NITTY - GRITTY OF THE RRSP MODEL
Americans - think IRA or 401k when you see RRSP, and Roth when you see TFSA
Why you should read this page.
Below is a detailed argument, supported by math, proving that what you probable understanding about 'how the RRSP system works' ... is wrong. It proves that RRSPs have five effects that must be evaluated and calculated separately. No one before has bothered to calculate the actual benefits of an RRSP over investment in a taxable account, or developed the math for calculating the cost/benefit of each effect.
|1) ||The tax reduction resulting from claiming the RRSP contribution as a tax deduction (called here "the contribution tax credit") is not a benefit. It is a loan from the government - a loan you must pay back along with all the income earned by it. You should not consider the whole account to be 'your own' money. The part funded by your after-tax savings is 'yours', but the part funded by the contribution tax credit is 'the government's'. All the profits earned by the tax credit belong to the government, not you. This is no different from your best friend Bob giving you his money to add to your own, for you to invest for him. If his money makes up 40% of the account's value at the start, he continues to own 40% of the account, and must be repaid 40% of the account's value at its closure. The taxes paid on RRSP withdrawals are an allocation of principal between its two owners, not a 'deferred tax on the profits earned'. |
|2) ||All profits earned in the plan are sheltered from tax, permanently. You benefit only from the sheltering of your after-tax savings - the portion of the account that is 'your own' - not the government's loan. This benefit exactly equals the benefit from a TFSA. |
|3) ||One's marginal tax rate when withdrawing cash may be higher (or lower) than the rate at which one claimed the original contribution credit. This creates a penalty (or bonus) equal to the amount withdrawn multiplied by the change in rates. There is no generality that can be made beyond ... Those who contribute at the top marginal tax rate have a pretty good chance of taking money out at lower rates and receive a benefit. Those who contribute at the bottom tax bracket, have a chance of taking money out at higher rates and pay a penalty. |
|4) ||Canada has a variety of programs available to retired people whose benefits decrease as one's income increases. By deferring the income until retirement, the additional taxable income created by RRIF withdrawals at that time may reduce those benefits. It is a subjective decision whether to consider this disqualification for benefits 'caused' by the RRSP or by your good fortune. Conversely, RRSP contributions may help qualify younger people for benefits like the GST rebate. |
|5) ||Claiming the contribution tax credit may be deferred until a later year, but there is a penalty that grows with the length of the delay. The penalty roughly equals the future value of the income the tax credit would have earned during the delay. The benefit from delaying (to claim at a higher tax bracket) may be more than wiped out in just a few years by this cost. |
|1) ||The contribution tax credit is a benefit. So ... |
a) The person contributing at the 40% tax bracket gets a larger benefit then the person contributing at the 20% tax bracket.
b) The tax credit can be compared to benefits from other programs, like the RESP's 20% matching contribution.
c) You can multiply your benefits by using the tax refund to (eg) make a donation that generates a deduction for charitable donations, or contribute to a TFSA, or paydown debt.
|2) || The RRSP's benefit comes from the deferral of taxes. You get to keep the income earned by the contribution tax credit during the interim. |
|3) ||Income earned inside the RRSP is taxed on withdrawal at full tax rates. So ...|
a) Dividend income that would be taxed at lower rates in a taxable account is wasted within an RRSP. And the dividend tax credit is lost.
b) Income already taxed at top rates in a taxable account (like bond interest) should be prioritized in an RRSP.
|4) ||Your investments in an RRSP grow on a tax-deferred basis - not permanently tax sheltered. So ... |
a) When choosing between contributing to an RRSP vs a TFSA, the TFSA (where profits are tax-free) should be the default choice.
b) The only important RRSP benefit comes from maximizing the bonus from withdrawals at lower tax rates.
|5) ||Most people's marginal tax rate in retirement will be lower than when they were working. |
|6) || You should delay claiming the contribution tax credit when you expect your marginal tax rate is higher in the future. There is no cost to the delay. |
The RRSP's $benefit can be calculated as the difference between the outcomes using RRSP and taxable accounts. That total benefit can be broken down into its three sources - the tax free growth, the cost/benefit of a change in tax rates, and the cost of delayed claiming the contribution. The Challenge For All Detractors spreadsheet calculates the $benefit and shows its breakdown by source as argued here.
For all people wedded to a wrong understanding, you are challenged to develop your own math to explain the RRSP benefit in the way YOU think is correct. If you think you are correct, prove it. All the industry players have been challenged to disprove this spreadsheet and provide their own math. This list includes the big banks, the Investor Education Fund, the Get Smart About Money website, the Financial Consumer Agency of Canada, Investopedia, the Competition Bureau of Canada, the issuers of the CFP designation, the CAs, the Canadian Bankers Association, and Ryerson University. None have attempted the challenge. They simply dismiss and stonewall.
There are also benefits/problems of an RRSP not covered on this page. Each procedural rule could be considered a benefit or cost. See the list of attributes discussed relative to the TFSA on the RRSP Decisions page.
What are the mechanics of the RRSP system ?
Most Canadian know the mechanics of the RRSP system. This webpage does not dispute them.
- Contributions to the plan create a tax deduction. The resulting reduction in tax is commonly called the contribution tax credit. Your marginal tax rate on contribution is a variable in the model. An arbitrary $1,000 contribution is used.
- Inside the plan, the contribution is invested and grows tax-free. The rate of growth is a variable.
- The government hopes the plan is left intact until drawn down gradually in retirement. That reality is too difficult to model, but there is an easier understanding. Consider that contributions are made each year of a working life - 35 years from age 30 to 65. Each year's contribution grows over a 35 year period and then funds one year's retirement income - 35 years from age 65 to 100. The model shows a default term of 35 years, but it is a variable because the fund can be collapsed at any time.
- When withdrawn, tax is levied on the total draw. Your estimated tax rate is a variable.
Some people are stuck in the 'mechanics' of the RRSP and believe the mechanics prove the 'benefits'. They are likely to state "Income earned inside the RRSP is only tax-deferred, not tax-free, because the total balance of the account is taxed on withdrawal.". This POV is wrong because it isolates only one part of the RRSP system and ignores the other parts.
The money you save must do a full round-trip into and out of the RRSP before you can use it. The mechanics of both the contribution and withdrawal come into play. The individual steps don't prove the net result. You cannot pick just one step and conclude it proves a benefit (or cost). It is the net benefit that counts.
The model developed below is interactive on The Model spreadsheet. Open it now and play with the variables as they are added below. This is a conceptual model of what is happening inside an RRSP, to show where and how its benefits are generated - what is going on and why. Models should always be structured to simplify as many factors as possible, even while reflecting reality. It does not try to be a worksheet for personal decisions.
So, to start. In The Model cash that creates an RRSP contribution is considered to be a combination of after-tax savings plus the tax credit. It makes no difference to the conceptual model if, in reality, you spend your refund, invest it elsewhere, or never receive it. Your contribution creates the tax effect, regardless.
People dismiss The Model because they think it does not reflect what they personally do with their cash flows. A common argument is: "I don't put the contribution tax credit back into the RRSP. I do ... something else ... with it. So The Model is wrong."
- When your wages and spending remain the same, you save more when your taxes are lower. The RRSP contribution will lower your taxes. Therefore you save more when contributing to an RRSP.
- The benefits of an RRSP accrue only for the $$ inside the account. This Model considers only $$ inside an account. When analyzing the benefits of using a specific type of account you must presume that the account IS used. It is nonsense to presume the account is only partially used, e.g. to presume that only after-tax savings go into an RRSP because some people spend the refund. With that logic you might as well presume that $0 goes into the RRSP because people over the age of 71 will not make any contributions.
- The division between the two columns within The Model is conceptual. It makes no presumptions about where the contribution cash comes from - much less where different parts of the contribution cash come from. Contributions comes from your chequing account, not from specific sources like a tax refund or a pay cheque.
- Don't get hung up on refunds. Lots of people don't get any.
* Consider the person with a taxable portfolio of investments. The RRSP refund calculated on his tax return gets wiped out by the taxes owing on investment profits. Similarly when there is additional income from renting out your basement suite, or from a small business.
* Consider the person who told his employer to lower his pay cheques' tax withholdings, and made RRSP contributions with before-tax savings during the year.
* Good tax planning aims to prevent any tax refund. A refund only indicates that you gave the government too much money too soon. If you delay recovery of the RRSP benefit that is your personal decision, not an attribute of the RRSP system itself.
- Don't get hung up on refunds. Lots of people cannot use them for any kind of additional savings.
* Consider the person who relies on the RRSP refund to pay his yearly property taxes. Because he did not have to save up for that expense from earlier paycheques, he was able to make earlier RRSP contributions instead.
* Consider the person who borrows to top-up his contribution and must use the refund to pay back the debt.
* Consider the person who temporarily draws-down the balance of his emergency fund to top-up his contribution, and must use the refund to replenish it.
What you do with any refund is irrelevant to the RRSP system itself. As long as you do not increase your spending as a result of any Contribution Credit, the Model is valid regardless.
- The source of savings makes no difference to how the RRSP works. It makes no difference whether the cash comes from a government cheque or a payroll cheque. The account is just a pot of dollars. Your RRSP won't work differently from JoeBlow's just because you funded them differently. The reality of the person who contributes $5,000 savings from pay cheques and uses his refund for a vacation, is no different from the reality of the person who first contributes only $4,000 from his pay cheques because he went on vacation - and then contributes his $1,000 tax refund as well.
Similarly, the person who creates a taxable investment account with the refund is simply making the decision to NOT use the RRSP. The fact that its funding came from an RRSP refund makes no difference. You are simply allocating your savings to a taxable account. The benefits of RRSPs accrue only for the $$ inside them.
- The timing of cash flows within a year makes no difference to an understanding of the RRSP. For simplicity they are all lumped together. It makes no difference to decisions you make between possible choices. Yes, the person who contributes in January will realize greater benefits from an RRSP than the person who waits till the last minute each year, but it won't change any decision/choice.
- Don't get hung up on the calendar date of events. For people who put the tax refund of one calendar year directly into an RRSP contribution dated the next calendar year, it is simpler to consider 'a year' to be a fiscal year that runs from (say) June to May. That way you can understand the refund to be integral to the original contribution.
- The math for the tax refund becomes an endless loop when you
a) make a contribution,
b) calculate the refund equal to the contribution * your tax rate,
c) contribute the refund,
d) calculate the refund equal to the second contribution * your tax rate,
e) contribute the refund, etc.
Much simpler to consider the refund to be integral in each contribution, no matter when or how or if it is received. Any delay is your choice and not an attribute of the system. Obviously the sooner you get the tax benefit the better, even if it is not used for a further RRSP contribution.
- If you insist for your personal reality, that recovery of the refund and its contribution is delayed a year, then use the model's variable for a delay in claiming the tax credit (discussed below). It won't be a perfect model of your reality, because the delayed contribution creates its own refund, but it is close-enough.
The Model tracks what happens to the two parts of the total contribution separately. The left column tracks the same after-tax savings that might have funded a TFSA contribution. This savings grows at the rate of return you use as a variable. It grows tax-free just like the TFSA would. Its Future Value will be exactly the same as the value of a TFSA.
What does the tax on withdrawal do ?
Now we have a model that shows what goes into an RRSP contribution and how each portion grows over time. We know the total $$ withdrawal tax, but how it is allocated between the two columns?
When you allocate all the withdrawal tax against the future value of the original credit, it zeros-out that column and leaves the after-tax savings portion in the left column untouched. This is exactly what the objective of the RRSP was - it's major purpose. The income earned by your after-tax savings is not taxed - not within the plan and not when withdrawn. Even the income on the government's loan is never taxed, but you don't benefit from that.
The RRSP system presumes that your tax rate at contribution is the same as your rate at withdrawal. This presumption is relaxed below, but for now just accept it. When the tax rates do not change, the tax on withdrawal will equal the future value of the tax credit. The tax reduction on contribution was never a benefit - it was a loan from the government. The withdrawal tax is not a tax on profits earned in the account. It is the paying back of a loan, along with all the income it earned. There was no benefit from a deferral of tax because all the income earned by the tax reduction on contribution goes back to the government on withdrawal.
If the analogy of 'a loan' is not clear, here is another analogy. Consider that your best friend Bob gave you $440 of his own money to invest for him, alongside your own $560. His money would then have funded 44% of the account. Over time, 44% of the account remains 'his', not 'yours'. When you close the account he gets his 44% back - regardless of the profits/losses earned in the mean time. His $6,506 portion is an allocation of principal, his ownership interest, not a tax on profits.
This concept of 'ownership' by the government (of a portion of your account) plays out in the Asset Allocation process. You don't AA the full value of the account. You only AA the portion owned by you. See discussion on the RRSP Decisions page
The 'experts' commonly say "While the income earned within the RRSP is not taxed - it IS taxed on withdrawal. This is wrong. Don't confuse the mechanics of the system with what its effects are. The left column is no different from the TFSA (where everyone agrees there is no tax on earnings - ever). This also disproves the somewhat-common idea that " You can replicate the benefits of an RRSP by holding assets without ever selling, so that capital gains are only paid decades later." That deferral of taxes does not replicate the RRSP's zero taxes.
RRSP's deferred taxes are not a benefit.
It is common to hear: "The RRSP's benefit is from deferring tax. A dollar today is worth more than a dollar tomorrow, so there is a benefit from the delay". But there is no benefit from deferring payment of a liability unless the income you can earn in the interim is greater than any increase in the liability. There is no benefit from the deferral when your $100 bill can be paid today or
This is the same situation in the RRSP. Your liability for the tax on employment income you didn't pay at the start, grows at the same rate as your investment returns. You end up no better off from the deferral. This shows in the middle column of the model. The taxes paid on withdrawal equal the all the original credit plus all the income it earned.
- invested at 10% but your bill becomes $110 at the end of the year, or
- invested at 20% but your bill becomes $120 at the end of the year, or
- invested to lose 5% but your bill becomes $95 at the end of the year.
The deferral of tax can cause two problems, both of which are discussed fully further below.
So, while the deferral of taxes is of no benefit, it does have possible downsides.
- The tax bracket you end up in when the taxes are eventually paid may be higher (or lower) than the rate of tax at which you originally received the credit on contribution.
- In retirement you are eligible for government benefits that may be clawed back when your income exceeds certain levels. The deferral of tax until that time may result in a reduced benefits.
The tax reduction on contribution has no value.
People mistakenly consider the contribution tax credit a value in itself. You are told it is - that getting it is a reason for contributing. You are told that you will have more money earning higher $profits as a result. You take the balance of their RRSP account at its face value and think that is your net worth. By now you should know that the contribution tax credit is an ephemeral value. It has no value for the taxpayer. It is a loan, not a gift.
The illusion of this 'value fluffing' of the apparent size of an RRSP can be a problem, especially for people contributing at the top tax rate. For them, almost half the portfolio's value is due to the ephemeral credit. When periodically calculating your net worth, you should subtract a rough estimate of the taxes you must pay on withdrawal. Exactitude is not necessary.
The fluffed-up value in the RRSP leads to errors in asset allocation when assets are held in both RRSPs and taxable accounts or TFSAs. The RRSP asset should not be 'measured' at it's account value when comparing it to an asset's value in a taxable account. The estimated withdrawal tax should be deducted. See the discussion on the RRSP Decisions page.
The illusion of this 'value fluffing' causes people to think they can game the system. They contribute free cash into an RRSP, then use the tax credit to fund a TFSA. They wrongly think they have magically grown their savings.
The claim that the contribution tax credit is a value leads to various wrong financial decisions;
- when choosing between long term TFSA or RRSP funding,
- when choosing between saving for a real-estate purchase in an RRSP, TFSA or outside both.
- when choosing between paying down a mortgage or contributing to an RRSP.
- when deciding to borrow the cash for a contribution.
- when comparing the value of tax reductions for charitable donations and tax reductions for RRSP contributions.
- when arguing the merits of government's social policy regarding pension taxation.
- when comparing the RESP's 20% matching contribution from the government .
Reconcile the different RRSP understandings.
Many readers will still be arguing that the traditional way of looking at the RRSP makes more sense to them. Here is a reconciliation of the two understandings. The same situation as above, is presented here with additional detail. The income earned in the account is calculated, and the taxes on withdrawal are broken down into the tax on the original employment income, and a tax on profits earned in the account.
(A) All the government sites claim the RRSP's main benefit is the original tax reduction on contribution. Here you can see that $440 is not a benefit because it is repaid on withdrawal. The claim is wrong.
(B) The second official claim is that the RRSP's benefit is from the deferral of that $440 tax on the original employment income. Yes, they list this in addition to (A) even though common sense argues that it must be one or the other - if either. Regardless, in this example, the $440 portion of the contribution is not repaid until withdrawn 35 years later. The time-value-of-money determines there is a value to that delay. That value equals $6,065 - the income earned by the $440 over the period.
(C) Traditional thinking claims that the profits earned inside the account are not sheltered from taxes - they are taxed on withdrawal at full rates. In the example, the $13,785 cumulative profits, taxed at full 44% rates, equals (surprise, surprise) that exact same $6,065. (B) and (C) will always exactly equal and offset each other. They essentially measure the same thing, but from different perspectives. In total they cancel each other out.
The calculation of (B) equals: Income earned by 44% of contribution.
The calculation of (C) equals: 44% of income earned by contribution.
The equations just rearrange the factors. All the tax-planning schemes that target (B) or (C) individually are pointless because in total they have zero effect.
So all the factors they talk about always add up to zero. No benefits at all. The original reduction in taxes on contribution is not a benefit - it is repayed. There is no benefit from its deferral - that is offset by the 'profits taxed on withdrawal'. What is left - what they refuse to acknowledge - is the benefit of income sheltered from tax - permanently - just like the TFSA.
(D) Most all websites make the claim that one of the RRSP's benefits is ..."The money in your RRSP grows and compounds without being taxed until you withdraw it from the RRSP." Investment income is normally taxed at preferential rates. Is there a benefit from delaying income recognition only to tax it at full rates? The answer is "NO".
You can play with the input variables using the spreadsheet if you like. Even if the conceptual model they use were true, the benefits they claim would not exist. But their model is wrong to start with.
What tax rates do you use in the analysis ?
The tax rate used as a variable in The Model spreadsheet for the CONTRIBUTIONS and WITHDRAWALS is not the marginal rate of the next dollar of income. Rather, it is the rate applied to the total RRSP contribution or withdrawal. First calculate your tax bill without any contribution or withdrawal. Then calculate it again with the contribution/draw. Subtract for the difference. Divide the $tax difference into the $contribution/withdrawal to get the marginal tax %rate.
(Difference in Tax $$ Paid) / (Contribution or Withdrawal $$)
The lower (higher) taxable income resulting from RRSP contributions (withdrawals) may change your qualification for other government programs, like GST rebates, OAS and GIS. The difference in expected $benefits should be included in the numerator of the calculation above. Some on-line tax calculators already include the OAS difference.
When using the In Or Out spreadsheet for practical RRSP decisions whether to contribute, the tax rate on WITHDRAWAL should be the incremental tax rate on withdrawal resulting from only the additional amount being considered for contribution. For example, a young person in the bottom tax bracket would expect to also withdraw those savings in retirement also in the bottom tax bracket, assuming CPP benefits use up the 0% tax bracket. But after 10 years of contributions and great rates of return, the account will have grown in size, The Minimum Required Withdrawals of the existing account are now expected to use up all the bottom tax bracket. Any additional contributions will be incrementally taxed at the second tax bracket on withdrawal. It is the second tax bracket's rate that should now be used for the decision to make an additional contribution.
Remember also that RRSP withdrawals qualify as 'pension income' that can be split between couples in whatever proportion they choose. This may lower the effective withdrawal rate when one spouse has unused tax brackets. Spousal RRSPs also allow for anticipated withdrawals at lower tax brackets than faced by the contributor.
The tax rate on INVESTMENT INCOME is a bit complicated. The statutory marginal tax rate for the type of income at your marginal tax bracket is just the starting point. You can look that up on the Tax Rate spreadsheet. Most people's portfolios generate interest, dividends and capital gains income in different proportions. Use the 'Weighted Average Tax Rate' tab on the Deconstruct Benefits spreadsheet to find the portfolio's average rate.
Capital gains may be realized each year, or delayed decades. If you turn over your holdings yearly, the rate needs no adjustment. But if you hold stocks for (say) 10 years, the effective yearly rate is lower. Use the 'Effective Capital Gains Rate' tab to find a 'good enough' calculation. This is the rate for capital gains to use on the 'Weighted Average Tax Rate' tab.
The In Or Out spreadsheet includes two variables for the TAX RATE ON INVESTMENT INCOME - for the periods before (while you are working) and after mandatory RRIF withdrawals. While you are working, investment income earned outside an RRSP would be taxed at increasingly higher marginal rates as your salary rises (hopefully), and also the size of a taxable portfolio increases. By the time you retire both can be very high. There is no one correct variable input because it will be changing. See the effects of different inputs. After retirement, depending on the size of your CPP and other pensions, your tax rate can be very low.
At retirement, some people conveniently forget all the past tax sheltering of investment income and start belly-aching about the high taxes they pay on RRSP withdrawals. They compare withdrawal taxes to the low investment-income taxes they would pay if the investments were outside the RRSP. This is disingenuous. The problem of facing a higher tax bracket in retirement has been widely discussed since at least 1985. The benefits from RRSPs will probably never accrue evenly or continuously over a lifetime. That does not mean the benefits are not great.
What happens when tax rates change ?
Inside an RRSP you get a bonus when your tax rate on withdrawal is lower than it was for the contribution. You pay a penalty when your tax rate rises. The amount of the bonus/penalty
= the amount withdrawn multiplied by the change in rates.
The difference in tax rates can be influenced at either end. You try to time your contributions to be at a higher rate, and you try to time your withdrawals to be at a lower rate. But be clear that just because you try to maximize the contribution tax credit does not mean it is a benefit in itself. It is the difference in rates that generates the benefit/penalty.
This effect (of a change in tax rates) is the major difference between the TFSA and the RRSP. Rate changes have no effect on a TFSA. In the example above, the value of the TFSA would be $8,280 regardless. It is most meaningful to calculate the penalty/bonus in relation to the amount that should have been available - the amount that would have been available in an TFSA - the $8,280. The formula is
( Change in tax rate % ) divided by ( 1 minus tax rate on contribution )
10 / ( 1 - 0.44 ) = 17.9 %
17.9 % * $ 8,280 = $ 1,479 bonus
Play with the variable inputs for the 'Tax Rates'. You will see that a same 10% difference in rates has a greater RELATIVE impact on people at the higher tax brackets. This is because it is calculated on the fund total. For high tax bracket earners, more of that total plan value was created by the government's loan. This makes contributing into a spousal plan (where the spouse will be withdrawing at a lower tax rate) even more important for high wage earners. Same with pension income-splitting.
This calculation shows how regressive the RRSP model is. Compare that 17.9% bonus from a 10% reduction at a high tax rate --- to the bonus from the same 10% reduction at a low tax rate. Change the contribution and withdrawal rates to 30% and 20%. The lower income persons gains only 14.3% above what they should have received.
( Change in tax rate % ) divided by ( 1 minus tax rate on contribution )
10 / ( 1 - 0.30 ) = 14.3 %
14.3 % * $ 10,350 = $ 1,479 tax.
The government could have got rid of this problem simply by stipulating all contributions(draws) be taxed at the first tax bracket's rate. Even during the pension reform discussions after the 2008 credit crunch there was NO discussion regarding this regressive tax or calls for change. It is likely that policy was written specifically to benefit the high income earners while hoping the poorer classes never find out.
What are the unknowns that will determine a tax rate change ?
The effect (positive and negative) of a change in tax rates is an unknown until the time of withdrawal. It is a risk. Anyone telling you that you will certainly withdraw at lower tax rates is a snake-oil salesman. The only people in a position to 'assume' a bonus are those contributing at the top marginal rate.
One way to deal with the unknown withdrawal tax rate is to make it the 'conclusion' instead of the 'given' in your analysis. Use the In or Out spreadsheet to find the break-even point in your particular decision. Make your best guess for all the other variables, then play with the withdrawal tax rate variable until both choices are of equal value. Then ask yourself "what is the probability of the future tax rate being larger/smaller than that variable?"
Some people say you should never assume your tax rate will rise on withdrawals because you could always withdraw the funds first, before your rates rise. There are two problems with this. First, most of the unknowns listed below will be unknown until they happen - too late. Second, withdrawing funds early would be shooting yourself in the foot - giving up the shelter from tax for all the future income earned.
Other people claim you should never assume your tax rate will rise because you should stop working and amassing wealth before that happens. Do you really need to told how silly that argument is? Some people will spend $100 to save $1 in taxes.
The only certainty is your tax rate on contribution. Following is a list of variables that will impact your tax rate on withdrawal.
- Your total income when withdrawing
- If you die the year you retire, the total RRSP balance gets taken into income in one year. That can push a lot of it into the top marginal tax bracket. The extend of this risk depends on the amount in the plan.
- If you withdraw larger amounts than the minimum required, they may push you into a higher tax bracket. E.g. if you need money to pay for long-term care, or take an around-the-world cruise.
- The amount you save inside the RRSP and the investment returns you earn.
- The bigger the fund at retirement the larger the required draws. Even if you worked and contributed in the bottom tax bracket, it is possible to create a fund large enough so that the draws push you into the next higher tax bracket, or even beyond.
- Your savings and investing results outside the RRSP impact the income you must consider BEFORE calculating the marginal tax bracket of the RRSP draws.
- Maybe you inherit a large sum.
- Maybe you realize a large gain from the sale of your principle residence that gets invested and earns income.
- Maybe you keep savings outside the RRSP because of the high cost of foreign exchange conversions.
- Government programs
- Any income you receive from Canada Pension Plan, Old Age Security, etc, should be considered as taxed first at the lower rates. The greater the government benefits, the farther along the tax-bracket scale the RRSP draw is pushed.
- Government programs providing income assistance in retirement (GIS, OAS) have clawback provisions that reduce the benefits when your income exceeds certain limits. This has the effect of increasing the effective marginal tax on any RRSP withdrawals. This effect does not exist for TFSA's.
- Tax rates can be changed by the government at any time.
- The deficit spending by governments in 2009 may force increases to future tax rates. Someone will have to pay.
- The economy in general and government deficits will determine political realities.
- The retired persons lobby groups may influence rate decisions.
- The width of tax brackets (how much income is taxed at each level before you get bumped up to the next tax rate) may increase with inflation, or not.
There are only two relatively safe assumptions that can be made from all these unknowns.
- Those who contribute at the top marginal tax rate have a pretty good chance of taking money out at lower rates. Therefore they benefit from the rate change.
- Those who contribute at the bottom tax bracket have a chance of taking money out at higher rates, if they are good savers and investors. Therefore they lose from the rate change. This reality is what prompted the government to create the TFSA.
Many government support programs are means-tested. Their level of benefit depends on your income. Reducing your income with an RRSP contribution may increase the Child Tax Credit or the GST Credit. Contributions made after age 65 but before 71 may increase GIS and OAS benefits. Increasing your income with RRSP withdrawals may reduce Old Age Security benefits, the Guaranteed Income Supplement or the Age Tax Credit. Your taxable income can also determine your costs for long term care or subsidized community housing. How you deal with this depends on how entitled you feel to the benefits.
For example, a huge fuss is made about the clawback of Old Age Security benefits. People need a reality check. The mean income in Canada is about $40,000. Anyone earning more than $60,000+ (at which OAS only starts to be clawed back) is far above 'normal' and certainly not deserving of taxpayer support, especially when the taxpayers are earning less than the people they are supporting. Consider how you feel about people who refuse to get a job because the paycheque earned will be offset by a reduction in welfare benefits. Do you think these people are doing 'good financial planning'? Or do you think they are abusing the system? Do you see the analogy?
Do you feel blessed to not qualify for benefits because you are wealthy? If so, just ignore the loss of benefits.
Now there is a choice to use a TFSA instead of an RRSP for new retirement savings. Two people in the same 'blessed' situation will now face very different outcomes because of their choice of account-type. Ignoring the loss of benefits is not logical in the choice between account-types for new savings. It is not the clawback of OAS that matters to the vast majority of us. It is the clawback of the GIS. While the RRSP withdrawal creates $1 claw-back for every $2 withdrawn, cash from a TFSA creates none.
You deal with this issue by treating the lost $$ benefits as an increase in $$ taxes paid - as an increase in your marginal tax rate. The middle column above shows the typical income in retirement. The GIS is not taxed so it starts below the bottom of the lowest tax bracket. The OAS is taxed, but not counted as income that causes a clawback of GIS. The CPP is roughly 20% of someone's ending wage ($40,000 is assumed here). RRSP draws would fall in the box of Other Taxable Income. The first $10,000 RRSP draws would clawback GIS at 50 cents on the dollar. So the marginal tax rate would be the sum of the statutory 22.5% plus the 50% clawback - total of 72.5%.
The Save in Which Account-type - RRSP or TFSA or Taxable? spreadsheet tracks yearly account balances with normal government benefits when saving starts at age 35 with retirement at age 65 and death at age 99. Since most all taxes and benefits are inflation adjusted the numbers should hold true over time. You see GIS clawbacks created by CPP income in all three account types. Notice that additional clawback is also created in taxable accounts because the portfolio's profits are taxable income.
There is no economic justification for the different outcomes between RRSP and TFSA savings. Given a choice the government should prefer people use an RRSP because it provides them with cash at the time they need it. Why does their policy do the opposite? The obvious solution to this inequity is to re-write the rules regarding qualification for benefits. Using the same Minimum Required Withdrawal from the RRIF system, the same percentages could be multiplied by any existing TFSA balance, and added to the taxpayer's other actual income.
Delay Claiming The Tax Deduction ?
The RRSP administrative rules do not force you to claim the tax deduction in the same year you contribute. Since the contribution credit is calculated at your top marginal rate, it seems intuitively better to delay the claim when you predict your marginal rate will rise in a few years. After all, a $440 refund (44% tax credit on $1,000 contribution) is better than a $340 refund (at 34%).
Use The Model spreadsheet as it stands so far (above). With those assumptions you see a bonus from the lower tax rate on withdrawal ( 34% instead of 44%). What happens when you must delay claiming the tax deduction because you only expect to be in that higher tax bracket a few years down the road? You pay a penalty.
Input a delay of 5.5 years. The middle column now shows the tax credit being invested for only 29.5 years - not the full 35 years. As a result it grows to only $4,260 (instead of the $6,506 when there was no delay). It is missing the profits that should be earned in an extra 5.5 years. Even with the bonus from the lower withdrawal tax rate, there is not enough money in the middle column to pay all the withdrawal taxes.
The penalty from lost profits grows, the longer the delay in claiming the tax deduction. Here, the $1,257 penalty from the delay wipes out all the bonus from the difference in tax rates ($1,254) within 5.5 years.
So what do you do when you expect your marginal tax bracket to rise in the future? The decision to contribute to an RRSP, but delay claiming the tax deduction, should consider all the alternatives. It will rarely be the optimal choice. The optimal choice will always be to stash the funds in a TFSA, but maybe you have no TFSA contribution room. Your other choices are between using a Taxable account or using the RRSP while claiming the deduction right away. This decision is modeled on the next webpage for RRSP Decisions.
Are the Tax Benefits of Investment Losses Lost ?
It is common to hear that the tax-recovery benefit of investment losses is lost when the loss is inside an RRSP. There is some truth to that, but not a lot.
First off, no one invests with the intention of losing money in the long run. And in the long run markets go up, so most everyone earns some kind of profit. In a taxable account both profits and losses generate tax effects, but over time you only pay tax on the net profits - profits minus losses. Neither the TFSA nor the RRSP is taxed, so their benefit equals their shelter from that same net tax. Losses have the same effect as profits.
But some loss of tax benefits may happen when market returns swing wildly from large losses to large gains (or vice versa). This is because tax effectively dampens returns. It reduces both after-tax profits and after-tax losses. Volatility reduces long-term return calculations. Reducing volatility (which is what tax does) increases returns. The box below shows the math. One year of 100% profits and the other year of 50% losses. It shows both the TFSA and RRSP ending up smaller than if the investments were in the taxable account.
But there are some qualifications to this understanding.
First - if the profit were 50% and the loss 25% the tax shelters would have beaten the taxable account - still a volatile return but with a net profit over the period. The volatility must be very, very extreme.
Second - the math depends on the taxes being paid from the investment account. But are they? In real life taxes are paid from your chequing account. A reduction in a chequing account may impact your decision to add additional savings to the investment account, but it will not impact the existing value at risk in the investment account. Maybe additional savings go into the RRSP and not the taxable account. There are a lot of maybe's.
Third - the math depends on the taxes being paid each year as profits are earned or lost. But they aren't. In real life an investment may be held for many years as its price increases. By the time the market tanks, that loss may retrace only part of the gains, leaving the stock price still higher than at purchase. There will be no tax loss.
Fourth - even if a loss takes the stock price below where it was purchased, claiming a capital loss in Canada will not generate at tax refund unless you have paid taxes on capital gains during the prior three years. Buy-and-hold investors who rarely trade may be out of luck.
Fifth - the math shown is reasonable for a single security. But a portfolio of assets will have some losses offset by some gains. Taxes paid are paid on the net gains of all the assets.
The appropriate conclusion should be to not worry about any lost benefits from capital losses.
RRSP Decisions and Choices
After the list of decisions grew too long, it was given a page of its own at RRSP Decisions.