Phased Retirement The CPP Experience La retraite progressive l expérience du régime de pensions du Canada (RPC)

Transcription

1 JUNE 2011 ANNUAL MEETING OTTAWA (SESSION 16) 1 Session 16: Séance 16 : MODERATOR/ MODÉRATRICE : SPEAKERS/ CONFÉRENCIERS : Phased Retirement The CPP Experience La retraite progressive l expérience du régime de pensions du Canada (RPC) France Panneton Yu Cheng Sari Harrel?? = Inaudible/Indecipherable ph = phonetic U-M = Unidentified Male U-F = Unidentified Female Editor s Note: we are sorry but the audio recording of the opening of this session is unavailable due to technical reasons. Speaker Sari Harrel:... and expenditures over a long period of time. For the demographic projections, we start with a difference of the populations of Canada and Québec to which are applied many demographic and economic assumptions in order to derive the projected labour force, number of earners, contributors and beneficiaries. As shown here, the number of contributors per CPP pensioner is projected to decrease in the future to a level of below two as the population ages from both a lower placement level fertility rate and continuing increases in life expectancy. The aging of the population is also evident from the increase in the pay-as-you-go rate over time. However, from 2000 to 2020 the net cash flows of the plan, that is contributions less expenditures, have been and are projected to be positive, leading to a rapid increase in the AE ratio shown in the last column. The net cash flows of the plan are invested by the CPP IB to further increase the level of assets. The AE ratio is projected to stabilize at a level of about five. That is assets are projected to cover about five years worth of expenditures over the long term. By 2020 the plan s assets are projected to reach a level of $275 billion. After 2020 net negative cash flows or shortfalls are projected. However these will be covered by a proportion of investment income. By 2050 a projected 29 percent of investment income will be needed to cover the shortfall. At that time, investment income will represent 27 percent of all revenue. The minimum contribution rate or MCR for the plan is just the sum of the steady state and incremental full funding rates. The MCR is thus the lowest rate that is sufficient to sustain the plan with respect to its two financing objectives. Under the 25 th Report the MCR is 9.86 percent until 2022 and 9.85 percent thereafter. The full funding rate of.02 percent until 2022 and.01 percent thereafter applies to the enhanced eligibility for the disability benefit for long-term contributors brought in In summary, despite the projected substantial increase in benefits paid as a result of an aging population and the retirement of the baby-boomers, the plan is expected to be able to meet its obligations throughout the projection period and to remain financially sustainable over the long term. I would like to conclude with a very important aspect of the valuation process for the CPP, specifically the independent peer review of the actuarial report. The external peer review process has been in place now for over 10 years and has generated valuable recommendations that have helped improve the quality and transparency of the reports. PROCEEDINGS OF THE CANADIAN INSTITUTE OF ACTUARIES Vol. 42, June 2011

2 2 JUIN 2011 ASSEMBLÉE ANNUELLE OTTAWA (SÉANCE 16) To further increase the transparency and independence of the process an external party has been responsible for the oversight of the peer review for the past three reports. The United Kingdom Government Actuaries Department, UKGAD, both selected the panel of independent Canadian actuaries to perform the review and provide an opinion on the work done by the reviewers. The most recent panel support provides opinions and recommendations on all major aspects of the 25 th Report as well as regarding the Supplementary 24 th Report. Among the opinions the reviewers found that the assumptions used were individually and in the aggregate within the reasonable range and that the methods employed were reasonable as well. UKGAD, in its opinion, determined the adequacy and reasonableness of the work done by the reviewers. The opinions and recommendations of the reviewers are all given due consideration by the OCA and are discussed with the plan s federal, provincial, and territorial stakeholders before any action is taken for future valuation reports. The next valuation date is set as at the end of December 2012 with the report to be completed and tabled before Parliament before the end of Thank you. We will now continue. (Applause) Speaker Yu Cheng: Good afternoon. I will continue to show you some stats on the CPP working beneficiaries. From the start, I would like to note that these statistics I m showing you are based on the 2005 experience. This is only a small subset of the work we ve done for the 2007 through 2009 CPP tri-annual review and the 24 th Actuarial Report. At that time only 2001 to 2005 data from CRA were available. In 2011 we just received another three years of data for 2006 to However, due to time constraints we didn t do any update on the statistics. At that time when we did the 24 th Actuarial Report we did analyze the data for year 2001 as well. And I will touch on some of those statistics. However, given the limited number of years, we didn t infer any trends. I will discuss the following characteristics of working beneficiaries, first by age and sex, and then by region, by level of pension and earnings, by employment status and sources of income. I will then present the projections of the 25 th CPP Actuarial Report regarding working beneficiaries. First, we ll start with age and sex. The upper table shows the number of CPP working beneficiaries in 2005 by age and by sex. The lower table shows the proportion of working beneficiaries, which is the ratio of the working beneficiaries to the total number of CPP retirement beneficiaries by age and by sex. For example, one quarter of the CPP retirement beneficiaries aged 60 to 64 were still working in The proportion decreases with ages, but it doesn t mean that absolute number decreases. For example, for ages 65 to 69 the proportion is lower, is at 17 percent, but in real numbers it s more than the age 60 to 64. Or 11 percent of CPP retirement beneficiaries, or, in other words, 329,000 were working beneficiaries in As a comparison, in 2001 the proportion of working beneficiaries to the total CPP retirees was 9 percent. By sex it s 12 percent for males and 7 percent for females. If we only look at those two years, there is a slight trend. But because we only have five years data, so we didn t infer any trends in our actuarial report. If we look at these two pie charts, there are more males than females working beneficiaries, 64 percent for males and 36 percent for females. The distribution by sex for all retirees is about 50 percent each for males and females. Forty percent of working beneficiaries are age 65 to 69, 32 percent are age 60 to 64, and 28 percent are age 70 and over. And next, if we look at regions, 46 percent of working beneficiaries are in Ontario and 44 percent in the western provinces, which includes Manitoba, Saskatchewan, Alberta and British Columbia. In Western Canada working beneficiaries makes up 9 percent of the population age 60 and over. The corresponding percentage for Ontario and Maritimes is 7 percent each. Vol. 42, Juin 2011 DÉLIBÉRATIONS DE L INSTITUT CANADIEN DES ACTUAIRES

3 JUNE 2011 ANNUAL MEETING OTTAWA (SESSION 16) 3 This slide shows the proportion of working beneficiaries by age and region. Western Canada has the highest proportion for each age group. Thirteen percent of all ages combined, Ontario follows with 11 percent, and Maritimes stand at 10 percent. These percentages are referred to the working beneficiaries or the total number of retirement beneficiaries of CPP. The first slide is the working beneficiaries over the population age 60 and over. The proportion decreases with age. The Maritimes have the biggest decrease from age 60 to 64, to age 65 to 69. Next, we will look at the working beneficiaries by level of pension and earnings. This slide compares the average retirement pension of working beneficiaries with non-working beneficiaries as a percentage of the maximum pension by age and by sex. Non-working beneficiaries are defined as working beneficiaries who stopped working after taking up their retirement benefits. Working beneficiaries have higher pensions, which means that their career average earnings were higher. This is especially so for female working beneficiaries with average pensions over 10 percentage points higher than for nonworking beneficiaries, which is 58 percent versus 47 percent. The reason behind this could be the white collar effect, which means the higher educated person would be working longer at older age, because those jobs are not labour intensive and they re professionals like accountants, actuaries, and they can possibly work at older ages. Know that female working beneficiaries represent about 36 percent of total working beneficiaries. And male working beneficiaries represent 34 percent, compared to non-working beneficiaries, distribution is males at 49 percent and females at 51 percent. Sixty percent of working beneficiaries have pension levels between 75 percent and 100 percent of the maximum. About 30 percent get a maximum pension. The higher the pension, the higher the proportion of working beneficiaries. Pension levels at 0 to 25 percent represents 3 percent of male working beneficiaries, 25 to 50 percent represents 7 percent. And 50 to 75 percent represents 16 percent, and 75 to 100 percent represents the remaining 74 percent for males. Pension levels at 0 to 25 percent represents 17 percent of female working beneficiaries, 25 to 50 percent represents 24 percent, and 50 to 75 percent represents 25 percent, and 75 to 100 percent represents 34 percent of female working beneficiaries. You can see there is a significant disparity between males and females. Only 34 percent of females have a pension of more than 75 percent of the max, compared to 74 percent of males. This slide shows the average post-retirement total employment earnings of working beneficiaries. I will just point out that if you printed out the slides two days ago, this is an updated version, so please note that. Employment earnings after retirement take-up include both employed and self-employed earnings. And this is the total employment earnings of working beneficiaries as a percentage of the Y&P by level of pension age and sex. Males have higher earnings than females for all pension levels combined. The higher the age the bigger the disparity is between males and females. This is also true for the earnings level between 75 to 100 percent of the Y&P. This could be due to the cohort effect; females of older generation work less, the gap will be reduced when the younger cohort works more and have higher education. For each pension level, the higher the age the higher the employment earnings for females. This is not the case for males. This slide shows the average post-retirement total employment earnings of working beneficiaries by level of pension and age for both sexes. The higher the age, the higher the total employment earnings. And the higher the pension levels, the higher the employment earnings. And next we will look at the characteristics by employment status. While a person has both employed earnings and selfemployed earnings, we define a person is employed if the employed earnings is greater. Here we can see more employed than self-employed, 71 percent employed versus 29 percent self-employed. In the regular contributors, which I mean is not working beneficiaries, but as a regular contributor between age 18 to 65, it s 90 PROCEEDINGS OF THE CANADIAN INSTITUTE OF ACTUARIES Vol. 42, June 2011

4 4 JUIN 2011 ASSEMBLÉE ANNUELLE OTTAWA (SÉANCE 16) percent employed versus 10 percent self-employed. You can see for the working beneficiaries, the self-employed percentage is much higher. Females are 77 percent employed and 23 percent self-employed, males are 68 percent employed and 32 percent self-employed. This chart shows the distribution of male working beneficiaries by employment status, level of pension and age. For instance, that at a pension level 0 to 25 percent for age 60 to 64 the sum of employed and self-employed is 100 percent. The proportion of working beneficiaries employed decreases with age, while the self-employed proportion increases with age. This is true for every level of pension for males. And this slide is the same chart, but for female working beneficiaries. For females, the self-employed portion increases more significantly with age for pension level below 50 percent. It does not increase as much by age for pension level above 50 percent. And next we ll look at the characteristics by sources of income. This slide presents the sources of income of working beneficiaries, total employment earnings for both employed and self-employed earnings represent almost 50 percent of all income or in other words about $35,500 per year on average. In comparison, the CPP pension is about $6,500 per year average or about 9 percent of the total income. Private pension income increases with level of the CPP pension. For ages 60 to 64, 3 percent of income is from private pensions for those receiving 0 to 25 percent of the max of CPP pension. This compares to 23 percent for those receiving 75 to 100 percent of the maximum CPP pension. RRSPs are level at about 3 percent of total income for ages 60 to 64 and 2 percent for ages 65 to 69. The proportion with RRSPs increases by level of CPP pension from 9 percent at 0 to 25 percent of the maximum to 20 percent at 75 to 100 percent of maximum. For ages 60 to 64 and similarly for ages 65 to 69, 9 percent for 0 to 25 percent of the max and 18 percent for 75 to 100 percent of the maximum CPP pension. Around 50 percent of working beneficiaries aged 60 to 64 have investment income and at 60 percent for people at ages 65 to 69, and these proportions are relatively stable across the four pension levels. This slide presents the average income of working beneficiaries aged 60 to 64 by sources of income and level of pension. Other age groups have similar trends. The higher the level of pension the higher the employment earnings. Also private pensions increase with the level of CPP pensions. RRSPs and investment income are somewhat stable. This slide compares the proportion of the working beneficiaries who receive the guaranteed income supplement, which is GIS versus the general population. In total, about 8 percent of working beneficiaries receive GIS compared to 36 percent of the general population. Compared to the general population, there is a much lower proportion of working beneficiaries receiving GIS at all ages above 65. There are more males than females in receipt of GIS in each age group of working beneficiaries. This is opposite when compared to the general population, where 31 percent of males and 42 percent for females that receive GIS. Moreover, the differences between the proportions of the sexes are more significant in the general population than for working beneficiaries. And next, I will show some results for the working beneficiaries projected in the 25 th CPP Actuarial Report. We assume the retirement take-up rates will be affected by the new early retirement adjustments and removal of the work cessation test. The retirement take-up rates at age 60 assumed to increase, peaking in year 2012, reflecting both anticipation of the higher early retirement reductions and removal of the work cessation test. After 2012, early pension take-up rates assumed to decrease and it s facing four five years and effect of removal of cessation test will be diminished. Ultimately, the long-term rate will go back to the same level as 2010, higher pension from the increase in general dropout and post-retirement benefits from working beneficiaries. And in the report we assume only 50 percent working beneficiaries will continue to contribute between ages 65 to 69. Because contributions are voluntary for those ages and there is uncertainty regarding the future behaviour of beneficiaries, so that s why we assume only 50 percent of working beneficiaries will continue to contribute. Vol. 42, Juin 2011 DÉLIBÉRATIONS DE L INSTITUT CANADIEN DES ACTUAIRES

5 JUNE 2011 ANNUAL MEETING OTTAWA (SESSION 16) 5 And this is the slide that s showing the retirement take-up rates. You can see an increase and then peaking 2012 for age 60 due to the anticipation of the higher early retirement reduction rates and also the removal of the work cessation test. And then gradually, revert back to the 2010 level by This slide shows the assumed proportion of working beneficiaries from 2012 onward. The proportion at age 65 to 69 already take into account the 50 percent assumption for our voluntary contributions. We did not assume any trends due to the uncertainty in behaviour after the implementation of the BOC 51 (ph). We will monitor the characteristics of the working beneficiaries closely. This slide shows the projected number of working beneficiaries and the post-retirement benefits for two scenarios. The first scenario assumes that 50 percent working beneficiaries aged 65 to 69 contribute, and the second scenario assumes all will contribute. Now, the conclusion. So from the stats, we know there are more male than female working beneficiaries; 64 percent versus 36 percent, and more than 70 percent working beneficiaries are under age 70, 90 percent of working beneficiaries are in Ontario and Western Canada, 60 percent of working beneficiaries have pension level 75 to 100 percent of the maximum CPP retirement benefits. Significant disparity between males and females. Seventy percent of working beneficiaries are employed, 30 percent are self-employed. Employment earnings make up almost 40 percent and the self-employed makes another 10 percent, so in total it s about 50 percent of the income of working beneficiaries. Share of employment earnings and private pension income increases with pension level. Number of working beneficiaries projected to increase from 320,000 in 2012 to 430,000 by The post-retirement benefits represent about 1.7 of total expenditures ultimately. And this is the end of the session. Thank you, everyone. (Applause) Moderator France Panneton: Well, thank you both, Sari and Yu. There were a lot of numbers and I m glad that the presentation is available on the website. And, to me, there were a lot of statistics that were very intuitive. That self-employed are working beneficiaries as opposed to more of the employed type. But some of the other statistics were not as intuitive. I didn t realize that people out west were working more than in Ontario, so it was very interesting. We had some conversations also with the folks representing the Québec Pension Plan and although they weren t able to come and present, their results were similar and their numbers were really showing that over the last 10 years there was a real clear trend that there was more and more working beneficiaries. And with respect to the Québec Pension Plan, and phased retirement, which is the reduction of 20 percent of your workload, there was really a clear trend over the last five years that more and more new retirees are asking to do a reduced workload and start receiving their QPP. And the numbers were I think about 10 percent of new retirees were asking to get the QPP and continue working. And I know it would have been very interesting to hear from plan sponsors or from the regulators what s really the pickup rate with respect to phased retirement. When we look at the CPP and the QPP we see that people do want to continue to work and start receiving their pension. But how does that affect the registered pension plan? And unfortunately, as of now, there s not a lot of people that are able to do so, either because the plan sponsors are not offering it or because of various reasons. And so maybe in a few years we ll be able to have a lot more statistics on the number of people in registered pension plans that are able to use the phased retirement. I can think of a few people that I know that are doing phased retirement, but it seems that these are more like in the government agencies. I think sometimes the government agencies are at the start of the wave with respect to those features. PROCEEDINGS OF THE CANADIAN INSTITUTE OF ACTUARIES Vol. 42, June 2011

6 6 JUIN 2011 ASSEMBLÉE ANNUELLE OTTAWA (SÉANCE 16) So I don t know if there was any questions from the audience with respect to the presentation for the CPP or any thoughts on what registered pension plan sponsors should be doing with respect to phased retirement or what will be the trigger? Will it be maybe the demography on the economic -- when those two arrive and when all the baby boomers are retiring, that s when plan sponsors might be focusing more on phased retirement perhaps. Any comments or questions? U-F: I don t know if I saw correctly, but you re saying that the early retirement rate will come back after a peak in 2012 and then they would come back to their regular level. I would expect that they would, you know, people would retire about Moderator Panneton: Excuse me, can you come to the microphone? U-F: I don t know if I saw correctly, but on one of the slides you were saying that the early retirement rate would peak in 2012 because of this change to the subsidy and then they would come back, yes, to exactly the same level, in fact, as they were in I would expect that people would retire later in those years. Speaker Cheng: Well, the thing is at the beginning, in 2012, when the reduction factor just started because there s gradually an increase of the reduction factor from 0.5 to 0.6 percent per month, so the people in 2012, that s the first year, so people will just rush into starting their early retirement. U-F: Yes, for sure, that I understand. But after... Speaker Cheng: But after that then the people will kind of get used to it and then they re like because the removal of work cessation test, so also the people may want to, OK, retire early and keep working because they get the PRB, the post-retirement benefits, recalculated year by year. They may choose that, so that s why we said, well, they might come back to the original level as So, yes, the increase of the actual reduction factor from 0.5 to 0.6 it really wants to push people to retire a little bit later, but U-F: Which is not reflected here. Speaker Cheng: which, after you consider the removal of the work cessation test and the PRB, which may lead people to come back, retire and then work. U-F: And the other one was a projection until 2030, I believe, in terms of working beneficiaries at the end. And I thought the number was not increasing that much from now to then. But then again, I didn t do the analysis or the math, but I would expect that we would have more working beneficiaries, the increase would be substantial in the next few years. I think it s the last slide the projection in the 25 th Report. Speaker Cheng: Here, because we assume a flat proportion of working beneficiaries as the total CPP retirees and this, we only assume, is 50 percent of the working beneficiaries will contribute at age 65 to 69. So that s why you see it s increased from 320,000 to 490,000. But if you look at U-F: (off mic) That increase, I would expect that it would be more than that in like over 40 years. So you re not expecting to have more working beneficiaries (?? 35:50)? Speaker Cheng: How the numbers increase depends on the distribution of the total CPP retirees by age. Because if you have more at younger age, between, say, age 60 to 64, then you have more working beneficiaries contributing. But if you have more people at age 65 to 69, then there are less. And for people over age 70, although they can work, but they don t need to contribute. So this is not the projection of the working beneficiaries, but this is really the number of working beneficiaries contributing to CPP. So there is a big portion, if they are over age 70, they don t need to contribute. Vol. 42, Juin 2011 DÉLIBÉRATIONS DE L INSTITUT CANADIEN DES ACTUAIRES

7 JUNE 2011 ANNUAL MEETING OTTAWA (SESSION 16) 7 Jerry Loterman: Jerry Loterman of Aon Hewitt. I was just wondering I guess it s a question for Sari what was the actuarial basis used for coming up with the new early retirement and late retirement factors with like interest, mortality, that sort of thing? I don t know if you were involved in that, but... Speaker Harrel: As I said, the factors were determined using something called the Steady-State Rate Contribution Method. And actually, this goes back to a study we did. A lot of work was done on this a number of years ago. If you go to our website you can look up actuarial study number 2 I believe, it was published in And it goes through detail, the various methods that were considered to determine the factors. So I would respectfully refer you to that study. The factors for this case, the specific basis, I would have to defer either to my co-worker here or my managers in the audience to go through the details of all the assumptions in setting those. U-M: No, we don t have all that much time. And I think that and I could be totally mistaken about this but I seem to remember the CIA had some criticisms of the factors in... I don t even remember what exactly the rationale was, but how do you justify, or did you justify, what happened versus what the CIA was saying? Speaker Harrel: We did discuss that, but I m not sure I m at liberty to discuss our actual response to that. That was discussed within the office. But that was based on... That was a little while ago and that was before the changes came into effect for the QPP now. Now the situation has changed, the conditions have changed. So the analysis done in that paper was based on assumptions about other changes that could be made for the QPP. But now the situation has changed, so there s that aspect as well. Bruce Michael: Bruce Michael. Along a similar line, you indicated that the factors were calculated on an aggregate basis as opposed to on an individual basis. I wonder if you could just go into that a little bit more in detail and what the implications of that were. Speaker Harrel: Michel? Michel Montambeault: Michel Montambeault of the Office of the Chief Actuary. In respect of that question about the CIA there you have to remember that at that time Québec was proposing to change its benefit formula to divide by 40 to make the contributory period more fair between people taking it at age 60 or age 65. So in that situation the.05 percent could have been thought to remain being a factor that is neutral. So for the CPP, instead of having to change the formula, which would have been a more complex issue to explain to people, the decision to change the factor seemed to be the easiest way to rectify the situation which is related to the existing subsidy that was existing prior to age 65 where the benefits were not reduced enough to be fair to the plan. And what is the method? Like I said, we should definitely refer you guys to that study that we did in It took many, many years before the CPP committee was in a position to make that change under the CPP regarding the factors. It s been a long time that we produced information that said that those factors might have been unfair, but they had other decisions to make also and different reviews prior to the latest one. The method is simply we are not looking at individual perspectives. We are looking at the plan as a whole. So the way the plan is financed is considered and this is the steady-state method where we look at stabilizing the funding ratio over 10 years and 50 years, 60 years after the evaluation date. So that s the basic steady-state financing approach. So we take that into account and also all the benefit provisions. So we make sure that whatever age people would take the benefit, that the resulting minimum contribution rate or steady-state rate would remain relatively the same. And this is how the factors are determined, so that is the basic nature of it. So it remains fair for everybody, because whatever people would have to pay, if there was only one age possible under the plan the answer for the contribution rate would be the same, whatever age it would be, 68 or 65. So this is the methodology simplified that I m saying here, but this is the basis for the final answer about what are those factors. And those factors of course are not the exact, they re approximate so that it s even easier to administer and explain to people. PROCEEDINGS OF THE CANADIAN INSTITUTE OF ACTUARIES Vol. 42, June 2011

8 8 JUIN 2011 ASSEMBLÉE ANNUELLE OTTAWA (SÉANCE 16) But they re definitely much fairer in general because they kind of follow more the exponential nature of those factors. In the past it was just.05 before and after, now it s a little bit more when... The more you wait the more you get. So this was one also of the selling points for the choice of those factors. Kim Young: Kim Young. Did you take into account in your working beneficiary projection that there s a decline in the private coverage of pension plans in determining how many? There may be a need for people to work more in the future to supplement their CPP than they re getting from private pension plans now. Speaker Harrel: We did not explicitly take that into account, no. However, I would add that our projections are based on... I m looking at long-term trends as well as recent experience. So in that aspect we do take many, many variables into account. But we did not explicitly make an assumption about what you re referring to. Moderator Panneton: Any more questions or comments? Well, thank you for attending this session and thank you, Sari and Yu. (Applause) [End of recording] Vol. 42, Juin 2011 DÉLIBÉRATIONS DE L INSTITUT CANADIEN DES ACTUAIRES