-RRSPs or other non-pension retirement savings to benefit from a VPLA option. In 2019, about 3.8 million people took part in a group-RRSP provided by their employer.

Recently, the federal government introduced the Budget Implementation Act, 2023, No. 1 (Bill C-47). This Act, if passed, will create a framework that will allow members of group RPPs, group RRSPs, and other persons with tax-deferred retirement savings accounts, to transfer their funds to a PRPP, which can in turn offer a VPLA.The BIA does not use the term VPLA within PRPPs, but rather defines variable life payments under a PRPP. It is possible that this could provide a framework for a more broadly available VPLA-like product.

Potential Premiums for an ALDA Purchased at 65 Years Old

It is important to recognize that the PRPP has not taken off as anticipated in the market, so creating a framework for individual access to a VPLA via a PRPP is not certain to be a success. It will be important that the regulations that come forward supporting the proposed legislation to allow individual access to VPLAs through the indirect use of PRPPs leads to an economically viable market solution.

The mandated use of PRPPs as an intermediary vehicle adds a layer of regulatory burden to a product that is new and already complicated on its own. In order to be economically feasible for market availability, the new regulatory framework should:

• avoid overly prescriptive and burdensome administrative requirements;

• allow for pooling of members across different pension jurisdictions (to help achieve critical mass); and

• allow for a simplified member registration process to transfer funds to a PRPP to elect a VPLA.

Additionally, this is likely to be a competitive market and, as such, the market should determine the appropriate fee level rather than having it constrained by regulatory minimum or maximum fees that may not align with the economics of making such a product available.


In addition to VPLAs, Ottawa recently created a new product called an Advanced Life Deferred Annuity (ALDA). An ALDA can be purchased from an insurance company via funds transferred from registered plansRegistered plans include registered retirement savings plan (RRSP), registered retirement income fund (RRIF), deferred profit sharing plan (DPSP), pooled registered pension plan (PRPP) and money purchase registered pension plan. and will allow a plan member to utilize a portion of their retirement savings – up to 25 percent (max. $160,000) – to purchase an income stream that would start at an advanced age, but no later than 85, and then use the remainder of their retirement savings to manage their income up to that age.

Similar to a payout annuity, an ALDA would be an insurance contract providing a guarantee from the insurance company. The investments backing the ALDA would likely be conservative fixed income investments.

The cost of an ALDA is therefore highly sensitive to long-term interest rates. Take the example of a purchase at 65 years old for payout at 85. Payments do not start for 20 years and then may continue for another 10-20 years. When tax rules were amended to permit an ALDA, interest rates were at all-time lowsFrom January 2019 to December 2021, yields on Government of Canada’s long-term bonds averaged only 1.5 percent – or negative 0.5 percent in real terms (adjusted for expected consumer price inflation). and product demand was not there. As interest rates have increased throughout 2022 it will be interesting to see if demand increases and how insurance providers will develop the product.

The table shows the cost of $1,000 per month ($12,000 per year), starting at age 85, for a 65 year old. Note that these prices do not include any return of premium should the annuitant die before age 85. The table demonstrates the high sensitivity of ALDA premiums to changes in interest rates. Note that actual premiums may vary and reflect potential anti-selection, as healthier individuals will be more likely to purchase such policies.

Even with higher interest rates, it is still unclear whether the product will be successful.Today’s higher interest rate environment at least currently reflects higher expected inflation. To the extent potential buyers care about real incomes, inflation expectations may dampen demand. The product still requires customers to relinquish capital with no assurance they will live long enough to collect the deferred annuity – although, for the cost of a lower annuity amount, tax rules permit a guarantee of return of capital in the case of death before annuity commences. A similar product has been available in the United States The US product is called a Qualifying Longevity Annuity Contract, or QLAC. for a number of years with limited success.

Deferred OAS or CPP

The longevity insurance benefits of an ALDA can be somewhat replicated by deferring receipt of guaranteed lifetime income from public sources; that is, electing to commence OAS later than age 65, and CPP/QPP later than age 60 (Genest-Grégoire et al. 2018). This option provides increasingly higher payments for each month the start date of payment is delayed. In the case of C/QPP, the adjustment factors are 7.2 percent per year before age 65 and 8.4 percent per year thereafter up to age 70. In the case of the OAS, the adjustment factor is lower at 7.2 percent per year from age 65 to 70. These adjustment factors provide a significant opportunity for many retirees to increase their total retirement income as well as to reduce their longevity risk. Currently, benefits can only be deferred to age 70 under both CPP and OAS (Quebec’s 2023 budget recently introduced legislation to allow QPP be to be deferred to age 72 starting in 2024).

Governments should allow Canadians to defer OAS and/or Q/CPP to older ages such as 75 or 80 to allow for a more substantial deferred income guarantee. If the boost in benefits is done at an actuarial-equivalent rate this should not hurt the financial solvency of these programs. Governments should also evaluate the desirability of a benefit increase rate that exceeds its actuarial equivalent, particularly in light of current labour shortages. Regardless of the extent of the boost, the increase in public pension benefits could serve as an encouragement for certain members who are in good health, and possess sufficient income and retirement savings to independently support themselves, to opt for longevity protection through this means.

Enable the Purchase of Life Annuities within Tax Free Savings Accounts

Due to liquidity regulations, it’s not possible to purchase life annuities within a TFSA. As an alternative to group RRSPs, many employers now offer group TFSAs to their employees. However, if someone invested in a TFSA wishes to purchase an annuity to hedge against the risk of living longer than expected, they would need to withdraw the funds and buy an annuity contract in which the interest portion of the payouts is taxable. This extra tax burden – i.e., the difference between paying taxes on interest income earned in an annuity vs no tax on such income earned in a TFSA – would possibly make it more tax-efficient to leave the funds within the TFSA and self-insure against longevity risks, which would eliminate the benefits of longevity pooling, as noted by Laurin (2019).

This cost distortion is unwelcome. It is unfair to those who opt to save for retirement in a group TFSA as opposed to an RRSP, in which the interest income portion of the annuity is tax-deferred until paid out. The federal government should amend the rules to make it possible to buy life annuities within a TFSA, including VPLAs once a standalone option is permitted, and ALDAs.

Extending the Age Limits for Retirement Saving and Decumulation

Life expectancy in Canada has been rising more than two years per decade since the 1960s, but current age limits related to retirement do not reflect this. Canadians (and their employers) now must stop contributing to tax-deferred retirement saving plans on December 31 of the year they turn 71 years old, which is also the threshold at which contributors must convert their retirement savings into a Registered Retirement Income Fund (RRIF) or an annuity, and begin drawing down their wealth the following year. Increasing the age from 71 to, say, age 75 may encourage older Canadians – who can – to stay in the workforce longer (Quebec 2013, Godbout 2022, Robson and Laurin 2023).

Modernizing Mandatory Minimum Withdrawal Schedule

Retirees invested in RRIFs also face a material risk of outliving their tax-deferred savings due to the mandatory minimum withdrawal schedule, notwithstanding improvements introduced in 2015 (Robson and Laurin 2023).

RRIF minimal withdrawal rates should be adjusted to reflect continued increases in longevity as well as recognize changes in investment returns due to a lower interest rate environment. Given the complexity of periodic or automated adjustments, Robson and Laurin (2023) recommend either introducing a basic withdrawal threshold amount below which minimums would not be required, or eliminating RRIF minimums entirely.

In addition, to eliminate the difference between having funds in a RRIF or an RRSP, RRIF minimums should be eliminated prior to age 71.

Equitable Eligibility for Tax Breaks

Pension benefits, lifetime annuity income from a registered savings plan, variable benefits from a DC plan, and RRIF withdrawals are eligible for tax breaks: the pension income tax credit and pension income splitting. Taxpayers aged 65 and over are eligible, unless they are pensioners from DB plans for whom an age exception is made: they can claim the tax breaks from age 55 or younger.It is generally not possible to start receiving a pension from a Registered Pension Plan (RPP) before the age of 55, unless an exception under the terms of the RPP is provided to allow for earlier access to pension benefits.

This age exception is unfair, and exacerbates the growing divide between public-sector workers, almost all of whom receive DB pensions, and those who do not. As long as the eligible retirement income is structured as lifetime periodic payments, either fixed or variable subject to maximums, it should not matter whether they are sourced from DB plans, registered savings plans, or variable pension benefits within a DC plan.

Fairness requires that all should have the same eligibility age.


A significant and growing number of current and future retirees will have limited access to a reliable pension and will instead depend on tax-deferred and tax-paid savings in personal and group plans to sustain their retirement. However, when it comes time to draw down their savings from capital accumulation plans (CAPs), very few retirees opt for a payout annuity. We propose several changes to federal tax rules that would expand the availability of new options for managing longevity and investment risks in retirement, such as Variable Payment Life Annuities (VPLAs) and Advanced Life Deferred Annuities (ALDAs). Additionally, we suggest liberalizing existing tax rules that govern the decumulation of tax-deferred retirement funds.

In the interest of tax fairness, we also recommend that people in CAPs be granted the same tax deferral opportunities during the accumulation phase as members of defined-benefit pension plans, as well as access to the same tax breaks during decumulation. These changes to tax rules would not only enhance retirement security but also promote greater equity among different classes of savers within the retirement system.


Baldwin, Bob. 2022. “The Evolving Wealth of Canadians: Who Is Better Fixed for Retirement? Who is Not?” E-Brief 327. Toronto: C.D. Howe Institute. May.

Benefits Canada. 2022. “2022 CAP Suppliers Report.” Available online: https://www.benefitscanada.com/archives_/benefits-canada-archive/2022-cap-suppliers-report-whats-on-cap-sponsors-wish-lists-for-the-coming-year/

Genest-Grégoire, Antoine, Luc Godbout, René Beaudry, and Bernard Morency. 2018. “Deferring Receipt of Public Pension Benefits: A Tool for Flexibility.” E-Brief 278. Toronto: C.D. Howe Institute. June.

Godbout, Luc. 2022. “Why Not 75 Years Old?” Intelligence Memo. Toronto: C.D. Howe Institute. February 14.

Hamilton, Malcolm. 2015. Do Canadians Save Too Little? Commentary 428. Toronto: C.D. Howe Institute. June.

Investor Economics. 2021. Group Retirement Savings and Pensions Report, 2021. ISS Market Intelligence.

Kesselman, J. Rhys, and Finn Poschmann. 2001. “Expanding the Recognition of Personal Savings in the Canadian Tax System.” Canadian Tax Journal 49 (1): 40-101.

Kesselman, J. Rhys. 2015. “Tax-Free Savings Accounts: Expanding, Restricting, or Refining?” Canadian Tax Journal 63 (4): 905-945.

Laurin, Alexandre, and Finn Poschmann. 2010. “Saver’s Choice: Comparing the Marginal Effective Tax Burdens on RRSPs and TFSAs.” E-Brief 91. Toronto: C.D. Howe Institute. January.

Laurin, Alexandre. 2014. “The Overlooked Option for Boosting Retirement Savings: Higher Limits for RRSPs.” E-Brief 183. Toronto: C.D. Howe Institute. September.

Laurin, Alexandre. 2019. “TFSAs: Time for a Tune-Up.” E-Brief 298. Toronto: C.D. Howe Institute. December.

Laurin, Alexandre and Nicholas Dahir. 2022. Softening the Bite: The Impact of Benefit Clawbacks on Low-Income Families and How to Reduce It. Commentary 632. Toronto: C.D. Howe Institute. November.

Office of the Superintendent of Financial Institutions Canada (OSFI). 2022. “Registered Pension Plans (RPP) and Other Types of Savings Plans – Coverage in Canada.” Ottawa: Office of the Chief Actuary.

Pantaleo, Nick. 2019. “It’s Time to End Over-Taxation of Lump Sum Pension Payouts.” Intelligence Memo. Toronto: C.D. Howe Institute. March 19.

Pierlot, James. 2008. A Pension in Every Pot: Better Pensions for More Canadians. Commentary 275. Toronto: C.D. Howe Institute. November.

Pierlot, James. 2012. “Could tax-free plans solve the pension puzzle?” Toronto: Benefits Canada. February.

Pierlot, James, and Alexandre Laurin. 2012. Pooled Registered Pension Plans: Pension Saviour – or a New Tax on the Poor? Commentary 359. Toronto: C.D. Howe Institute. August.

Pierlot, James, with Faisal Siddiqi. 2011. Legal for Life: Why Canadians Need a Lifetime Retirement Saving Limit. Commentary 336. Toronto: C.D. Howe Institute. October.

Quebec. 2013. Innover pour pérenniser le système de retraite - Un contrat social pour renforcer la sécurité financière de tous les travaileurs québécois / Comité d’experts sur l’avenir du système de retraite Québécois. Régie des rentes du Québec.

Robbins, Jenna, and Michael Veall. 2002. Registered Saving Plan Holdings as a Government Asset. Backgrounder 63. Toronto: C.D. Howe Institute. October.

Robson, William B.P. 2010. “Cutting through Pension Complexity: Easy Steps Forward for the 2010 Federal Budget.” Backgrounder. Toronto: C.D. Howe Institute.

Robson, William B.P. 2017. Rethinking Limits on Tax-Deferred Retirement Savings in Canada. Commentary 495. Toronto: C.D. Howe Institute. November.

Robson, William B.P. and Alexandre Laurin. 2016. Where the Bucks Stop: A Shadow Federal Budget for 2016. Commentary 447. Toronto: C.D. Howe Institute. March.

______________. 2023. Live Long and Prosper? Mandatory RRIF Drawdowns Raise the Risk of Outliving Tax-Deferred Saving. Commentary 641. Toronto: C.D. Howe Institute. April.

This E-Brief is a publication of the C.D. Howe Institute.

Alexandre Laurin is a Director of Research at the C.D. Howe Institute.

George Turpie (FCIA) (FSA) is Senior Vice-President, Group Retirement Savings & Investments, Canada Life. He contributed to this E-Brief in a personal capacity and his views don’t necessarily reflect those of the company.

This E-Brief is available at www.cdhowe.org.

Permission is granted to reprint this text if the content is not altered and proper attribution is provided.

The views expressed here are those of authors. The C.D. Howe Institute does not take corporate positions on policy matters.

Alexandre Laurin

Alexandre is the Director of Research and leads the fiscal policy program and the pension policy program at the C.D. Howe Institute. He joined the C.D. Howe Institute in 2008 and became Director of Research in 2014. From 1999 to 2008, Mr.

George Turpie (FCIA) (FSA) is Senior Vice-President, Group Retirement Savings & Investments, Canada Life.