Session PD-16: Future of Defined Benefit (DB) Pension Plans Séance TR-16 : L avenir des régimes de retraite à prestations déterminées

Transcription

1 Future of Defined Benefit (DB) Pension Plans (PD-16) 1 Session PD-16: Future of Defined Benefit (DB) Pension Plans Séance TR-16 : L avenir des régimes de retraite à prestations déterminées Moderator/Modérateur: Speakers/Conférenciers: Jerry N. Loterman Stephen P. Bonnar Paul G. Forestell Moderator Jerry Loterman: Welcome. Hi, my name is Jerry Loterman; I m currently a pension consultant at Hewitt Associates, having spent many, many years in the actuarial industry and primarily in the pension industry. Our topic today is the Future of Defined Benefit Pension Plans. I m sure a topic that s near and dear to everybody s hearts. Our speakers today are Steve Bonnar and Paul Forestell. Now, amongst Paul, Steve and I we have over 70 years in the actuarial business; so a couple of years there. Steve is currently a consultant with Towers Perrin; has been there for almost his whole career. He concentrates on retirement systems, consulting, and also does a lot of asset consulting, primarily in the asset liability modelling area. Paul Forestell graduated from the University of Toronto and has been with Mercer his whole career; been there almost 15 years and is currently the leader of the Canadian Retirement Professional Group. All three of us were on PPFR a few years ago, together at the same time and that s a little bit of background. At breakfast this morning I was speaking to Brian FitzGerald and Tom Levi around the breakfast table and they ve actually been around longer than Steve and I for sure and Brian mentioned that when he first got into the actuarial business everybody was converting from Defined Contribution (DC) to DB plans. What were the reasons? The reasons were because DC plans were providing inadequate benefits. The rates of return were quite poor and generally there was a lot of dissatisfaction with DC Pension Plans. Tom mentioned that back in the 50s and 60s the DB plans that existed were primarily funded by Group Annuities and as Tom has mentioned a couple of times today they were very well funded, there were no unfunded liabilities because you basically bought the benefit that you paid to the insurance company and that s the benefit you got at the end of the day. Very well funded but I think the industry realized that those weren t the best models for continuing into the future. So we really went into a golden age of defined benefit pension plans probably in the 60s and 70s, and probably midway through the 80s, at least in Canada and lots of new plans were created during that time. Of course with the 80s came pension reform, provincially and then pension tax reform at the federal level in the early 90s and I think we started to see no new plans really being created after that and a very slow decline in existing defined benefit pension plans and coverage. So as David Blitzstein said recently in a pension section news article, our tragedy is in the unravelling of the defined benefit pension system. It looks like we ve come full circle. We started with DC plans back in the 60s and 70s and then converted to DC and now it looks like we re going back again.

2 L avenir des régimes de retraite à prestations déterminées (TR-16) 2 So without much further ado we re going to have our speakers, Stephen and Paul. Steve is going to be talking about how we got to where we are today, what some of the reasons are for the decline. Basically some key messages about what s going to happen with DC plans, what s been happening with DC plans? Paul will continue with some projections into the future, based on current trends with some suggestions for all stakeholders to work together We ll have about 25 minutes for each speaker and then questions afterwards. Speaker Stephen Bonnar: Thanks, Jerry. What I m going to do, as Jerry mentioned, is take you through a little scene-setting, how did we get to where we are today and then present a potential vision of the future. A vision that, in my mind, is likely to occur if the status quo continues and that s a vision of largely DC arrangements. Now, as I thought about this I kind of felt like I fell into Charles Dickens, A Christmas Carol and I m playing the part of the ghost of Christmas Future. This is a discussion of a potential future outcome, not a guaranteed future outcome. Whether it becomes the actual future outcome or not depends on the actions of all stakeholders in the pension system; actuaries being just one of those many stakeholders. So with that, let me just take you back and go through some of this. So a lot of this here is not new news. We ve got some relatively serious problems with the pension system. I m going to show you some statistics on the next page but at this point I m just mentioning that pension coverage is decreasing and why is that happening? We ve got a very complex regulatory framework. My guess is that most of you in the audience work in the system and so you know that we have this Byzantine constitutional structure that has many masters, if you will, all of the provinces, as well as the feds. We ve got a fair lack of flexibility in pension rules and I m seeing that from the perspective of a pension consultant. I m not sure you d necessarily have the same reaction from a pension regulator or supervisor. But, you know, a lot of the provisions that we have in pension legislation were put in place for good and valid reasons but they don t necessarily move with the times or keep pace with the times and so that s what I m talking about in terms of lack of flexibility. The plans that were designed in the 50s, 60s and 70s and that were established in a regulatory environment in the 50s, 60s and 70s have difficulty in the current environment where circumstances have changed dramatically. We don t have single employers for our whole working career, notwithstanding me as an example, being with one employer for virtually my whole career, at least hopefully for virtually my whole career. We ve also got alternatives right now that avoid pension regulation. You ll see some of that in the stats, that group or RSPs in general, group RSPs as employer sponsored arrangements appear to be growing in their use as a retirement tool for employees. It s difficult to actually get information on those statistics because a group RRSP is nothing more than a collection of individual RRSPs so you can get all kinds of information from Stats Canada on RRSPs, you just don t know how many of them are employment related and how many of them are up to the individual. We ve got employer perceptions that DB plans are too risky. You know, if they heard, it would get some of my clients mad at me but I m going to recall that back in 1998, 1999 and 2000 when these plans were just as risky but the risk was on the good news side and employers in general weren t really concerned about that. And the perceived value, the value that individual employees, individual participants in these plans perceive of them is relatively low. And so if you think of it from a total compensation perspective, the amount of pay they re prepared to give up for a defined benefit pension plan is relatively low and that s not because the value of a DB plan is that small. It s because their perception of that value is quite small. Délibérations de l Institut canadien des actuaires, Vol. XXXVIII, n o 1, octobre 2006

3 Future of Defined Benefit (DB) Pension Plans (PD-16) 3 Given an employee s perception is relatively low, employers feel difficulty in making that compensation trade-off of the true value of the plan. And so while DB plans have certain very good HR management aspects to them, employers more and more are saying that those advantages aren t sufficient to offset the negatives, the uncertainty of costs, the regulatory environment and the like. So we ve got a gradual process in Canada, moving away from DB to DC. Not necessarily DC Pension Plans but to DC arrangements. And I think of that in a broad scheme of the word because the vast majority of very small employers that have no pension arrangements at all implicitly are placing it on their employees to do their own saving for retirement. Well, that s kind of the ultimate in DC. Lack of member understanding of the value, I talked about inappropriateness of the traditional DB design to today s workforce. That s not a comment about a DB not being appropriate to the current environment but the way plans were designed they were really targeted for single career employees, basically delivering significant values to those that retired from the employer and relatively modest value to those that terminated from employment prior to retirement. In an environment where the expectation was that you were going to work for virtually your full career for one employer, that made a lot of sense. In an environment like today where that s just not the reality, it no longer makes a lot of sense. I talked before about cost uncertainty of DB plans. Again, when you think of cost uncertainty the problem exists when the costs go up, less so when the costs go down. When I originally wrote this chart we had proposed changes in accounting rules in the U.S. We have now finalized changes or at least the first phase of accounting changes in the U.S. that will in effect bring the surplus or deficit in a pension plan onto the sponsor s balance sheet. In an announcement from the Canadian Accounting Standards Board that they expect to follow suit fairly quickly, next year in fact. That will have very material effect for a number of plan sponsors because just the way the mechanics have worked the many plans in Canada have prepaid pension expense on the balance sheet as an asset on the sponsor s balance sheet in respect of plans that are in deficit. When you wipe out that asset and further take the deficit onto the balance sheet that s going to have a material effect on shareholders equity. Now, the process in Canada has been fairly slow relative to the U.K and the U.S. In the U.S, for the last two years, at least, we see monthly, if not weekly, headlines of another large sponsor shutting off the DB plan at least, to new entrants and in many circumstances just shutting it off for existing employees as well. In the U.K. it s been more of a closing of plans to new entrants. I m not sure exactly what the proportions of plans is that are still open to new entrants but it s really quite small. Canada hasn t quite followed that fast of a pace. So, with that, what have we actually seen? We ve seen reduction in number of pension plans and reduction in coverage. The numbers I ve got on the slide here show the numbers of pension plans. If you wonder why I m just looking at pension plans with more than ten members, the information is coming from Stats Canada and they got more aggressive at collecting information for very small plans partway through the period. So the numbers on plans with less than ten members was not consistent throughout the period. So if you look at public sector plans with more than ten members, they have grown, about a 25 percent increase over the 12- year period, from 1992 to But in spite of that increase in the number of plans the percentage of the public sector that is covered by DB plans has dropped from, roughly, and I m rounding quite generously, roughly 90 percent to about 80 percent. Now, part of that drop may be because of more contract employees in the public sector who don t have DB coverage. In the private sector though the number of plans has decreased by almost 15 percent, from 9,800 to 8,400 over that period of time, from 1992 to 2004 and the coverage in the private sector has dropped more dramatically than that.

4 L avenir des régimes de retraite à prestations déterminées (TR-16) 4 In 1992 almost 30 percent of the private sector was covered by DB plans and by 2004 that had dropped to a little over 20 percent in that period of time, so significant reduction in coverage. So what we re seeing is that for those who are going to be retiring in the future, a good chunk of their pension income will not be coming from DB plans and that s going to be a material change. We looked at tax return information and I think that s from 2004 but it may be slightly earlier than that, for retirees and we looked at people in the age range 65 to 70, we couldn t really tell whether they were retirees or not but they had relatively low employment income. But the piece of the pie for that group comes from pensions is 50 percent for those in the 30,000 to 40,000 range, that s income in retirement and over 70 percent for those in the 70,000 to 80,000 range and most of the balance is made up of government pensions and only very small amounts coming from RRSPs. That s in spite of RRSPs having been around for 30 plus years at this point. So that s a little bit of scene-setting. If we say, okay how do we move from here? What are some principles that we would like to see in a pension system? I m just going to take a short while and go through these things. But if you think, okay what would I like from a pension system? I d like to have three strong participants or stakeholders in a pension system. I d like to have the employer involved, I d like to have the employee involved, or the individual involved, and I d like to have the state involved as well because it would be useful to have some form of a broad safety net for pensions. That sharing of responsibilities among those three stakeholders is kind of important and how we ve evolved from where we started to where we are today is that the historical design of DB Pension Plans in Canada was a pretty generous contributory define benefit pension plan and it basically dealt with, for career employees, all of their needs in retirement, over and above what they were going to get from government pensions. That s really changed quite dramatically. Even setting aside the movement from defined benefit to defined contribution, you look at the evolution of the defined benefit design. More and more there s been a removal of employee contribution requirements, together with a reduction in benefit level so pushing back onto the individual more responsibility for the retirement savings. If you look at those organizations that have actually moved from DB to DC and what level of benefits provided from the DC side, you re going to see, again, lower still, much lower still, levels of income that you could expect to come out of the DC plan in retirement. And I ll take you through some specific numbers on that later on in the presentation. So what does the world look like if we actually do move completely to DC? And remember, this is a presentation of what the world might look like as opposed to what the world will look like, I hope. Let s think of what are the elements that affect an individual s ultimate pension; when they start contributing. There are a number of groups in the audience, those in the audience, people with hairlines like mine and there are those in the audience with much nicer looking hairlines and for the younger group in here, think of the competing needs for disposable income; potential house acquisition. For the very young of us, you know, paying off of some debts accumulated while we were at university and the like. A lot of that contributes toward deferring the age at which we start saving for retirement. The rate at which people contribute each year affects the ultimate pension. The size of how much you re getting from your employer. And interestingly enough, just as an aside, the design of how that employer match affects how much individuals choose to save for retirement. Perhaps not as scientific an experiment as you might like but basically if you look at a plan design where the employer matches 100 percent on the dollar, up to 3 percent of pay and lets individuals continue to contribute beyond 3 percent of pay but just without a match, and you compare that to a design with a 50 percent match, up to 6 percent of pay, that latter design will attract much more in the way of employee contributions than the first design, in spite of the fact that the money coming in from the employer is identical. So, just kind of a word about what the right design might be. Délibérations de l Institut canadien des actuaires, Vol. XXXVIII, n o 1, octobre 2006

5 Future of Defined Benefit (DB) Pension Plans (PD-16) 5 Investment strategy employed by the member is obviously going to effect income in retirement and I ll show you something on that later on. The age at which they retire, which may be the whole safety valve or safety release in the DC system is clearly going to affect the ultimate pension. Whether the member buys an annuity at retirement or not is going to affect, not necessarily the pension of an individual when they start to receive their pension but certainly affects the potential for individuals to outlive their retirement savings, which then is the last point, the members longevity. Each of these risks, largely speaking, in DC plans are shouldered by the individual plan member with very little scope for pooling of the risks. There is scope for pooling of expenses. Most of the evidence suggests that members are negatively affected by this so let s kind of take them one by one, some of which I ve talked about. Young members often delay contributing, obviously when it s optional. The younger folks often contribute well below the maximum amount. It s easy to say these people should know better. It s easy to say that plan sponsors should educate people better. It s difficult to get people to say, you know, I have living expenses, I need to live and this is what this money is for. As I mentioned before, contributions above the level that attracts an employer match is subject to each individual s personal decisions and often they decide not to contribute. And depending on the design in the program members often have choice to withdraw money that was put away for retirement savings and use it for something else. Their retirement risk, and as I mentioned before, I think that s going to end up being the safety valve in the system that people will not have sufficient income to retire, when they would like to retire, so they end up working longer if they have that choice. And two things happen: earnings continue to accrue, contributions continue to come in and the period over which you need to receive that income shortens, so it kind of all balances out. That s assuming that the individual has the opportunity to continue working. The biggest issue, in my view, is the mortality risk. You can t pool it in a DC arrangement. Certainly not pre-retirement, postretirement arguably you can but via annuity purchase but that s a big, big issue in DC plans that I don t think has received sufficient airplay. And, of course, the inflation risk. It s difficult, I guess for those of us that work outside of the Province of Alberta to think of inflation as being a big, big issue. But, even at 2 percent levels of inflation, we re living long enough that that s going to take a very sizeable bite out of our real pension income over not too many years. And the big ticket items in my mind are the investment risks and there are a whole range of them set out here on the screen. If you go through the history of DC design,we had plans with relatively modest investment choices and in some cases, no investment choices for plan participants to more and more and more options as employers looked for ways potentially to reduce litigation risk they offered tons of choices. A lot of it is just overload on the part of plan participants and not really understanding what are the true differences between the various options that are being offered. Individuals also tend to invest more conservatively than groups as a whole because they have no real ability to offset their risks or to pool their risks with others, their investment risk that is. What s clearly seen is that people don t adjust their asset mix, other than those that are in investment vehicles that automatically do it for them. They just get drift in their asset mix. They may have started off at something that made some sense and over time it just drifts away from something that s appropriate. And if you think about it the other way around, if you have a very systematic approach to rebalancing back to what the right mix is, if you assume mean reversion and I ll leave that out there as a question mark - I m not necessarily going to defend that - but

6 L avenir des régimes de retraite à prestations déterminées (TR-16) 6 if you do assume mean reversion, rebalancing adds a significant amount of value. If you think about it in terms of a typical DB plans asset mix that probably adds 25 to 50 basis points of return, just rebalancing again. Again, if you assume mean reversion and that over a full career is very useful. Members tend to chase returns, in effect buying high and selling low and that s also not useful for returns over time. Similarly, not fully aware of the quality of investment performance; they re not fully aware of whether good performance was created by chance or by skill of the manager. Again, there s a question as to how much information to load onto employees and when does it become overload. Sponsors have a reluctance to replace poorly performing products; they tend to add more products, not just replace poorly performing products. Individuals don t get as much fund disclosure as paying sponsors and depending on how the investment advisors are compensated, there s a potential bias there to target or to move people towards more highly commissioned funds. So what does that actually mean? What I ve got up here is a summary, it s actually as of the end of 04 but the numbers at the end of 05 are very much the same. I just didn t have minimums and maximums at the end of 05, the average is very much the same. This is information we ve collected from eight major providers in the Canadian marketplace and there are a few key or very interesting messages here. You look at the mix, the balance between equities and bonds and it s more or less 50-50, which was a surprise to me when I first looked at this. That s the good news side. The bad news side is on the equity side it s almost all concentrated in Canada or the vast majority of it is concentrated in Canada. I m not trying to belittle the Canadian equity market but it s a very concentrated equity market and more global exposure makes some sense. If you look at the fixed income side there s a lot of that in term deposits and money markets funds which arguably should be in longer term bond funds instead. My guess is I m going to get the hook from Jerry in a sec, so I m going to have to end shortly. What we did do is we looked at the median pension plan. We do an ongoing survey of all benefits and the median DC plan has an employer contribution of 5 percent. The median DB plan provides a final average earnings benefit of just under 1.5 percent. And if you admittedly make some assumptions and look at what kind of income gets produced, the DB plan for a career employee provides double the income, together with some risk transfer back to the employer. Even if you say, well, you know, a 30-year career isn t reality, three 10-year careers in the DB plan, assuming it s the same DB plan still provides you with 50 percent more pension. So what that means is in an environment where we do move to DC plans, people will either need to save more or work longer. What are the key messages? We have some fundamental problems in the Canadian pension system, I don t think they re insurmountable but they are real and they are serious. If we don t address the problems, this view of the future will really happen; we will have inevitable move to DC. I m not trying to stand up here and argue that DB is the only right alternative, though financially for actuaries that might be a good thing. But it should remain as an option for employment pensions. We need significant change, part of which Paul is going to talk about and that s going to require collaboration from all stakeholders. Thank you. Speaker Paul Forestell: Thanks Steve. I ll do my best to get us back on time. Speaker Bonnar: Thanks, Paul Speaker Forestell: I was relying on Steve going over on his part so I prepared a very short presentation. I came across this quote Délibérations de l Institut canadien des actuaires, Vol. XXXVIII, n o 1, octobre 2006

7 Future of Defined Benefit (DB) Pension Plans (PD-16) 7 from Ben Franklin, I have absolutely no idea what he was talking about and I m very certain it wasn t defined benefit pension plans but the quote itself could be applied to defined benefit pension plans over their history. Steve really talked about today s inconveniences and how they re leading to a change to DC. So the questions in my mind about making predictions about the future of defined benefit pension plans are, what are the inconveniences of the future and will DC plans address them? Are DB plans better suited to address them or is it some combination, or potentially nothing at all, which as Steve said, is the ultimate DC plan? So if there are going to be a number of changes, what has to happen in order to either preserve the DB environment, hasten its departure or at least make it balanced? Both Steve and Jerry have talked about changes that have happened in the past. Jerry went all the way back to group annuity contracts; with very low accrual rates, very low discount rates used to value them and how they weren t appropriate. And then plans moved out of that and as they did that they went to a mismatch mix. And, because pension plans appeared less expensive they increased the accrual rates, they improved early retirement benefits. If you go back to the 70s, 30 and out provisions weren t in the pension plans then, those are a relatively recent phenomenon. And pension plans went from being contributory by employees to non-contributory. Many of these plans happened in the 80s, these changes happened in the 90s as plans had surpluses due to the mismatch in assets and liabilities. I read a quote on the plane, on the way here yesterday, where somebody said that; In an increasing stock market, risk tolerance is infinite and in a decreasing market risk tolerance is infinitesimal. I think that is a very true statement when it comes to pension plans. When we look at the present for defined benefit pension plans, plan sponsors are talking about risk. They re saying I can t take the risk because equity markets have turned against them. More importantly, bond markets have turned against them. Pension plans are more expensive today than they were five years, ago, than they were a year ago, actually probably two years ago, they re about the same expense as last year. So what are sponsors doing today in response to this change in environment? They ve closed their DB plans, they, in some cases, have frozen earnings for the people who are in their DB plans. That s been much more common in the U.S. than either the U.K. or Canada but that s not to say it hasn t happened in both places. So why have they done that? They ve done that because they ve seen their costs increase. Now I d submit that as a profession we haven t done a great job of separating costs for plan sponsors. That s there s really a compensation cost which is the current service cost for their plan. And by freezing their plan, or changing it to DC that s what they ve addressed. They ve taken a current service cost that might be 10 percent of pay and turned it into 5 percent of pay, so they ve saved themselves 5 percent of pay. Often that s not what prompted them to make the change. What prompted them to make the change was that they saw their special payments go from zero to a significant percentage of pay and they re contributing 30 percent of pay in some cases. So what s going to happen in the future? What is going to happen with these plans? Now the motto of the Society of Actuaries is to substitute facts for impressions. For the rest of my presentation, I m not going to be able to do that because I m going be talking about the future and what it might look like. So as Steve said, this is just my view, one possible view of the future. A number of things will have to happen to make it come true; some of them will, some of them won t and in all likelihood I will be wrong. At Mercer, about two years ago we surveyed all of our consultants and asked them for each of their clients what type of pension plan was in place. Now obviously, in the retirement business at Mercer we re very heavily weighted towards defined benefits plans. So this is the distribution that we got. It wasn t a particularly scientific study but it did give a general idea of the status of pension plans among our clients. So less than 30 percent of the plans were standalone DB plans but over 50 percent had at least some defined benefit in them.

8 L avenir des régimes de retraite à prestations déterminées (TR-16) 8 Now, then again, the more interesting question is what s going to happen to these plans as we move forward? So we also asked our consultants to make a prediction for what was going to happen with their clients plans. So we took the first column of the current standalone DB plans and asked consultants to make a prediction about what those plans would look like five years from now. So we re two years into this process, we haven t gone back to actually test and see how good their predictive abilities were but we probably will do that over the next year or two. What you can see is that only half of the plans or less than half of the plans are expected to remain as standalone defined benefit plans. And these are for non-union employees. Obviously if we re talking about union employees we would get a much different result on that. The vast majority of the plans would be viewed as being closed to new entrants and having straight DB wound up or at least a hybrid component introduced into it. A number of plan sponsors, if they haven t been able to make the full switch away from DB have at least put any new money into defined contribution plans. Now, looking at this and given what s transpired in the last two years it could be said that this might have even been an optimistic view now, that looking back on it, perhaps our consultants were more optimistic than they should have been about how many plans were going to remain defined benefit. Steve mentioned the U.K. and the U.S. The stat I have for the U.K. is that over 70 percent of plans there are closed to new entrants. And I don t have an exact statistic for the U.S. but it is not that far off it. So what we re seeing isn t unique to Canada, in fact, you could argue Canada is falling behind in the trends globally but it is a real issue. So, given that the title of this was The Future of Defined Benefit Plans I will make some predictions about the future. Now it s interesting, when I sent this presentation internally within Mercer for review, one of my senior colleagues in Montreal came back and said he disagreed with the second sub-bullet for non-union plans, the move to DC will likely change in the longer term. His view was that there will be no DB plans in the future; that they will all be eliminated. Now, he may be particularly jaded and I must admit, he s got a hairline more like Steve s. Speaker Bonnar: Thanks. Speaker Forestell: And he s not as worried about how long that future might be and he s probably going to retire from our well-funded defined benefit pension plan. But what I ve seen with my clients is that the trend of DC is continuing in the short to medium-term. Clients are exiting DB if they re able to, often based on a push from the CFO rather than from HR. But what I will predict, I think, is this trend will change. It was interesting, that since I put this presentation together I have spoken to three different clients who have come with different perceptions on this. I spoke with one client who took the view that for a 20-year old hire, pensions have no attraction. They don t care whether they have any pension plan at all so why would he bother providing one. But, he felt his defined benefit plan did provide a significant competitive advantage for him in hiring 45 year-old employees from other companies that were getting rid of their DB plans. Now he s in a unique situation in that he resisted, his company has resisted a lot of those changes I talked about in the 90s and still has a relatively affordable defined benefit plan with relatively high employee contributions but he firmly believes that the employees where pensions do matter will prefer the defined benefit plan and the ones who don t really care about pensions, he s not that worried about whether they re going to come or go because of the pension plan. His view was that for hiring younger employees he was better off giving an extra week of vacation, which would cost him 2 percent of pay than to put a defined contribution pension plan in place. The other thing that s always been clear is the unions know the value of defined benefit plans. For everybody here who has had to sit across from either the CAW or the United Steelworkers and try to suggest defined contribution for their members, it s Délibérations de l Institut canadien des actuaires, Vol. XXXVIII, n o 1, octobre 2006

9 Future of Defined Benefit (DB) Pension Plans (PD-16) 9 not the most fruitful discussion. With one of my clients we tried to put it on the agenda for the discussion and they told us to remove it from the agenda or they weren t even going to come to the meeting. Now, you can say what you want about unions but often they are thinking about what is the best for their members; certainly they re thinking about what s the best for themselves. There was an article in the Globe and Mail on Saturday just past talking about pensions and taking the view that defined benefit pension plans are clearly better for employees. I m relying somewhat on some of these views to make their way through to the thinking of plan sponsors. So in the longer term I think DB plans will remain. I think hybrid plans will also increase in existence. In the public sector I think DB plans are here for a long time and then without a significant change in the unionized environment I don t expect to see a move away from DB there. Now, that being said, a lot of things do have to happen to help make this vision of the future come true. I don t think any discussion of defined benefit pension plans is complete without a discussion of asymmetry. Everything starts off with asymmetry. Plan sponsors are saying I don t want to put money in my pension plan because if I happen to be unfortunate enough to wind it up in the future and there s too much money in the plan, I m going to have to give it away. So without legislative change to correct this problem and I think it can only come through legislative change, there will be issues about plan sponsors funding the plans the way they should in order to manage their plans to scale them to the size of their business as they move forward. I think that legislative change will also have to happen to provide funding flexibility. Now this could just be a transition issue, today, given today s low interest rate environments. With significant deficits the contribution requirements that are put on plan sponsors are significant and in some cases, may make the difference between companies surviving or not. So, we have seen this happen already with respect to funding flexibility. The federal government is considering it, Alberta is considering it and Quebec did include letters of credit, along with a whole bunch of other things that aren t so great for the pension system but at least letters of credit were there to improve funding flexibility. The last one is just that something else that is a nice to have would be a national pension regulator. Plan sponsors dealing with ten different funding rules, ten different minimum standards understandably throw up their hands. So I think there could be a move to a national pension regulator with reasonable legislation and legislation that does deal with the funding asymmetry will help to ensure that defined benefits plans continue in the future. Before I move on I do want to just come back to the asymmetry question because I know if the unions were in the room they would react poorly to a suggestion the surplus should belong to the plan sponsor. My point is, often the courts are relying on 60 year-old trust documents that were written in a particular way, due to income tax legislation that existed in the 50s that required you to say that the funds were for the exclusive benefit of the member or you didn t get the tax deduction. It wasn t anything to do with this is deferred compensation so therefore it should be held away from the employees. So my view is the legislation should make it clear that surplus belongs to the person who owns the deficit. If you re responsible for funding the deficit you should have the benefit of any surplus assets. Individual employee groups and companies could contract out of that in terms of how they share the surplus but that shouldn t be the default. I also think that benefit design will be extremely important to ensure the future of defined benefit plans. What the last five years have demonstrated to me is, the designs we have now weren t designed to handle the shocks that have been thrown at the pension system.

10 L avenir des régimes de retraite à prestations déterminées (TR-16) 10 As I said earlier, benefit rates got increased, employee contributions were removed, early retirements benefits were improved; then interest rates dropped 2 percent. All of a sudden what everybody thought would cost 7 percent, 8 percent of pay cost 12 to 15 percent of pay. So I think the new designs will be important. It s not just the actuary, it s the consultants and the plan sponsors who have to keep their minds open to new designs. Iit is also the regulators who have to be more open-minded around new pension design. I was talking to one of our senior lawyers at Mercer who said that he hadn t seen a new pension design, a materially new pension design since 1990 because of the income tax regulations. Now I would argue flexible pension plans that came in in the mid to late 90s could fit the definition of a materially new or distinctively new pension design. But that s probably the only one we ve seen. So flexibility and design, I think, will be key. Things that I ve thought of that would help plan sponsors deal with the shocks that are thrown at their plans are maybe having the accrual for the plan tied to interest rates so that they can decide their compensation cost that they re willing to spend 8 percent of pay on pensions, therefore the accrual under the defined benefit plan will be set based on the current interest rate environment. Or, have your plan be an old career average plan that gets updated when the plan sponsor can afford it. Another way that the shocks can be dealt with is to have variable employee contributions. I mean employees can share either in the current service cost or potentially in funding the deficits. Now we ve seen in the public sector, certainly in Ontario, some of the issues around sharing deficits, as Ontario teachers have seen their contributions go from 9 percent of pay to 12 percent of pay. But, they do still maintain a defined benefit plan that promises them very good benefits. As we look at converting to defined contribution, one of the things that I find interesting is we end up with a DC plan that has a 5 percent employer contribution and often a 5 percent employee contribution to it, if it s 100 percent match on each dollar of contribution. So we re still looking at 10 percent a pay but the employer shifted 5 percent of it to the employee. In some cases I wonder whether employees would rather still maintain their defined benefit plan and make contributions to it. Ancillary benefits are also an expensive piece of defined benefit plans and as demographics change and employers want to retain employees rather than shed them at alarming rates, they may want to change the early retirement provisions in the plan to help balance that. So I ve talked about plan sponsors and regulators. I think unions will need to be opened to alternatives as well. That they can t continue to say we won t even discuss defined contribution plan, that it may be inappropriate to add-on to the defined benefit plan they already have. Together, I think there are some other changes that will need to happen within the system. I think from a funding and investment perspective for pension plans that employers have talked about the risk of the pension plan. So they ve got a mismatch of the pension plans assets and liabilities. With a better match they wouldn t be subject to the volatility in past service costs. As well, if they had a higher funded status, remember what I said about asymmetry earlier because I don t think anybody is going to move to 100 percent funded without some guarantees that they can access the surplus. But a lot of the volatility and the challenges companies run into with defined benefits plans wouldn t exist with a fully funded pension plan. At that point your pension plan is scalable. If you scale back your active employee population your costs go down because you re paying lower current service cost and your past service cost is matched off. It s interesting that there s been a lot in the news about General Motors in the U.S. They re essentially a mutual fund company that makes cars on the side. Their main issue, if you actually look at their financials really isn t their pension plan. While it s under-funded it s not a significant under-funding. The bigger problem they have is with their post-retirement benefits that aren t pre-funded at all and represent a huge liability for them. So, they re now trying to support a huge unfunded liability with a much smaller employee base and a much smaller market share. Délibérations de l Institut canadien des actuaires, Vol. XXXVIII, n o 1, octobre 2006

11 Future of Defined Benefit (DB) Pension Plans (PD-16) 11 So, I think moving pension plans closer to fully funded, combined with better matched investment will result in a more sustainable defined benefit system, maybe with lower benefits but more secure benefits. So, to conclude my portion of the presentation I think the future of pension plans will be different than it is today. I think that with some of the changes I ve talked about there will be a future for defined benefit plans because they do serve and contribute and give value to employees and if plan sponsors can realize the value that the plan provides to their employees, getting back to Steve s point about perceived value, that no employer is going to pay 15 percent of pay when employees attach a 2 percent of pay value to it. But, that being said, without significant change, both from legislative, from plan sponsors thinking, unions thinking, and our thinking as actuaries and consultants, about what the appropriate design is, how to appropriately manage their pension plan, the march towards DC may not be reversible. With that I ll turn it over to questions. Moderator Loterman: Thanks, Paul and Steve. Well, we ll open the floor for questions. I think this should be almost an open forum. I think if anybody has some ideas, let s hear them or questions of the participants, please feel free to go up to the microphones. Mr. Yvon Chamberland: I will start the discussions. I think any hope of having legislative changes moved back away from the asymmetry and away from what has been developed in the past is wrong thinking. I don t think there s going to be any political basis for the politicians to move to a national pension regulator or to move back the question of surplus. I m not even sure the courts would allow the situation to change. I would think that all of us here really believe in the advantages of defined benefit pension plans. They do provide an immense service to the public. But that we are probably the only ones to think that way, plus the unions. You ve got to think outside the box. Forget the past, think what -- based on the legislation that we have in place now. What is going to be best for the public in the future? I think that s the issue, the direction we need to follow. Moderator Loterman: Any comments from the panel on that? Speaker Bonnar: I found it a little difficult to hear everything you said but I think I caught most of it. And I agree with a lot. That it would be very difficult, if not impossible to create, I think, either the national regulator or deal with surplus or some aspects of surplus issues. But in terms of your last challenge which was to say, let s take what we ve got now and what design makes sense, in that kind of an environment to provide defined benefits. In my view, I haven t got a lot of people to agree with me but not for the first or the last time. In my view, the design that works in the current environment that is scalable, truly scalable is a design akin to a multi-employer plan design. And before I get sort of stoned up here, let me put a little more meat on the bone here. What I m talking about is a plan where individual employers can choose to participate, where the design is such that there aren t cross subsidies between employers for things that are in the control of the employer. Things like people s wage increases, people s rate of retirement so that what s pooled is the investment risk and the mortality risk. If you think about it a little more what that means is the design looks very much like the design of the Canada Pension Plan but we ll leave that aside for now. But that kind of an arrangement exists in the current environment. It gets rid of a lot of the negative aspects. How you get from here to there is another story but that kind of design, in my view, may be the only kind of design that works, you know, any number of years out into the future, outside of the union environment or public sector environment. We ll see whether my employer gets mad at me for saying that. Mr. Pat Johnston: Pat Johnston from Pension Strategies. I like a lot of what you said but there s a couple of elements to it, I think. One is that it s retention of defined benefit plans and stopping them from converting to defined contribution.

12 L avenir des régimes de retraite à prestations déterminées (TR-16) 12 But even that in itself won t really change the mix unless there s new defined benefit plans come into place. A lot of what you ve identified I see as being issues purely with existing plans. For instance, the asymmetry of the surplus is largely an existing defined benefit problem. I d say the regulators are largely an existing defined benefit problem as compared to a new defined benefit problem because companies that are looking to set up a retirement program don t know about those problems yet. The issue of a national regulator, probably a national standard would likely affect many national firms but certainly it doesn t affect the provincial firms at all where they really don t care that there s different rules in Ontario than there are in Alberta, if they ve only got Alberta employees currently anyway. The issues I see though on setting up the new plans really has nothing to do with a lot of those existing issues, it s the employees. And the employers demand for employees and the need for employees. They purely want to bring in employees and attract them either from other firms or get them to stop going to other firms. That s purely been in the defined contribution side of things. The employees, you talk about the defined benefit plan overall maybe being worth ten percent and the defined contribution being five and five and the employees are paying for half of it. But at least those employees are seeing that they re getting five percent and they aren t getting it on the defined benefit. You mentioned that they d rather have the week s vacation. Where I see changes occurring is in the executive side of things, and the senior manager side where notwithstanding having a lot of defined contribution plans, the executives and senior managers are having defined benefit supplemental plans added on top of that. And that may be where the change comes in the future, is, as you have the aging workforce and trying to get the age 45 and up hires who are really appreciating the plan. There s nothing we can take action on there, it just may be what happens in the future. So those are my comments. I really don t see a lot that can be done with respect to how it goes, except perhaps educating the 30 year-olds and 25 year-olds that they re a lot better off in a defined benefit plan than a defined contribution plan. You comment about the unions being flexible but they re probably the only ones out there doing that right now. Moderator Loterman: Thank you. Speaker Forestell: It s interesting, the point you made about employees appreciating their defined benefit plan. My experience, my recent experience has been the first time they really appreciate it is when you take it away and you re going out to explain to them why the five percent DC plan is much better for them. Mr. Johnston: On that point I fully agree but that s on the side of keeping the defined benefit plans. I ve been involved with two mid-size companies recently, both who have been looking to introduce a retirement program and we went through all the options. In both cases they opted for defined contribution arrangements, one was a defined contribution pension, one was group RSP. And in both cases they had absolutely zero interest in defined benefit and it had nothing to do because of the regulators or not having a national regulator or the asymmetry of it, it was purely because where the perceived value was for the employees that they were trying to hire. In that case the choice between group RSP and defined contribution pension was the employer that went with the pension, wanted all of the regulators standards on that with respect to locking in the money which Alberta has now unlocked half of it. But they wanted to make sure the money was there for retirement. Also, the other thing on the defined benefit that put them off was the pension accounting rules which likely aren t going to change, in fact they re going to get worse in the near term rather than changing but it had nothing to do with a lot of the other elements that are talked about here. Moderator Loterman: I wonder if as we move into an environment of greater and greater labour shortages where we re going to need, in order to run the economy as a whole, it s going to be different, employer by employer but in order to run the economy Délibérations de l Institut canadien des actuaires, Vol. XXXVIII, n o 1, octobre 2006