[Sampson]

The series of articles in the Globe and Mail about the growing problem of underfunded public and private pensions has got me thinking.

Are DB really that good? I know from some of MDJ's posts that Federal Government Employees can get 70% of their average salary from their last few years of service, but I think that's rare. In my public pension plan, it appears that % of salary paid out are about 55%, 47%, 39% for 35yrs, 30yrs, and 25yrs of service respectively.

Given that my contribution rates keep going up (20% next year - half from employer), I was wondering whether I'd be better off with my employer simply giving me that money.

I made some preliminary models, and I'm actually finding it difficult to see an obvious benefit to the DB model. My rates of returns over the years of service were very modest (4-5%) - and considering many DB pension plans have severe penalties for early retirement, I'm starting to think I wish I had a DC plan (with similar rates of employer matching funds).

Taxes obviously play a major role, but lets assume all the money goes into an RRSP so is not taxed in either the pension, or in my own hands.

I'm I way off base here? Thoughts?

 

[MoneyGal]

You are focussing on one aspect of retirement income streams -- expected return.

In my view, this is not the appropriate way to evaluate whether you'd be "better off" with a DB or DC pension.

DB pensions protect against two three major risks to the sustainability of income in retirement -- longevity and Sequence of Returns. If they are inflation-adjusted, they protect against inflation risk as well.

 

[yyzvoyageur]

You're starting to sound like me:

http://www.canadianmoneyforum.com/showthread.php?t=1060

If I had the choice, I would take all of the money going toward my pension and CPP and invest it myself within an RRSP. I'm convinced I would be better off.

What pension plan is that? The percentages sound really odd.

 

[Sampson]

DB pensions protect against two three major risks to the sustainability of income in retirement -- longevity and Sequence of Returns. If they are inflation-adjusted, they protect against inflation risk as well.

Hadn't given these factors too much thought. I suppose the age/longevity assumptions have to be made, but it wouldn't be very nice I'm 95 and end up without pension money. Perhaps protection against the risk of sequence of returns is their biggest advantage. I know my pension is not fully indexed to inflation so only limited protection there.

Pension plan is Alberta Public Service Pension Plan. Only started to look into payout %'s, so those are rough estimates.

I think it may not work well for me because I plan to retire early, and not actually participate in the pension contributions that much

Oh well, I guess I'll have to keep planning as if I won't be getting a pension

 

[ghostryder]

You're starting to sound like me:

http://www.canadianmoneyforum.com/showthread.php?t=1060

If I had the choice, I would take all of the money going toward my pension and CPP and invest it myself within an RRSP. I'm convinced I would be better off.

What pension plan is that? The percentages sound really odd.

There is an interesting perspective in the book "Filthy Lucre" on this topic.


"when we purchase life annuities as a group through an employer they are no longer called annuities. Instead, they are called "defined benefit pension schemes". This change in nomenclature creates all sorts of confusion; there is an inclination to regard pension schemes as savings arrangements, rather than insurance products. Nevertheless, an annuity is essentially what is being purchased: in return for an upfront payment (the "pension contribution"), the employer guarantees a fixed periodic payment from the time of retirement until death.

.... In the debates over "privatization" of social security in the United States, for instance, people routinely compared the rate of return of money that was saved and invested in the stock market with the rate of return of money paid into Social Security. Yet analyzing the latter in terms of rate of return involved a category error. It amounted to comparing an investment to an insurance policy....


<paraphrasing here> what proponents of "privatization" of Social Security in the US were reccomending was not really privatization of the system. That would be people purchasing their own annutities individually. What they are actually reccomending is that people stop purchasing insurance and simply save for their own retirement.....their goal was simply to undo a mutually beneficial risk-pooling arrangement for no other reason than an ideological hostility to government."


Maybe you could do better on your own, but are you comparing apples to apples? A DB pension, especially a public sector one has a "guarantee" like no other.

 

[AndrayDomise]

I find that the DB vs. DC debate often misses the point. There's no reason the two can't be reconciled. Many employers still using DB plans offer some form of DCPP or group RRSP as well, to which you can make automatic contributions.

If not, open your own and make contributions. Even with the pension adjustment, there ought to be ample room available to build a complementary retirement nest egg.

DBPP and DCPPs are meant to cover two different forms of risk, and there isn't any reason you can't do both.

 

[andrewf]

Can't you solve the longevity risk with annuities, and both longevity and inflation risk with indexed annuities? Sequence of returns remains a risk for individual investors, just as sponsor insolvency largely remains a risk for private sector DB recipients. I won't say anything about public sector DB plans, except that it is part of the general problem where public sector workers are paid far more than their private sector counterparts, even without benefits, and once benefits are factored in, it becomes a bit embarrassing.

 

[MoneyGal]

AndrewF: yes. It is more expensive to purchase annuities privately than to get the same kind of longevity insurance through a DB pension plan, though.

 

[CanadianCapitalist]

It's interesting that those in DB plans think DC is better and those in DC plans think DB is better. Isn't the grass always greener on the other side?

I work in the private sector and would not be very comfortable being in a DB plan because we are seeing what's happening to pensioners at Nortel, GM and other troubled companies. I'd rather be in a DC plan and assume the risk of a shortfall.

Even before we get to the risks that MoneyGal is talking about, there are 2 risks with DC plans: (1) Saving risk and (2) Investment risk. Employees in DC plans may not be saving enough for their retirement and typically they are on their own when it comes to managing their portfolios.

Despite these risks, I don't see how the private sector will go back to DB plans. So many have been burned as these plans turned out to be far more expensive than initial estimates.

 

[Four Pillars]

I don't think there is anything magical about DB plans. The main "benefit" of DB is that for some people, the employer makes some of the contributions.

I guess it boils down to - do you consider the employer contribution to be a "bonus" or just part of your compensation that you have no control over?

 

[HaroldCrump]

I don't think there is anything magical about DB plans.

The magic is the longevity insurance, and in many cases, inflation insurance that it provides.
To replicate the kind of retirement income stream, longevity insurance and inflation indexing that gold-plated DB plans provides, a corresponding private sector worker with a SD RRSP in similar job would have to work much harder, longer, make much more gross income and have a higher % of savings put away each month.
For someone who begins working for the federal govt. in their mid-to-late 20s, is easily able to retire at 55 (or earlier, depending on when they started) with the full force of the govt. pension behind him/her.
The RRSPers are faced with very stiff risks (esp. equity market turmoil) that most RRSPers will not be able to overcome.
The best case scenario for them is hoping they have enough to buy an annuity.
I don't know the going rates of annuities, but I'd imagine that getting an annuity that has the inflation indexing and longevity insurance features of a DB pension at age 55 will be extremely expensive, esp. these days with low interest rates.

 

[MoneyGal]

The magical thing about DB pensions is that they are cheaper for employees to participate in and employers to buy.

Why? Forced participation.

People who buy annuities privately tend to be healthier than the population as a whole, which drives the price of annuities up relative to the longevity insurance provided by DB pension plans.

In a DB pension plan, *everyone* participates, whether they are long-lived or not, and whether they personally have no longevity risk aversion (and thus would opt out of the company pension plan if they could). And then those who die earlier (and stop receiving pensions) subsidize those who live longer than average.

No privately-purchased personal pension can ever match that. This is the same reason why CPP is a "good deal" for recipients, and pays more than a similar lump sum (the discounted value of CPP entitlements at retirement) would buy in the open market.

I think the "grass is greener" arguments arise because people compare two different things. If you look at longevity risk, DB pensions are a no-brainer winner. But if you are looking at retirement savings through an investing lens (not a longevity risk lens), all of a sudden DB pensions look "riskier" because the payout depends on an uncertain and random date of death (plus loss of liquidity, credit risk of the issuer, etc).

In addition, I think a couple of issues are being collapsed in this discussion: one is - which is the best way to save for retirement? And the other is, which is the best way to provide for a guaranteed lifetime stream of income once you are IN retirement?

I think these two questions need to be considered separately, and suggesting that somehow DB and DC pensions are "the same" or "there's no winner" is evidence of these two aspects being collapsed.

JMO.

 

[MoneyGal]

X-post with HC, who makes excellent points. An inflation-indexed annuity purchased at age 55 (especially for a woman) is staggeringly expensive and the true value of public sector pension plans is enormous.

 

[MoneyGal]

An indexed pension of $50,000 for a 55-year-old woman costs about $1M today.

 

[Sampson]

It's interesting that those in DB plans think DC is better and those in DC plans think DB is better. Isn't the grass always greener on the other side?

Not necessarily a 'grass is greener' situation for me necessarily, I'm just a control freak and would like plan the remainder of my savings/investments accordingly.

What got me first going on this issue was looking into the 'milestone' years of employment that are required to have bumps in the DB payout.

Because I have plans to semi-retire very early, having worked between 15-20 years, the payouts (although protected against inflation and longevity risk) will not be nearly as good as those working 30+ years (i.e. the payout is not linear to time worked).

If I had a DC plan instead, at least I could plan my asset allocations properly. Who knows how these public pensions will fair, and what the rate of contribution increases will be over my years of employment.

I agree with both MoneyGal and Harold, re: annuities and also forced participation. These are likely overwhelming benefits to most in this pension plan, just not necessarily beneficial to me

 

[OhGreatGuru]

The series of articles in the Globe and Mail about the growing problem of underfunded public and private pensions has got me thinking.

Are DB really that good? I know from some of MDJ's posts that Federal Government Employees can get 70% of their average salary from their last few years of service, but I think that's rare. In my public pension plan, it appears that % of salary paid out are about 55%, 47%, 39% for 35yrs, 30yrs, and 25yrs of service respectively.

I've taken a quick look at the Alta Pension Plan booklet, because the numbers you quote are at variance with other typical public pensions, such as the federal public service. The difference seems to be in how they handle coordination of benefits with CPP. The federal plan is nominally 2%/year of service to a maximum of 35 years, or 70%. But that pension is reduced at age 65 by a formula that approximates your CPP entitlement. The Alberta formula, on the other hand, seems to incorporate the CPP integration right off the bat - the pension is 1.4%/ year of service for the average salary up to the YMPE, and 2% for the portion of salary over the YMPE. But then there is no reduction at age 65.

The Alberta method might present cash flow problems for people who are retiring before age 60, but in the long term it probably amounts to a similar payout.

 

[MoneyGal]

Sampson: the forced participation is a benefit to *everybody* in a DB plan during the accumulation stage. You only lose out if you die before 50% of your cohort.

And I don't know why you'd say "I don't know how these public pensions will fare" - they are backed by the strongest counterparty going, the one that can actually print money to back up the promise.

But yes, public sector pensions are not linear. They are among the only pensions that pay based on best five (or whatever) years of salary. So what that means is a clerk (for example) who started with a salary of $32,000 but finishes with a salary of (for example) $62,000 will receive a pension as if she had made contributions on a $62,000 salary for her entire career.

It's really a staggering benefit. The last actuarial valuation of the federal public service plan showed that at the then-current yield of 1.73% (on a RRB), the plan was worth more than 33% of pay. Discounting at today's rates (I usually use 2% to estimate a long-term inflation-adjusted rate) would produce a contribution rate of exactly 30%. That's the cost of the public sector guarantee.

 

[andrewf]

An indexed pension of $50,000 for a 55-year-old woman costs about $1M today.

Isn't that about the liability faced by the fed gov't due to public sector pensions? Considering that the federal DB plan requires contributions of ~30% of salary, this seems not far off the mark. My rough calculation tells me than 30 years at 30% contribution on a $50k real salary requires 5% real ROR to yield $1 million. Doesn't sound far off the mark to me.

 

[MoneyGal]

Yes, but:

1. public sector workers do not contribute 15% of the totals required to fund their pensions (because of the last-best earnings issue I described earlier) and

2. what 55-year-old female private sector worker earning $60-$65K do you know who has been able to build up a lump sum of $1M by making maximum RRSP and TFSA contributions?

 

[fraser]

I believe that the debate over whether DB's or DC's are better very much depends on the individuals age, their propensity to manage a DC plan, their ability to save, etc. I am in a DB plan, which ends this year. Ten years ago my employer offered us an option to move to DC while at the same time closing DB to new membership. I was 48 and had 15 years of service. I declined the offer because I thought that that the offer was too little. Many people accepted...it was the time of the internet bubble and 15 point returns were common. A year later, things went south. For me, the DB was really one leg in the three leg stool of personal savings, CPP/OAP, and DB. The DB that I will be getting will form the backbone of my retirement planning even though is is not indexed. Fortunately, the plan is well funded, secure, and not large. I think that we need some immediate changes in provincial and federal regulations governing DB plans to make them better and more secure for both employers and employees.