CPP and OAS

[brucecohen]

CPPIB's 4% real return target is derived from the assumptions used by the Chief Actuary in CPP's triennial valuation. It's literally a moving target that's subject to review every three years after the Chief Actuary's staff does their thing. Several years ago the target was 4.2%.

Essentially, this is the annual rate, averaged over 75 years, that CPP needs to sustain the current benefit structure at the current employee-employer contribution rate of 9.9%. It's one piece of the puzzle that includes a wide range of other assumptions about employment rates, pay hikes, mortality, etc. See the CPP actuarial report on the Chief Actuary's website.

The Chief Actuary expects CPP contributions coming in to exceed benefits going out until 2020. After that CPP will begin drawing income from the fund administered by CPPIB to subsidize the contribution rate. The draw will start small and rise gradually -- I believe to 30% of income in 2050, but my recollection of that percentage might be a bit low. At maturity, the CPP Fund is expected to cover up to 25% of liabilities while the rest continue to be covered by contributions under CPP's pay-as-you-go structure. (This heavy reliance on the Canadian economy is one of the reasons why CPPIB invests almost all new money outside of Canada.)

The federal and provincial finance ministers (ex Quebec) review CPP funding every three years after the Chief Actuary's office submits its valuation. If the current 9.9% contribution rate -- dubbed the "steady state rate" -- is deemed inadequate, the ministers must increase it, reduce benefits or do both. If they can't agree, an automatic fallback kicks in. I forget the details, but it hits everybody -- including current retirees whose inflation indexing gets reduced.

If I understand correctly, one of the big drivers behind CPPIB's big push into private equity, infrastructure and commercial real estate is the cash flow from active businesses. Their annual report stresses that, unlike typical private investors like Romney's Bain Capital that face redemption deadlines from investors, their long horizon enables them to buy operating businesses as very long-term holds, though -- as shown by their sale of Skype to Microsoft -- they can be induced to sell well before they intended.

BTW, the CPP Fund is now the world's 7th largest pension fund, according to CPPIB.

[Dejavu]

Thank you Bruce for this very helpful inside look into CPP. Very clearly explained.
Dejavu.

[ockham]

Thank you Bruce for this very helpful inside look into CPP. Very clearly explained.
Dejavu.

+1

[ghariton]

CPPIB's 4% real return target is derived from the assumptions used by the Chief Actuary in CPP's triennial valuation. It's literally a moving target that's subject to review every three years after the Chief Actuary's staff does their thing. Several years ago the target was 4.2%.

The more I read, the more it seems that you are right -- literally. The 4% is set as a target because it is the minimum return that will keep the plan viable for seventy-five years with no increase in contributions and no decrease in benefits. If it turns out that the 4% is not reached, then benefits can be reduced, including reducing or suspending indexation to inflation. I have found no details as to the formula for such reductions, just that "this" would happen.

That's interesting. It sounds as if inflation indexing of CPP is a "best efforts" promise and not an absolute guarantee. Given that the promise depends on reaching a 4% real return, those are interesting "best efforts".

On the other hand, politically, reducing indexation of benefits would be extremely difficult. I think that, if a shortfall were to occur, contribution rates would rise instead.

If I understand correctly, one of the big drivers behind CPPIB's big push into private equity, infrastructure and commercial real estate is the cash flow from active businesses.

Yes, that is the impression I get from the annual report.

I found it mildly amusing that, for the fiscal year ending in 2012, the CPP's best performing asset class was inflation-indexed bonds (RRBs) at 16.3%. Non-marketable bonds were second at 14.4%. Real estate was third at 13.0%.

I do wonder how non-marketable bonds are valued. The notes to the report say that, if there is no market for an asset, an attempt is made to find close substitutes that does trade on markets. If that doesn't work, they will look at the value of discounted future cash flows. But that begs the q2uestion: Which discount rate? Unfortunately the report doesn't say. I do see a potential for circularity here.

George

[George$]

IOW this too shall pass. Maybe not in your (or my) lifetime. But most likely over the course of the next several decades of CPPIB's projections.

Anything is possible over a 75-year horizon, I suppose. But my impression is that rates of return have been falling over time. This is not just a short-term phenomenon. It's been going on for some 3,000 years now.

If one believes that, in future, the unput in short supply will be labour, not capital, then that should put downward pressure on rates of return for the next century or so. After that, who knows?

George

YES. I've been wondering about similar possibilities.

In physics, the "g" on our earth, or "conservation of engery", etc are the same for all time and all people (both the poor and the rich).

But the investment returns for affluent in North America for the past 100 years may not remain (like in physics laws) - may not be the same going forward - as the average rate of return for the ever larger number decreases per individual.

[brucecohen]

I do wonder how non-marketable bonds are valued. The notes to the report say that, if there is no market for an asset, an attempt is made to find close substitutes that does trade on markets. If that doesn't work, they will look at the value of discounted future cash flows. But that begs the q2uestion: Which discount rate? Unfortunately the report doesn't say. I do see a potential for circularity here.

I believe the note you paraphrased is the general one. IIRC, there's a note that specifically addresses valuation of the non-marketable bonds. I believe it essentially says they're priced on the same basis as marketable bonds of the same issuing jurisdictions and that price is then "adjusted" (aka reduced) to reflect the illiquidity and the rollover provision.

[brucecohen]

I have found no details as to the formula for such reductions, just that "this" would happen.

From a talk given by the Chief Actuary:

(Slide This leads me to the other side of the coin. What could happen if, in
future actuarial reports, the calculated minimum contribution rate is higher than
9.9%? The default provisions in the Canada Pension Plan Act may result in
adjustments being made to the contribution rate and, perhaps, benefits in
payment if the federal and provincial governments are unable to reach an
agreement in response to the actuarial determination of the minimum contribution
rate. If the new minimum rate is 10.1%, one half of the excess of the new
minimum rate over the 9.9%, that is 0.1%, will apply to an increase in the
contribution rate and the other half will apply to non-indexation of benefits in
payment in order to keep the steady-state rate at 10.0%. In other words, the
contributors and the beneficiaries would equally support the additional cost
shown in the actuarial report.

[pmj]

The wording in that slide is somewhat tentative! Does the legislation & regulations cause any such changes to occur - or does it just allow such changes? IOW, would a political decision be required - and if so what would be the process?

[brucecohen]

The wording in that slide is somewhat tentative! Does the legislation & regulations cause any such changes to occur - or does it just allow such changes? IOW, would a political decision be required - and if so what would be the process?

Here's my understanding, based on what I was told by someone at CPPIB:

-- A political decision is the preferred route. It's up to the "CPP stewards" who meet every three years. The stewards are the federal and provincial finance ministers (ex Quebec). I don't know if their decision requires parliamentary/legislative consent. It probably does technically, but that would be pro forma.

-- If the finance ministers cannot agree on what to do, the CPP reform legislation requires application of the default fix described by the Chief Actuary.

I just checked the actual Canada Pension Plan Act but my eyes glazed over very quickly. I noticed that the word may occurs throughout the text. I think it has a specific legal meaning; perhaps ghariton can explain. I'm guessing that legislation customarily uses may because, in theory, everything's up to the Queen.

[ghariton]

I just checked the actual Canada Pension Plan Act but my eyes glazed over very quickly.

I checked it last night and it certainly is not the most straight-forward read.

I believe that the provision we are looking for is in Section 45(2):

(2) Where any benefit has become payable commencing with a month in any year, the basic monthly amount of the benefit shall be adjusted annually, in prescribed manner, so that the amount payable for a month in any following year is an amount equal to the product obtained by multiplying

(a) the amount that would have been payable for that month if no adjustment had been made under this section with respect to that following year,

by

(b) the ratio that the Pension Index for that following year bears to the Pension Index for the year preceding that following year.

The Pension Index is defined in Section 43(1) & (2):

(1) Subject to subsection (2), the Pension Index for each year shall be calculated, in prescribed manner, as the average for the twelve month period ending October 31 in the preceding year of the Consumer Price Index for each month in that twelve month period.

(2) For any year for which the calculation required by subsection (1) yields a Pension Index that is less than the Pension Index for the preceding year, the Pension Index shall be taken to be the Pension Index for the preceding year.

Tha doesn't allow for reduction or elimination of indexation of benefits. I've also checked the Regulations. While these give more details on the indexing, e.g. the number of decimal places, they don't touch on partial indexation either. (Normally regulations can't overrule the statute, but I've seen it happen, e.g. the Ontario Securities Act.)

A mystery......

the word may occurs throughout the text. I think it has a specific legal meaning; perhaps ghariton can explain. I'm guessing that legislation customarily uses may because, in theory, everything's up to the Queen.

Statutes may say "may" or they may say "shall". The first gives the government discretion to act or not. The second binds the government and forces it to act.

I notice that, in several places, the CPP Act gives discretion to the Minister to act, on the advice of the Chief Actuary. I suppose that the Minister could act contrary to the advice, but he or she would have to justify that pretty strongly.

George

[George$]

Beginning with GeorgeH's post on Thursday (May17) for the G&M report on the CPP - I think both Bruce and and George have raised some basis issues. They really are thoughtful and good at this. I have not followed it all - but I will try and do more so in time.

Here are some issues ... like it reads in the G&M ...

CPPIB chief executive officer David Denison told reporters Thursday the fund is still on track to earn the rates of return it needs to remain sustainable for the next 75 years. The CPPIB needs to earn a “real” rate of return after inflation of 4 per cent annually to meet its payout targets in the future, which currently means average annualized returns of about 6 per cent annually before inflation.

The fund’s 10-year annualized rate of return is 6.2 per cent, but its five-year rate of return is just 2.2 per cent after years of recent turmoil in the markets. Mr. Denison said he is confident the fund will earn the average returns it needs over the long term despite recent short-term results.

For example, what problems me is "75 years" -- "the fund is still on track to earn the rates of return it needs to remain sustainable for the next 75 years". How can this make sense to me?

While assets are growing, everything in a silly way seems "possible". Like a ponzi scheme. But when the payouts income exceed incomes - and annual returns no longer meet their 4% real expectation - how will it reduce or adjust? I don't know. Like ponzi or Madoff the promises are no longer possible.

Will inlationtion increases and so reduced purchasing power result in time - when the average returns no longer are possible?

I don't have the answers. I guess at the moment I feel more unease about it all and what will result.

[brucecohen]

For example, what problems me is "75 years" -- "the fund is still on track to earn the rates of return it needs to remain sustainable for the next 75 years". How can this make sense to me?

The 75 years is simply a statutory endpoint. Every projection needs an endpoint. Parliament, or more likely the civil servants who really do the work, chose 75 years. It's no different than a 25-year-old basing her RRSP planning on the expectation of living to 95. (Actually, it is. The 25-year-old's endpoint is concrete because she will definitely die at some point and so will her non-investment income stream.)

The important point is that it's not a 75-year projection that's frozen. Rather, it's redone at least every three years. And re-dos between triennial valuations occur whenever a CPP benefit change is considered, such as the change in disability rules last year or the year before. CPPIB also reviews the projection on an ongoing basis using a complex CPP cash flow computer model that was developed with help from the Chief Actuary and validated by outside risk experts. A CPPIB boffin told me the model does thousands of simulations, including recreations of past black swans, and identifies portfolio allocations that are likely to perform "the least worst."

FWIW, Waterloo actuarial science professor Rob Brown once wondered why people who view CPP as a ponzi scheme express no concern about the future lack of police, fire, military, education, roadway maintenance/construction and health care services. All are govt commitments funded entirely -- with CPP's exception -- on a pay-as-you-go basis. CPP is exceptional in that, while it is primarily PAYGO, it's the only commitment on the list that has an actual reserve. The purpose of that reserve is to subsidize the future contribution rate to hopefully keep it at 9.9% of covered earnings. If that's not possible, the rate has to be increased and/or benefits have to be cut with the preferred reduction being a reduction in inflation indexing. How is that worst-case any different than the worst-case for a DC pension or RRSP -- notwithstanding that DC and RRSP accounts are more risky in that they're finite pools with a real though unknowable endpoint, a finite flow of contributions and no subsidy from those who die early?

While assets are growing, everything in a silly way seems "possible". Like a ponzi scheme. But when the payouts income exceed incomes - and annual returns no longer meet their 4% real expectation - how will it reduce or adjust?

Based on the current expectation, we'll get some idea in 2021 when CPP is scheduled to start tapping the fund's income. IIRC -- and here I'm running strictly on memory -- the long-term projection is that in 2050 CPP will be drawing somewhere between 30 and 50% of the fund's income. I don't remember the percentage, but the key point is that all the principal accumulated since 1998 will still be untouched...and the first-born members of the boomer cohort will then be over 100.

Will inlationtion increases and so reduced purchasing power result in time - when the average returns no longer are possible?

Normally, inflation occurs only when the economy is expanding and there's demand for credit. If the economy's expanding, some/many/most asset classes should deliver decent returns. ISTM the bigger concern is deflation/depression. In that event, CPP's cash flow from contributions will fall substantially while many/most of its investments suffer. But, on the other hand, it will save a lot on inflation indexing and indexing of the starting pension payment to the YMPE.

[StuBee]

Yes, thank-you Bruce Cohen for sharing your knowledge concerning CPP. You are very reassuring...at least for those who depend on CPP. Is the situation the same for the QPP? Are the different aspects that you describe more or less identical for both plans or are there some fundamental differences whereby QPP is perhaps less trustworthy?

[Shakespeare]

IIRC the Caisse de dépôt et placement du Québec, which manages the QPP money, also has a mandate to promote Quebec investments and suffers for it in poorer returns.

[brucecohen]

IIRC the Caisse de dépôt et placement du Québec, which manages the QPP money, also has a mandate to promote Quebec investments and suffers for it in poorer returns.

Yes. Also, QPP faces serious demographic challenges. The Quebec govt addressed that last year with a bunch of changes. Among other things, for the first time, the QPP contribution rate will be higher than the CPP contribution rate though I don't recall the details.

Note too that the QPP fund took a disastrous hit a few years ago when its managers misjudged the asset-backed commercial paper (ABCP) market.

[StuBee]

IIRC the Caisse de dépôt et placement du Québec, which manages the QPP money, also has a mandate to promote Quebec investments and suffers for it in poorer returns.

Yes. Also, QPP faces serious demographic challenges. The Quebec govt addressed that last year with a bunch of changes. Among other things, for the first time, the QPP contribution rate will be higher than the CPP contribution rate though I don't recall the details.

Note too that the QPP fund took a disastrous hit a few years ago when its managers misjudged the asset-backed commercial paper (ABCP) market.

I am aware of all of these things... My question is whether the QPP has the same mechanisms in place as you have so well described upthread to adress all of the sustainability issues. Do they have the same "statutory end point"? I believe that they do "re-dos" avery three to five years... Do they use the same "independantly validated" computer model with multiple simulations?

The QPP issues that you and Shakespeare have mentioned (including the underperformance), should they not be appropriately dealt with if they use the same "system" as CPP uses. As a consequence, this would be the rationals for the recent QPP ajustments...

[George$]

Thanks Bruce for the response. I will try and get back later - and do it some justice. Other matters at the moment. G

[brucecohen]

Do they have the same "statutory end point"?

Probably. I think a 75-year horizon is common for DB pension funds. I think they do triennial valuations, also common for DB plans. AFAIK, they do not have a portfolio risk mgmt model like CPPIB's.

[ghariton]

I think a 75-year horizon is common for DB pension funds.

I am told (by an actuary) that the 75 comes from the U.S. It was traditional there for many years. U.S. Social Security was first subject to automatic indexation for inflation starting in 1972; before, Congress had authorized increases from time to time on an ad hoc basis. But when they indexed in 1972, they indexed to current wages AND built in an inflation adjustment, thus accounting for inflation twice. As a result, by 1977 Social Security was facing a deficit so large that it would have forced cuts to benefits within five years. The Carter administration set out to resolve the problem, but matters were made much worse by the deep 1980-81 recession. The Readan administration finally came up with a bipartisan solution, but after a lot of argument and compromises. A key element of the compromises was to keep the traditional 75-year horizon, and attempt to solve the deficit within that time span. So 75 was more or less enshrined in the U.S. Canada tagged along.

George

[Bylo Selhi]

Inside the Canada Pension Plan's $153-billion portfolio

Check out this sneak preview from the June issue of Report On Business Magazine... The model for investing that money has gone through a rapid evolution, leaving the sleepy world of government bonds far behind and culminating in something the CPPIB’s managers call “risk budgeting.” All told, CPPIB has emerged as one adventuresome investor.

It’s all run by investment professionals like Wiseman and Bourbonnais, guys who, for some mysterious reason, think it’s more rewarding to invest on behalf of the Canadian public than make far more dough on the other side of the Street. As a Bay Street professional myself, I want to know what makes these guys tick....

[brucecohen]

RE: When to start taking CPP (and OAS when its deferral option becomes available)

Here is a short paper published by the Boston College Center for Retirement Research. While it deals with US Social Security, the logic is applicable to CPP.

The author says the decision to delay taking benefits amounts to buying an annuity from the govt plan...at a rate much more favourable than those on annuities from insurers. The govt annuity lacks a number of the overhead costs an insurer has to include in its pricing. Also, the "pricing" on the govt annuity is based on average longevity while insurers use above-average longevity.

Bottom line: Now that taking CPP does not exempt you from having to pay contributions on new employment/self-employment income, it pays to wait unless you're longevity-challenged.

[Insomniac]

RE: When to start taking CPP (and OAS when its deferral option becomes available)

Here is a short paper published by the Boston College Center for Retirement Research. While it deals with US Social Security, the logic is applicable to CPP.

The author says the decision to delay taking benefits amounts to buying an annuity from the govt plan...at a rate much more favourable than those on annuities from insurers. The govt annuity lacks a number of the overhead costs an insurer has to include in its pricing. Also, the "pricing" on the govt annuity is based on average longevity while insurers use above-average longevity.

Bottom line: Now that taking CPP does not exempt you from having to pay contributions on new employment/self-employment income, it pays to wait unless you're longevity-challenged.

Thanks, Bruce. Very interesting.

[steves]

Here is an example..... our guy is 60, retired, and has a $500K RRSP.

He can opt to take his cpp at 60 and oas at 67, or he can delay both to age 70.

The graph shows his capital running out at age 90 in the first case, and lasting to age 100 in the second case. I don't know why, but most individuals I talk to opt for early CPP. Myself included.

Early or late CPP/OAS?

[AltaRed]

I don't know why, but most individuals I talk to opt for early CPP. Myself included.

It is simple. A bird in hand is worth two in the bush. We might all hope to live to a healthy 90 or 95 but the actuarial tables say otherwise. I took mine early too with no regrets.

[CROCKD]

It is simple. A bird in hand is worth two in the bush. We might all hope to live to a healthy 90 or 95 but the actuarial tables say otherwise. I took mine early too with no regrets.

+1