Canadian Money Forum banner

Value of CPP

21K views 30 replies 10 participants last post by  valueindexer 
#1 · (Edited)
For reasons that would not seem sane to anyone who doesn't check this forum every day, I'm trying to model the value of the CPP to see if it's worth contributing to it when given the choice. This includes the employer contributions, which affects you directly if you own your employer or indirectly if you can't negotiate a higher salary because the employer is paying this cost.

If anyone who knows more can find anything wrong with this that would be great to know :)

Basics:
Maximum monthly CPP payment, in 2012, when retiring at 65: $987
Annual contributions including employee and employer for maximum eligible earnings in 2012: $4500+
Ages: starting at 25, retiring at 65
Cost of an annuity worth $987/month in 2012, with inflation indexing: $200k-250k (guessed from indirect sources)

The model:
Instead of paying into the CPP, invest $4500/year for 40 years at a 3% real after-tax rate of return
End up with $339k, adjusted for inflation
Buy an annuity paying $1673/month with inflation indexing?

The result:
The CPP is worth up to 42% less than investing yourself, with conservative estimates? That almost seems to put it in the same range of financial quality as universal life insurance. If you plug in some slightly higher rates of return you can get as much as 5x the income by investing yourself.
 
See less See more
#2 ·
Welllll...there's a lot to comment on here.

First, it is possible to get the actuarial present value of a future income stream. That is, you can get an estimate of the total value of a lifetime income stream even when the end date is random (insurance companies do this all the time).

However, the value is extraordinarily sensitive to the assumptions you use - the mortality assumptions and the discount factor you use for inflation indexing. You need to be cautious when projecting values which are sensitive to assumptions over long time horizons.

Secondly, I'm not sure what sources you used for your annuity quote, but you cannot find a lifeco paying $987/month with a CPI adjustment at age 65 for $250K. You are looking at more like $350K for a male, and higher for a female.

Thirdly, keep in mind that your CPP contributions (and those made by your employer) are made with pre-tax dollars, and you'd need to save with post-tax dollars, unless you save exclusively through an RRSP. (Tax is something of a red herring in this discussion, but to fully model your scenarios, you'd need to include it. Then you'd need to model the tax implications for the kind of annuity you purchase at age 65.)

Finally, for what it's worth, CPP provides a form of longevity-insured income. To some extent I think it is useful to think about this as providing a form of risk transfer - you are transferring the risk of outliving your funds to a (very strong) counterparty - versus thinking of the CPP income stream as "an investment."

By way of example, without a doubt I would be "better off" financially by NOT purchasing house insurance, but instead investing the funds I use now to buy insurance.

If I extend that process over a long enough time period, and make generous enough assumptions about the rate of return I will earn, then I will have enough to cover the cost of replacing my house - so long as it does not burn down while I am foregoing the insurance.

The same thing applies (in reverse) with longevity-protected income: you can think of many scenarios in which you would be better off foregoing the insurance rather than purchasing it (in your example, not paying CPP premiums vs. paying them) - but they are all predicated on the basic assumption that you don't need the insurance.

In your example, you suggest buying a private annuity at age 65 (or whenever) to provide longevity-insured income, you have covered that issue off. But if you want longevity-insured income, you will not be able to find it more cheaply or efficiently than through CPP, if you fully model your contributions over time. The private life insurance business is not very efficient.

I think this discussion is very worthwhile. But there's lots to consider.
 
#3 ·
Great points. I wasn't sure if I estimated the cost of CPI adjustments for an annuity correctly since all the prices I saw didn't mention them. On the other hand if you can get more than a 3% real after tax return the annuity you can buy climbs quickly. Or if you want to keep an investment portfolio, $987/month is a 3.5% withdrawal rate on $339k.

If you aren't getting a salary but are getting the same total income, you would need to be using other forms of income that may have lower taxes. For example at an average income level small-business dividends seem to potentially reduce the amount you pay to the government by $4000-5000 which gives you the investment amount. And if you had no other investment plans a TFSA would hold it nicely but we all have that maxed out already :)

Since an annuity with CPI adjustment seems to be the closest substitute for CPP benefits it sounds like it's at least comparable unless there's a big tax effect somewhere (but it doesn't give you the benefits of government backing and the possible election of a socialist government when you're 64). I wouldn't consider it risky to buy an annuity or withdraw under 4% from a portfolio annually.
 
#9 ·
You're right that applying the correct model is essential. What I tried to do is take a model similar to what you would do when comparing expensive life insurance products with "buy term and invest the rest".

With option #1 you hand money every year to a wise and benevolent partner (that also has the power to change the rules at any time). After accumulating 40 years of value this way you start to get money back for as long as you live.

With option #2 you keep your money and put it into a portfolio every year. After accumulating 40 years of value this way you can withdraw from it using a fixed or variable withdrawal rate, or hand it over to a wise and benevolent partner (that can't change the law) and get money every year as long as you live.

Other than that it uses inflation-adjusted numbers everywhere and assumes the CPP rules stay the same in relative terms. It's hard to see the CPP benefits increasing significantly unless we manage to produce a lot more contributors or practice political timing. They could be reduced or delayed. Do you see any specific points where one of the options doesn't match what the other is doing?

By the way, it looks like the FP has picked up the story as well and quotes an effective 3.6% rate of return on CPP contributions: http://business.financialpost.com/2...le-have-to-foot-the-bill-for-boomer-pensions/.
 
#12 ·
OK.... in rrifmetic, you can do either of:-

-turn off/on salary CPP withholding
-turn off/on CPP income

For a 27 year old earning $65K, planning to retire at 65, turning off cpp withholding and eliminating cpp income, results in a 'die-broke at 95' annual ATI of $39,280. Taking part in the CPP scheme results in a $41,242 ATI.

Seems like a pretty good deal to me.

(above based on an investment rate of 4% and cpi of 2%.)
 
#14 ·
OOPS... I guess I should have read the article. Our engineer may have a point. When you run the numbers using 8% instead of 4%, then the "CPP: no thanks" option does outperform somewhat. This is no surprise when you think about it.

Now, all we need is a strategy that guarantees 8% return for the next 50-60 years.

BTW, I trial and errored it, and it appears that 7% is the tipping point.
 
#15 ·
I have never bothered with it because I had little choice in the matter. My only concern now is when to take it.

Would the rate of return not vary depending on age, and more specifically how many years of post 1997 contributions you have made? This is when, as I recall, the CPP percentage rates started to climb. So, in my case only about 2/3 of my contributions were made prior to this period-at lower rates.
 
#19 · (Edited)
TO OP: try reading some of the triennial reports of the Chief Actuary on CPP. You can fidn them here:
http://www.osfi-bsif.gc.ca/osfi/index_e.aspx?DetailID=499

The CPPIB was created in 1997 to invest the surplus revenue of the CPP fund, in order to increase investment returns, and thereby reduce future costs. (Annual CPP contributions currently exceed CPP benefit payements and other expenses, and this is expected to continue until about 2020.)

Prior to the creation of the CPPIB the surpluses were "invested" in government bonds. The government essentially lent the money to itself at prevailing central bank rates, and credited the account accordingly. (I thnk Provincial governments borrowed from it too, but I stand to be corrected.) In any event, the Chief Actuary's reports have tables with historical investment yields before the creation of the CPPIB as well afterward. (See Table 10 of the 25th report for example) It also has reports and projections of real return for various periods.

Your mother's ROI should be calculated by what her CPP benefit will be, compared to her contributions, not the present annual earnings rate of the CPPIB.

Table 34 makes interesting reading:

The results presented in Table 34 are rates based solely on contributions paid and benefits received; ...
Table 34 Internal Rates of Return by Cohort
(annual percentages)
Birth Year Nominal Real
1940 ------ 10.4 ----- 6.3
1950 ------ 7.1 ------ 4.2
1960 -------- 5.3 ------ 3.0
1970--------- 4.7 ------ 2.4
1980 -------- 4.6 ------ 2.3
1990--------- 4.6------ 2.2
2000 --------- 4.6 ------ 2.3
The higher internal rates of return of the earlier cohorts mean that they are expected to receive better value from the CPP than those who follow.... However, the rates stabilize for cohorts born after 1970. The real internal rate of return of about 2.3% for these cohorts provides a competitive return in a low interest rate environment. The CPP is able to provide such a return due to the Plan’s size, financing method, and the resulting ability to mitigate
risks.
 
#20 ·
The breakeven point is 5% only if you hold the end date constant at age 95. (And any one individual's likelihood of dying at age 95 and not before and not after is actually quite low.)

The breakeven point is much higher if you die earlier (which is statistically likely) and lower if you live longer, naturally.

You can measure or approximate the embedded alpha in a life annuity (whether CPP or a privately-purchased annuity) using the implied longevity yield.
 
#22 ·
OhGreatGuru: that seems to back up what I found, with everyone under 40 getting a real return under 2.5%. It's not clear if they based this on actual benefits since you can't retire at 40 under the CPP although they could compare to the current level of benefits like I did. And I'm not an engineer :)

If we accept Jeremy Siegel's finding of a long-term 6.8% real return for US stocks as being within the limits of reality, going from 6.8% to 2.3% would seem to be extremely conservative. Maybe even unreasonably conservative at a personal level unless you're chasing hot mutual funds every year.

The 5% real return breakeven that Steve found would be much lower when considering the doubled inflows (just like it's not accurate to say a bank is profitable because it's solvent after it gets a bailout). Since CPP benefits are taxable I'm not sure if the comparison even needs to use an after-tax rate of return. You can't generate RRSP room without participating in the CPP and none of us would have that much room left in the TFSA, so regular unregistered investments are the most likely way to do this.

If you only generate capital gains when you sell to get income, and you have a dividend yield of 2-3%, that might mean you're taxed on approximately half your income at withdrawal if you keep the portfolio instead of converting to an annuity. Which might be even better tax treatment than CPP benefits depending on where your total income is. And that doesn't count the possibility of dividend optimization to reduce the taxes. The only exception would be if the capital gains and dividends during your accumulation years lower your return by a few percent.

The ILY is an interesting measure. That paper might even set the bar too low for the required returns if it doesn't compensate for the extra drag of monthly withdrawals in a down market. It could be applied to the CPP using the difference between early and delayed payments, but that's only accurate if they are affected by actuaries and not politicians.

It's possible the CPP benefits will grow if it has good investment returns or if it gets overfunded somehow and doesn't have to compensate for past undercontributions. Otherwise, I still don't see it. It could even get worse with lowered payouts and higher retirement ages. Even in politics 40 years is a long time. Can't a large insurance company provide an annuity with mortality credits and a diverse coverage pool?
 
#23 · (Edited)
The ILY is an interesting measure. That paper might even set the bar too low for the required returns if it doesn't compensate for the extra drag of monthly withdrawals in a down market.
Which is why you should really only compare CPP income streams with other guaranteed income streams. Since you liked the Milevsky/Aronoff piece, you might also like this other Milevsky piece on the true cost of retirement:

http://www.advisorone.com/2011/09/01/what-does-retirement-really-cost
 
#26 ·
So if you could choose to hand over your full investment portfolio and all future investments to be managed by the CPP under the same terms as the regular benefits, you would do that? If nothing you can do privately is comparable that seems like the best way to go.

Interestingly the last piece on the cost of retirement quotes a $1000/month inflation-adjusted annuity at $230k, right in the range I guessed initially. This seems to be from 6 months ago.
 
#28 ·
Interestingly the last piece on the cost of retirement quotes a $1000/month inflation-adjusted annuity at $230k, right in the range I guessed initially. This seems to be from 6 months ago.
I double-dog dare you to find that quote in Canada. :) That's a US quote for the US market, and the US and Canadian annuity markets are very distinct (typed from a lounge in the Denver airport)
 
#30 ·
I think you don't necessarily need indexed annuities if you annuitize over time and keep a significant portion of your portfolio in inflation sensitive assets like utilities, consumer staples and real estate. As I recall, indexed annuities are very expensive, and have a cap on the inflation rate they will insure against, so are a limited hedge against inflation risk.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top