A new report by the C.D. Howe Institute is recommending the federal government amend legislation to allow Canadians to buy life annuities within a tax-free savings account.
It also suggested the government modifies the draft legislation creating advanced deferred life annuities and variable payment life annuities to make them both available within a TFSA.
“More than 45,000 employers offer group [registered retirement savings plans] to their employees, with an increasing number also giving their employees the option to invest in an alternative group TFSA,” said the report. “Many of these plans involve employer contributions on top of employees’. Someone having funds accumulated in a TFSA (group or not) wanting to buy an annuity to insure against the risk of living a long life would have to withdraw the funds and purchase an annuity contract in which the interest portion of the payouts is taxable.
“This tax on the interest portion reduces the benefit of the annuity contract and makes the alternative of leaving the funds within the TFSA to self-insure against the risks of longevity — negating the benefits of longevity pooling — more tax-effective.”
The report, which comes as the TFSA turns 10, used data published between 2009 and 2015 from the Canada Revenue Agency. “After a decade in existence, there is now enough data and empirical analysis on TFSA utilization to enable a quick assessment of the extent to which the product is reaching its policy objectives, and how its evolution compares to pre-existing registered retirement saving plans,” noted the report.
It found the fair-market value of all investments in TFSAs reached almost $233 billion by the end of 2016, about 20 per cent of all assets held in RRSPs, registered retirement income funds and locked-in retirement accounts.
The increase in assets reflects the strong growth in the number and share of tax-filers holding a TFSA, noted the report. Data showed the number of TFSA holders increased from 4.8 million in 2009 to 13.5 million in 2016, with annual growth slowing down as the program matured.
The popularity of TFSAs is evident when compared to RRSPs. In 2016, a total of $55 billion was contributed to TFSAs compared to $42 billion to RRSPs. Since 2013, annual TFSA contributions have exceeded those in RRSPs.
The report also found TFSA contributions per contributor increased much faster than RRSPs during the period. Among those aged 20 to 29 and 30 to 44, TFSA contributions per person increased at an average annual rate of 5.2 and 5.8 per cent, respectively, compared to 2.6 and 1.8 per cent for RRSPs. For older age groups, the growth of TFSA contributions was even higher, reaching 8.4 per cent per year on average for those aged 45 to 64, and 9.6 per cent for those aged 65 and older — again, more than double the rates of RRSP contribution increases.
Average annual TFSA contributions are sizeable and follow the same age pattern as RRSP contributions, said the report. For instance, in 2016, the average TFSA contribution for those aged 20 to 29 was $4,600, compared to more than $5,000 for those aged 30 to 44, more than $7,000 for those aged 45 to 59 and more than $9,000 for those aged 60 and older.
This suggests that at least a portion of TFSA accumulation is long-term saving, possibly for retirement purposes, said the report, which raises the question: to what extent are TFSA contributions displacing contributions that otherwise would be made to an RRSP?
The report highlighted a considerable overlap between TFSA and RRSP contributors — in 2016, 32 per cent of households with a TFSA also had an RRSP. But it also noted there’s considerably less overlap with respect to contributions than account holding, which suggests some displacement of RRSP contributions in favour of TFSAs.
Indeed, across different demographic groups, the data showed Canadians have diverted a significant portion of new savings from RRSPs and into TFSAs since their introduction in 2009 — every one per cent increase in TFSA contributions reduced RRSP contributions by 0.4 per cent.
In addition, the report noted TFSA savings may be more likely to be withdrawn early compared to RRSP savings. From 2009 to 2016, aggregate TFSA withdrawals among those under age 30 and those from ages 30 to 44 accounted for about half of aggregate contributions. In contrast, RRSP withdrawals accounted for only 13 and 22 per cent of contributions, respectively, among this younger age group.
Among people aged 45 to 59, the report found no material differences between TFSA and RRSP withdrawal rates, which account for about 30 per cent of contributions for both. “The fact that withdrawals rates are similar between TFSAs and RRSPs in this age group suggests that a sizeable portion of pre-retirement contributions and investment income accumulating yearly in TFSAs will be available to Canadians as they move into retirement,” said the report.
Among older age groups, RRSP withdrawals are predictably high, but TFSA withdrawals amount to only a quarter of contributions, the lowest rate of all age groups. However, with interest rates so low, the after-tax investment income on fixed income assets suitable for capital investment in retirement may not even cover the increase in the cost of living. For seniors, TFSAs are a tax-effective tool to decumulate their retirement capital and insure against longevity risk, noted the report.
In addition to suggesting the government allow TFSA contributors to buy annuities within the plan, the report also recommended the government make amendments to create a new tax-free pension account to encourage a greater number of younger and low- to mid-income workers to save for retirement on a tax-effective basis; permit a surviving spouse to use the unused TFSA contribution room of their deceased spouse, within limits; and resolve Canada-U.S. tax issues to improve the operation of TFSAs.