Government Operations Committee on Oct. 2nd, 2018

[The Chair (Mr. Tom Lukiwski (Moose Jaw—Lake Centre—Lanigan, CPC)), Liberal, Yasmin Ratansi]

Colleagues, even though not everyone has arrived yet—I believe they're on their way—we do have quorum, so I'll start.

For the first hour today, Mr. William Robson, President and Chief Executive Officer of the C.D. Howe Institute, will speak with us and answer any questions we may have on public sector pensions.

Mr. Robson, I think you know how committees work. We'll entertain your opening statement, which I believe will be approximately 10 minutes or less. Then we'll go directly to questions from all our committee members.

With that short introduction, Mr. Robson, the floor is yours.

[William Robson, President and Chief Executive Officer, C.D. Howe Institute]

Thank you for the invitation to appear in front of the committee today. I don't know whether I'll succeed at entertaining the group, but I hope this will be informative. I'll also be glad to answer your questions.

You will know, I think, but I'll underline, that the federal government's pensions are a major component of the compensation of its employees, including members of Parliament. Ensuring that we have meaningful measures of the value of that compensation matters. It matters for getting the compensation right, paying federal employees properly for the work they do, and giving taxpayers value for their money. Meaningful measurement of those promises also matters when it comes to ensuring that we're funding them properly and allocating their costs fairly over time.

We don't have to look very far away or very far back in time to see how important it can be to get these things right, or conversely, to see the problems if you don't. In the United States in the last few years, we've heard a number of cautionary stories. Members of the committee will know of municipalities in California that have gone bankrupt. Closer to the Canadian border, we had Detroit. Puerto Rico made the headlines recently for similar reasons. All of these jurisdictions had great trouble actually coming up with the cash to pay their employee pensions, then discovering that the cost of these promises was much higher than they expected.

The common element in all those episodes was that the pension plans were underfunded. The main reason they were underfunded was that they discounted the payments that they had to make in the future, using discount rates that were much too high. They did that because it shrinks the liabilities on their balance sheet and it justifies charging contribution rates that are too low.

The managers liked it because it made everything look good. The elected representatives liked it because it helped keep taxes and market borrowing down. And the workers, or at least their representatives, liked it because it made their pensions look secure and left more money for current compensation.

Everybody likes it, until the Ponzi game of making payments not from returns on the investment but from incoming cash gets exposed. When that day comes and cash payments must be made but there isn't enough cash there to make them, nobody likes it.

The obvious question is this: Could that happen here in Canada? Happily, many pension plans in Canada's public sector are not like those that have caused trouble in the United States. They're not all equally healthy, but the shared risk plans in the broader public sector, in most provinces, have two key features. They use economically realistic, meaningful discount rates when they look at the future payments they have to make. They also have the flexibility for contributions and benefits to respond if the financial reports indicate that they should.

Unhappily, however, the federal government's pensions are not like that. They are pure defined benefit plans. They're relatively generous by the standards of other public sector plans in Canada, and private sector plans certainly. The financial reports we get on them use discount rates that are too high. Their liabilities on the balance sheet are too small. Their ongoing costs that we see in the federal government's statement of operations in the federal budget are too low. Those pensions are, in short, more valuable to their recipients than what the federal financial statements indicate, and they are correspondingly more costly to taxpayers.

I will mention, because you're members of Parliament, that the pension plans for federal members of Parliament are a particular problem. They are completely unfunded. You could be forgiven for not knowing that. You'd need to read the chief actuary's reports on them very carefully to discover that. The assets in these plans are simply bookkeeping items. The contributions that go into the plan actually buy no assets. The benefits that go out of them are funded by current revenue and market borrowing. The cost of the payments in the future will be higher than what you see in the financial statements.

That is not true of the big federal employee plans for the public service, the Canadian Forces and the RCMP. Those plans are partly funded. Contributions since 2000 have bought assets, but the pre-2000 obligations in those plans are completely unfunded. The value of all of the obligations, pre- and post-2000, are understated because of these excessive discount rates.

Why do I say the discount rates are excessive? The promises to pay these pensions are unconditional. They are like any other federal government debt that is an unconditional promise to pay. They are indexed to inflation, and that means there is a ready comparator for that kind of an obligation. They are like the federal government's real return bond.

If I were a federal public servant and someone offered to buy my pension from me, I would use the discount rate they yield on the federal government's real return bond in thinking about how much it's worth. If I were a taxpayer thinking about how much money I need to set aside, because one day I'm going to have to pay taxes to pay these pensions, I would look at the same thing. I'd look at the asset that matches the liability, and that 0.7% real return is the return that is applicable. That's the accumulation rate I could expect.

The problem is that the federal government's pension liabilities are discounted at much higher discount rates. I can go into that in response to questions. That used to be a common practice in many pension plans, but it's becoming less prevalent over time because of some of the problems that have emerged with pension plans.

If we use an economically meaningful discount rate, and if we think about the real return bond as the right benchmark to use in valuing the federal government's pensions, the federal government's pension obligation and its net debt are understated at the end of the last fiscal year by about $100 billion—$96 billion.

To put this in perspective on the way forward, recent reforms raised the share of participants in funding these plans—that applies to the MP plans too, thinking about the contribution rate—and it aims at a fifty-fifty split, more or less. But there's a problem with that formula because it's based on an ongoing cost that is held artificially low because of these high discount rates used to estimate the value of these plans.

Even as we get to 50% contributions by the employees, the true economically meaningful split of the cost will not be fifty-fifty. The taxpayer will still be left with the bulk of it, and when I say “the taxpayer” I mean much more the future taxpayer than the current taxpayer. We're storing up trouble for the future.

More economically meaningful reporting of these plans, benefit values and their cost to taxpayers, I think, would be a valuable step forward in helping the federal government think about its ongoing obligations and how to fund them.

I will add as a footnote, and then I'll close, that the Public Sector Accounting Board is currently consulting over key questions that relate to this, both the timing of recognition of changes in the value of a pension plan's assets and liabilities, and also the discount rates used. There may be changes in public sector accounting standards coming down the road that would encourage or mandate the federal government to move in this direction. Whether they do or not, I think it would make sense for the federal government to look at these things on an economically meaningful basis. I think if it did, there would be better funded pensions for federal employees, including members of Parliament, and it would also provide protection for taxpayers against risks that few of them know that they run, and against costs that they don't currently know they're shouldering.

Thank you for the opportunity to testify. I'd be glad to take your questions.

[The Chair, Conservative, Tom Lukiwski]

Thank you very much, Mr. Robson.

I can assure you that your comments about the members of Parliament's unfunded liability in the pension plans certainly have the attention of all of the members of this committee.

We'll go directly to questions from committee members.

Mr. Jowhari, you're up first for seven minutes, please.

[Majid Jowhari, Liberal, Richmond Hill, ON]

Thank you, Mr. Chair.

Mr. Robson, I had the opportunity to read the report, and you summarized it well. In my mind when I was reading the report and listening to you, I had two take-aways. One was that the discount rate needs to be looked at and the method that we evaluate our assets and recognize our liabilities also need to be looked at.

Can you help me understand two things? One is the transition that has been taken since 2000. The other one is regarding what jurisdictions are applying different methods that are much closer to what's being suggested. How has their performance been as far as mitigating the risk associated with taxpayers being on the hook is concerned?

[William Robson]

I want to pay tribute to past Canadian governments for having tackled some of these public sector pension issues the way they did. We not only had the reforms to the Canada and Quebec pension plans affecting the broader population, but in 2000 there were major changes made to the federal government's plans.

Since then, some of the contributions going into the plan have flowed into assets. Effectively there are now two accounts in these major plans. If you look at the actuarial reports you'll see that they're segmented out. You've got the pre-2000 obligations and then the post-2000 plans. The post-2000 plans have assets as well as liabilities. That's commendable.

You can go quite a bit further than that if you look at what has happened in the major provinces with the broader public sector plans. The Ontario Teachers' Pension Plan is one of the better known ones. It was one of the first. There are many shared-cost plans in Quebec, British Columbia and Alberta, and a different approach in some of the maritime provinces.

One of the key things in those plans is typically they've adopted quite conservative discount rates by the standards of public sector pensions, and certainly compared to what I was describing in the United States. Part of the motivation for that is just to be conservative. If you think of your pension obligation as to pay the pensions—if that's your primary aim—then it makes you think differently about the types of assets you'll hold. You'll typically focus more on bonds. You're not just trying to chase returns, because you've actually got to have the cash on hand.

In support of that, in some of these broader public sector plans you'll see more conservative discount rates. They focus much more on the funding ratio because, as I say, they're thinking much more in terms of wanting to pay those benefits when they come due.

In the federal government's case there has been a sense of...I won't say it's because we control the central bank, because that's a bit contentious. It's more of an assumption that the tax revenue will just always be there, and that there'll always be the resources there to pay the pensions. Perhaps it seemed less of a challenge to actually think about paying the pension promised from the plan itself.

As a result, at the federal level you still see these discount rates that are just based on history, on an assumption about return on assets. To me that makes no sense. If part of your plan is unfunded—and certainly part of the plan is totally unfunded and part of the rest of the plan is underfunded—there are no assets to earn those returns. You really ought to be thinking about the nature of the obligation. That's the right way to think about it.

You asked about different methods. I would just underline how innovative and successful Canada has been to date with these broader public sector plans. They're actually world-famous. The federal government could usefully imitate some of what we see at the provincial level.

[Majid Jowhari, Liberal, Richmond Hill, ON]

You brought up the point that the MPs' pension is completely unfunded. I recall that when I became an MP, one of the choices I had to make was the amount of my contribution, and I could max up to 50% of the contribution. Can you shed some light on that? Right now, where is the money that we're paying going? What's happening to it?

[William Robson]

MPs could certainly be forgiven—if it's my business to forgive anyone—for not knowing that these plans are unfunded, because they have the trappings of a funded plan. As you pointed out, you're paying contributions, and if you look at the financial statements for the plans, there appears to be an asset against the actuarial liability that's based on how long MPs are likely to live, how much income will be replaced and so on.

The awkward fact is that the MPs' plan is still like the pre-2000 plans for the public service, for the RCMP and for the Canadian Forces. Those contributions are not buying assets. They are simply going essentially into the government's current cash flow. The payment for the pensions is coming out of the government's current cash flow.

That's unfortunate. People like to talk these days, for good reason, about tone at the top. It's awkward that the federal members of Parliament do not participate in a pension plan that is appropriately funded. It also provides a key lesson in the problem you have when the financial statements do not report in a way that ordinary people—non-experts and non-actuaries—can understand when they are wondering whether their pension is properly backed or not, or if a taxpayer is looking at the federal government's finances and asking, “Am I on the hook for something here that I didn't know about?”

I'll elaborate briefly in case you want to follow this up. The chief actuary does reports on the federal MPs' pension plan. You'll see a line in it that says there is no separate fund held in these plans. The accounts are simply bookkeeping entries, and I regret to tell you there's no money in them.

[The Chair, Conservative, Tom Lukiwski]

Thank you very much.

With those positive words, we'll turn it over to Mr. McCauley for seven minutes, please.

[Kelly McCauley, Conservative, Edmonton West, AB]

There's going to be a scramble to see who can retire first and grab their pension.

Mr. Robson, thanks for joining us. It's a pleasure to have you with us.

I read with dismay your report, “Taxpayers on Hook for Ottawa's Pension Shortfall”. Has the $96 billion of hidden unfunded liability that you talk about been backed up by any other organizations? Is it solely through the C.D. Howe Institute that you've come up with the number or is it generally a well-known secret?

[William Robson]

I wouldn't say it's very well known. Pension accounting—and I'm delighted to be talking to your committee about this—isn't a topic that has been a big grabber in the past because valuing these payments that will occur a long time in the future is a genuine difficulty. It's why there's debate about it.

The place that I would refer you to for a similar type of analysis would be in the United States. There are some think tanks there, some academics, who have been paying attention to this problem at the subnational level, particularly in the United States. I mentioned them already, so I won't go on at length, but I will say that the problems there are worse than what you find in Canada because the discount rates are so high and many of the plans are extremely brittle. That's why you see these problems.

The debate that we're part of here is about valuing the liabilities. The people who have looked at the U.S. plans and said there's a problem and they're understating their liabilities, make the same case that we do. If you think about the value of this promise, including the one that we are making to our members of Parliament as they work, it's unconditional. There's no flexibility in that benefit should the funded status of the plan turn out to be less than was anticipated. In that respect they're unlike, say, the Ontario Teachers' Pension Plan or some of the shared-risk plans.

We also have this tendency to discount at assumed rates of interest. They're not as high in Canada as they are in some of these U.S. plans, but there's a real question as to why you would just pick a number like that when you have an obligation that is so clearly comparable to the promise that you're making in the federal government's other bonds.

[Kelly McCauley, Conservative, Edmonton West, AB]

That's my next question. How did we arrive at this arbitrary incorrect number instead of using the RRB? Is it just a past practice that's continued along?

[William Robson]

Yes, it is past practice. It used to prevail not just in the public sector but in the private sector as well.

Any of you who have in your legislative lives or elsewhere dealt with any of the pension-related issues will know that as a result of some very high-profile bankruptcies, both in Canada and elsewhere, the pension regulators began to take a dim view of this use of high discount rates. I referenced earlier the convenience of high discount rates on both sides of the table, if you like, management and labour. In the short run it's attractive. It makes the plan look well funded. Management, if you're in the private sector, can pay higher dividends or higher current compensation. The workers can take higher current compensation. If you're a government, it makes the bottom line look better. But that's an artifact of choosing a high discount rate.

If the discount rate were lowered more to what I think is appropriate, from the point of view of both the value of the promise to the employee and also what the taxpayer faces—what we would have to do as individuals in order to have that same kind of income—it does show that the value of these promises is greater than what we've been saying.

[Kelly McCauley, Conservative, Edmonton West, AB]

Have the provinces all moved away from an arbitrary number into a more realistic number? Is it just the feds that are continuing along down the wrong path, so to speak?

[William Robson]

Well, the federal government does stand out for the size of the discount rate that it's using, and also because it's using that discount rate to value promises that are unconditional. An unconditional promise is quite different from the promise that you would see in a shared-risk pension plan, such as you would see in the broader public sector in many of the provinces, where the participants in the plan know that, if circumstances don't work out well, they might have a reduction in their benefits. For example, they might lose part of their indexation or in some cases they might see the base benefit whittled down. That creates quite a different kind of a promise for those pension plan participants because there's some risk involved; it's not an unconditional promise. Under those circumstances you could justify having a slightly higher discount rate than what you would get if you used a government bond, but the essence of the problem is to think: how valuable is this promise?

I will say—and I apologize that I'm branching off from your question—if you think of the situation that a Canadian who does not work for the federal government would face in trying to provide for herself or himself a similar kind of pension to what the federal employee gets, the right kind of asset for them to hold would be the real return bond because it's an unconditional promise and it's indexed to inflation. So the amount they would need to save is vastly higher than what they can save because they would need to match the contribution rates that you'd have to pay to fund the plan properly in that kind of an environment.

There are important ramifications from this for retirement saving more generally.

[Kelly McCauley, Conservative, Edmonton West, AB]

You mentioned that the Public Sector Accounting Board is looking into this. I read somewhere they are looking at trying to change to an RRB or something similar. Have you chatted with them? Do you get any sense that this is how we're going to go forward to address the issue or is it just a pie-in-the-sky hope right now?

[William Robson]

I'm a member of the advisory group on compensation to them, so I've seen the consultation papers. I wouldn't prejudge the results of that consultation. There is a debate. I've presented one side of it and there are other people who take a different view. The fact that these consultation papers are out there at all, and the fact that the Public Sector Accounting Board is looking at these issues, signals there is, at least among some constituents, discomfort with the way this is being done.

It's been a long time since the accounting standards they're reviewing have been looked at. We've had some very cautionary tales out of the private sector. A recent example in Canada is Sears, where there was an understated pension obligation. There are other examples in the public sector, abroad more than in Canada. There is ample reason to think that these standards haven't been serving us as well as they might and that we should be looking both at the timing of recognition, which has to do with how quickly you recognize changes in the status of the plan, and the discount rate itself.

[The Chair, Conservative, Tom Lukiwski]

Thank you very much.

Madam Benson, welcome to our committee.

You have seven minutes.

[Sheri Benson, Saskatoon West, NDP]

Thank you very much, Mr. Chair. I'll do my best to stay on top of what seems like a complex journey here.

I am going to ask some basic questions, the answers to which I think Canadians and workers would be interested to hear.

When you talk about the nature of a defined pension plan, the first thing is the unconditional obligation or contract with the employer and employee. As I think we can all appreciate, for those who are in those plans, those kinds of pensions are valued very highly. I'm not going to get into the argument about whether they're good or bad.

In what you've brought forward, and in the fact that we're not using a rate that gives us a realistic projection into the future and we're therefore not basing it on what contributions need to be made for people right now, is there a way to have a defined pension plan that is more realistic in its rate of return or are you saying that defined pension plans are something we're not going to have in the future? Are they something that only happen in government because governments don't go bankrupt?

I'm just trying to figure out where that fits into your thoughts.

[William Robson]

The key question we should be asking about pension plans of this kind is whether we want to make that absolutely unconditional guarantee that we'll pay. Of course, there are always circumstances—the end of the world, for example—that would prevent a pension plan from delivering. However, excluding extreme possibilities, I think the right way to think about it is to ask what asset you would hold, such that if you suddenly needed to pay the obligations, you could do it and be able to put your hand on your heart and say to your plan participants, “We have you covered on this.”

In the broader public sector plans where there's a little bit of benefit flexibility, I think the joint governance probably helps to have this kind of conversation. Many of them, including some of the more recent ones on the block, like OPTrust in Ontario, for the public service there, the members of OPSEU, have explicitly said to their members that their job is to pay their pension benefit. The idea is partly to get people's attention off the rates of return on the assets because that's a bit of a sideshow. Everybody's pension plan is different. Employees may be older, younger or have different characteristics. You want to make sure that your assets are suitable to pay those obligations.

The sticking point with the pure defined benefit plan, the one where there's absolutely no flexibility, is that the type of asset you should hold to meet that promise is a very low-risk one, like the federal government's real return bond. Low-risk assets yield low returns. To buy enough of those assets to guarantee the pensions, you have to buy a lot of them. That makes the contribution rate high, and that has been the point at which people have bought. They've said that looks so expensive. They don't think about it as saving.

I'd say it's saving. If you want a guaranteed pension, save a lot. You would do the same in your private life. It looks like a cost, and people balk at the cost, and that's where you see this high discount rate creeping in to make it look affordable, but then what you're holding in the plan isn't actually going to be sufficient to meet it.

I don't have a problem with generous pension plans. I think you want to consider a balance between deferred and current compensation, but I would like to see Canadians in general able to save more. To do that, they need to contribute more. Sometimes, in order to contribute more, they have to be shown the cost of the pension and how much they should be paying. It's a bit shocking to see how much, but if returns are low, that's realistic. If you want that kind of pension, one with a real guarantee— close to certainty—of getting it, that's what you have to do. You have to contribute more.

[Sheri Benson]

You're saying there's been a bit of a cultural change. People need to have these conversations about the level of contribution in order to have that kind of pension. We're shying away from those conversations because for a lot of pensions, it would mean quite an increase in the contribution, both on the employer's side and on the employee's side, if we were more realistic on the future rate of return.

[William Robson]

Yes.

[Sheri Benson]

You talked in your article about steps toward looking at this unfunded liability or being more realistic in our future projections of what the assets in the pension plan will earn.

Do you want to talk about that? What's the process? What are the steps to get to a better place? Is the first conversation about using the best guess of a discounted rate for the future? Is that a conversation we are having all over the place? Is it something that you're a part of? Is that the first step?

[The Chair, Conservative, Tom Lukiwski]

Mr. Robson, you only have about a minute for your response.

[William Robson]

It starts with the numbers, because the numbers are very powerful in presenting us with a situation, and helping us think about what to do with it.

The focus within the federal government is on estimates that reconcile the budget. It sounds like a green eyeshade exercise, but it's very important, because it helps people figure out how this fits into the broader fiscal plan. I would start with that. The accounting matters a great deal. It would matter if we saw the ongoing cost of these plans.

With respect to the earlier conversation, if you look at the broader public sector plans, their joint contribution rates, employer plus employee, are now 23%, sometimes 24%, and that's with a lower level of benefit. We've seen pension plans making that adaptation. A realistic presentation of the numbers was a key step for them.

[The Chair, Conservative, Tom Lukiwski]

Thank you very much.

Now we'll get some questions from one of our millennials, who may have about 35 or 40 years before he starts collecting pensions.

Mr. Drouin.