Problem
The first problem is that the current plan has a significant deficit. The second, but more pressing problem is that the University is threatened with special payments to cover the deficit computed under the solvency rules. Because coming up with lots of cash at short notice is costly and disruptive, they want to avoid it.
Deficit
The law requires pension plans to undergo two types of actuarial valuations: going concern and solvency. The two valuations measure the same thing (the deficit) in two different ways, under different assumptions, and are designed to address different potential problems with the plan. In the case of our plan, each of these two valuations shows a deficit, with the solvency gap being substantially larger The most important numbers (as for 1 July 2016) are collected in the Table below. This information comes from the most recent Actuarial evaluation of the University of Toronto pension plan.
Table Actuarial valuation of UofT pension plan at 1 July 2016
Going Concern
Going concern valuation evaluates the ability of the plan to make good on its promises assuming that the plan is going to function exactly as it does now. That means that there will be a sponsor, who is responsible for the management and for future surpluses and deficits, and, in particular, who can invest assets in the best possible way so to use the returns to pay for future benefits. It means that the actuary should use a mix of average stock market return rates and bond rates as the discount rate to evaluate the present value of the liabilities (this sentence is explained in a section about actuarial calculations).
Solvency
Solvency valuation asks a hypothetical question, how secure is the plan and current benefits assuming that the employer closed down today. The purpose of the test is to protect the plan members from a potential bankruptcy. The scenario assumes that there will be no more plan, no management and, most importantly, no future investment of the assets.
There are two differences with respect to going concern. The first difference is that the actuary uses bond rates as the discount rate to evaluate the present value of the liabilities (this sentence is explained in a section about actuarial calculations). Because the bond rates are much lower than the expected stock market returns, the solvency calculation leads to a larger value of the liabilities, and larger deficits. (For instance, Actuarial evaluation of the University of Toronto pension plan assumes the stock market return at 5.75% and the bond rate at 2.85%.)
The second difference is that under the solvency rules, the actuary can assume that future pensions are not indexed (this is also called as exclusion of escalated adjustments and it is permitted by the Pension Benefit Act). This reduces the value of liabilities comparing to going concern. In fact, in normal times, the second difference dominates and the going concern valuation of liabilities is higher than the solvency valuation. On the other hand, whenever there is an expectation of low inflation rates, the exclusion of indexation plays smaller role, the first difference dominates, and the solvency valuation of liabilities becomes larger than the going concern valuation.
Hypothetical wind-up
Additionally to going concern and solvency, there is a third type of an actuarial valuation: hypothetical wind-up. It relies on the same assumptions about discount rates as the solvency valuation, and it does not exclude indexation of benefits. So, if the benefits are indexed to inflation, the actuary will take it into account when performing valuation of liabilities. Not surprisingly, the valuation of liabilities under the hypothetical wind-up are the largest of the three.
Which number is more important?
The two tests are designed for different purposes. The going concern valuation is a natural measure when you think that bankruptcy is not an issue (at least, it is not an issue at this moment, and unlikely in a near future). The solvency valuation is natural if you are really worried about bankruptcy.
Some people argue that the solvency test is not well-suited for a broader public sector, where bankruptcies are less common. It is difficult to imagine a bankruptcy of the University of Toronto. In the same time, the bankruptcies in public sector do happen (just look at Detroit).
A problem with the solvency test is that it is very volatile. A small change in the bond rate may lead to a large change in the present value of liabilities. For instance, the large value of the deficit on 1st July 2016 is due to very low rates at the date of the valuation. The rates were slightly higher on 1st July 2015, which led to a solvency deficit of $1 bln (which is substantially smaller than $1.7 bln in 2016). Presently, the rates are slightly higher again. If the current trends continue, the solvency problem might be significantly smaller (even if still large) by the next actuarial valuation on July 1st, 2017.
On the other hand, the solvency number has an additional meaning, beyond bankruptcy, that is very relevant to our discussion about pensions (see the treatment of old plan liabilities). In our current plan, the University is solely responsible for paying all the benefits that are earned till now. This is a promise that the University made. Suppose that the University would like to get rid of this promise. They cannot just break the promise, because it is a promise. But they could pay somebody else to take this promise away from them. $5,811 mln* is how much they would have to pay for anybody to be willing to take over the responsibility for the promise. This is an open market value of the University pension promise. (*That is not completely correct. The exact value of the promise should also include the indexation part and is given by the hypothetical wind-up calculation of liabilities - see above. But, we simplify for the sake of discussion.)
The solvency deficit is much larger, and because of this, it is a much bigger source of worry for the administrators. In fact, as we explain below, the University has a grip on the going concern deficit. The solvency deficit is the main reason, why we are where we are.
Why did we get into trouble?
There are four main causes: under-funding, bad investment decisions, longer life expectancy, and historically low rates.
Under-funding
Prior to 1987, the University was obliged to contribute $2.55 for each $1 of faculty contribution to the pension fund. That means that effectively, the plan was jointly sponsored, although the amounts of contributions were not equal.
That changed in 1987, when the University took full responsibility for future deficits and surpluses. The expectation was that the University would keep on contributing $2 for each $1, but this did not happen. In fact, the University immediately started taking contribution holidays (from 1987 to 2004), partially counting on the dot.com boom lasting for a longer time than it did. In multiple years, the holidays were extended to plan members (see also Is there really no risk?). Some past information can be found here.
On top of the low contribution rates from the University, the faculty contribution rates were (and remain) low: They raised from between 4.5-6% before 2013 to 7.15-9.50% after June 2017. Compare it with the contribution rates at masterfully run Teacher’s Plan OTPP, which remain consistently 3-4% higher (OTPP is a jointly sponsored plan, so employees contribution is matched by the employers). See also graph on page 3 of this paper.
Bad investment decisions and 2008
Around 2000, the University created a new entity, UTAM, to manage the University endowment and pensions. Its performance, especially in 2008 (they lost much more money during market crash than other funds) and 2009 (where they failed to catch-up with the recovery) was a disaster. Jackman's report blamed the performance on the lack of oversight from the university.
UTAM underwent a series of changes around 2011, which includes new leadership, structure, multiple layers of oversight. Since then, the returns came back to normal - which means that they are roughly in line with other Ontario public pensions funds.
Longer life expectancy
Since 50-ies, the life expectancy in Canada grows steadily at rate of 1 year for every 5 years. Longer lives mean longer pensions. Some of the increases in the life expectancy were clearly not anticipated by the past actuarial evaluations. When actuaries realized that the life expectancy is longer than anticipated, this led to a sudden increase in the value of liabilities.
Historically low interest rates
The lower are the interest rates, the higher is the present value of liabilities . The interest rates remain at historical low. (The last time that the bond rates were so low, we were preparing for a war.)
Whose fault is it?
A short answer is that it is the University’s fault. As the single-sponsor of the plan, they were solely responsible for the management of the plan and its financial standing.
At the same time, the faculty and their representatives are not completely blameless either. They didn’t watch the University to make their share of contributions to the plan. Their rate of contributions were relatively low (see above).
Many other Ontario university pension plans have similar problems. More generally, single-employer, single-sponsor, defined benefit plans were all designed in the glorious past, when future looked bright, growth rate was >4%, societies were full of young people, who smoked and drunk more and died quicker. Since then, such plans brought down numerous companies, and not so few public entities. It is not entirely fair to blame the University for wider social trends.
Immediate consequences of the deficits.
The main consequences are for the University who makes special payments to cover the deficit computed on the going concern basis.
University: As the single-sponsor, the University is required by law to make special payments to cover the deficits. The law (the Ontario Pension Benefit Act) says that
Any deficit determined on a going concern basis must be eliminated through special payments over a period of no more than 15 years. The University pays around $100 mln each year to cover the deficit computed on the going concern basis - see Special payments below.
Any deficit determined on a solvency basis must be amortized within 5 years of its appearance. Since 2011, the University received various form of relief regarding the amortization of the deficit determined on a solvency basis - see regulatory environment below. As a result, the University did not have to make any payments above the going concern payments. Given the current regulation, the University won't have to make solvency payments until 2020-21. If the relief were not to be extended after that date, and the interest rates remain so low, the University faces an extra $50-150 mln for the solvency payments (the exact number depends on economic situation and the results of the actuarial valuation on 1st July 2020). A concern about the latter is the main reason why the university pushes towards the reform of pension plan. The new plan is expected to receive a permanent solvency relief.
Plan members: There are no immediate consequences for the plan members. In the two previous bargaining rounds, the University asked and got increased contribution rates. At this moment, the University does not ask for increased contribution rates. However, the contribution rates remain on the low side relative to many other pension plans.
Retirees: An immediate consequence for the retirees is that it is very difficult for them to get a full (i.e., at 100%) indexation. However, given that all the alternatives on the table either eliminate indexation at all or make the 75% indexation conditional on the health of the fund, the consequence does not seem very costly.
Regulatory environment
The University of Toronto is not the only institution that faces large deficit of its pension fund, and, potentially crippling, emergency payments. Because of the economic situation (very low bond rates), many other private and public pension plans face the deficit and, in particular, the need to amortize the deficit according t solvency rules. For this reason, since 2011, the government is working on ways to address it and reduce the burden on the pension plans. They follow a two-line approach:
Temporary measures: Since 2009, they offered various and changing ideas about how to provide temporary relief from the need for solvency payments (see UofT Budget Report 2016-17 and Solvency relief regulations.).
Originally, starting from 2011 actuarial valuation, the U of T plans would have been required to commence full solvency deficiency payments at July 1, 2015 and to make those payments over a 10–year period.
Under Stage 2 of the solvency funding relief, after 2014 actuarial valuation, the government permitted a three-year deferral – in our case to July 1, 2018 - however, the solvency deficiency would then have to be funded over the subsequent seven years, increasing the annual payments because of the shorter period. The University has opted to take this option.
Under the subsequent solvency funding relief, after 2017 actuarial valuation, the University will have to start making payments that are equal to 1/7 of 25% of the solvency deficit. This means that given that the University has already been making going concern payments, and they are larger than 25% of the projected solvency payments, the University is likely not have to do anything extra till at least 2020-21 (again, this is conditional on the details of the actuarial valuation on July 1st, 2017).
According to the current rules, after 2020 valuation, the University will have to amortize the remaining part of the solvency deficit over 4 years.
Permanent solutions:
Government created a special commission led by D. Marshall to review the existing rules of solvency funding for public sector pension plans. The commission is preparing a report that is due to appear in June 2017. One of the goals of the commission is to propose rules for permanent solvency relief (see Discussion paper).
In the same time, the government pushes the industry to reform pension plans. The Ontario Ministry of Finance provided financial grants to design a multi-employer, jointly sponsored plan. This is how the new plan was developed. See below for discussion of what the government may want.
Rules for permanent solvency exemption
In 2011, the government permanently exempted six jointly sponsored pension plans in the broader public sector from solvency funding requirements. While there are no stated criteria for exempting these JSPPs from solvency funding requirements, most of these plans are large multi-employer plans. Since then, some criteria were proposed. The present new plan does not satisfy some of these criteria. The expectation is that as long as the University forms a multi-employer, jointly sponsored plan, it will get a permanent exemption.
The design of the rules for permanent exemption was left to Marshall's comission. The report is expected in June 2017.
Special payments
The next Table presents all the current (regular and special) payments to the current plan (sources Actuarial evaluation of the University of Toronto pension plan and UofT Budget Report 2016-17 ).
Table: Going concern special payments
*The payments over 2016-2019 increase progressively. A fraction of the payments include interests for the loan taken in 2011-13 for the first tranches of the special payments.
Under current regulations, the University is not required to make any additional payments to cover the deficit evaluated on solvency basis till 2020-21. To illustrate how important the rules are, we show how much the University would have to pay every year between 2018-22 on top of special payments from the previous Table if there was no subsequent solvency funding relief (hence, the relief ended with Stage 2 - see the regulatory environment above.):
Table: Solvency special payments with no solvency relief.
The required payments depend very heavily on the economic variables, like discount rates. The current economic trends look good for our plan (Trump bump on the stock market as well as sings of inflation), and there is an expectation that the 2017 actuarial report will look more like 2014, then 2016.
How does university cope with special payments?
The special payments reduce budget available to units throughout the university, money on programs, hiring, etc. Is the squeeze felt at the university? Also, borrowing costs. TBA
What does the government want?
At every opportunity, the government said that their preferred option would be a sector-wide JSPP. The reasons why that's what they want are not obvious. But, we can speculate. Here are some thoughts:
- One sector-wide plan simplifies regulations: instead of 17 problems, the government has to deal with one. The only way to create a meaningful university-wide plan is when the largest university, i.e., the University of Toronto joins.
- The government believe size makes pension plan more stable and secure (although the size argument does not apply to our situation).
- Some of the most successful pension funds in the world are Ontario jointly-sponsored multi-employer pension plans (although some others are not so much). It is easier for the government to justify a conversion to a model that is widely recognized as successful.
- In case of a employer-sponsored plan, the employer is responsible for the liabilities all the way till bankruptcy. It means that in case of significant deficit, it is possible that the University would have to sell its assets (endowment, real estate) in order to pay out pensions. In the same time, it is impossible to believe that the government would allow the University to go bankrupt. More likely, in such a case, the government would need to step in and bail out the University. In case of JSPP, the protection is weaker (see reduced pension security). In particular, neither the University, and certainly not the government is responsible for our pension. That takes the government off the hook.
Relevant long-term uncertainties
Here are some things that can happen in longer term, beyond 3-5 years.
Problem may disappear. The interest rates are historically low, if they go up, there might be no solvency deficiency in 2-3 years. At this moment, the trends are good. First, there was a significant raise in the asset prices as Wall Street is still celebrating Trump elections. Second, there are signs of inflation rising, which may cause central banks to increase the rates.
The economic situation on 1st July 2017 is particularly important. This actuarial valuation done on this day is going to be used for filing with the government and it will be determine how much solvency payments will need to be done for the subsequent years (see regulatory environment above).
If the current economic trends remain till July 2017, the going concern deficit might be substantially smaller than in July 2016. As a consequence, the University could be significantly ahead in terms of the schedule of going concern amortization. That would leave some extra money on the table for the University, at least relative to its current situation.
Additionally, if the current economic trends remain, the University won’t have to make any solvency payments till 2021-22.
Uncertainty about regulatory environment may resolve. Since 2009, since the solvency problem started to pop up, the government keeps on revising regulations under which the troubled pension plans receive an exemption from the law that requires them to amortize the solvency deficit over 5 years.
In 2016, Ontario government created a commission led by D. Marshall in order to review the existing rules of solvency funding for public sector pension plans. The commission is preparing a report that is due to appear in June 2017.
There is an expectation that the Marshall commission is not going to recommend full solvency relief. So far, the government’s actions suggest otherwise: there has been an emergency solvency relief to public sector since 2009.
The Marshall’s commission will also propose the rules for permanent solvency exemption for jointly-sponsored plans. This is relevant because it is going to affect how our plan is formed.
Political uncertainty may resolve. Ontario elections are coming up in 2018. In 2012 and 2013, Tim Hudak and Ontario Tories advocated for defined contribution plans for public employees. Some fear that Tories win will mean that the University will be forced to move to a something like the Possible Threat-point described below.
It is far from guaranteed that Ontario government will change.
Even if they win, there has been no public comments about defined contributions plans after 2013. Tim Hudak is gone. That does not mean that Tories changed their mind, but it also does not mean that they didn’t. It is worth paying attention what they say.
Finally, it is worth noting that if so happens, the University of Toronto won’t be the first one or the last one that goes that path.
Possible threat-point
In order to be able to evaluate whether it is prudent to change the new plan, we need to be clear what is the alternative. Unfortunately, the alternative is not clear.
First of all, it is not entirely clear that the current plan cannot survive (see the above discussion about the relevant uncertainties). Second, we have no idea what would happen if all the relevant long uncertainties resolved against the current plan (i.e., economics would work against us, there would be no solvency exemption, Tories would win, and would try to force all public plans into defined-contribution).
A possible speculation is that we would get a something like a hybrid defined-benefit and defined-contribution plan. Something like McGill's pension plan, for instance. Roughly,
the University would remain completely responsible for all the benefits accrued before the new plan,
all the old employees would earn new pensions in the form of a defined contribution. That means that at the retirement, the old employees would receive a part of pension that would depend on their highest salary multiplied by the serviceable years before the new plan started plus a pot of money from the contributions after the new plan started.
all the employees hired after the beginning of the new plan would be entirely under defined contribution.