Lessons from Europe’s pension stress test

For many people, the mere mention of European Union rules conjures up images of red tape and bureaucratic minefields. So it’s hardly surprising that Europe’s pension industry wasn’t thrilled with the newly released results of the first-ever European Union-mandated stress test of occupational pensions.

Critics charge the information the exercise provided has been available all along.

They also question the validity of the results, arguing that administering a common test in an area where each country has different pension rules produces an imperfect comparison. It’s a complex situation that offers lessons for Canadian pension funds.

“It simply doesn’t make any sense to use a single method for valuing pension schemes when there are 28 different pension systems across Europe and 28 different regulatory regimes and tax regimes. Inevitably, you end up comparing apples and pears,” says James Walsh, a policy lead at the Pensions and Lifetime Savings Association in Britain, a London-based organization that represents British workplace pension plans.

Read: What can Canada learn from Britain’s pension reforms?

He’s not the only one who’s unhappy. “The whole methodology used in these exercises is far away from the methodologies used today in the German regulatory environment,” says Stefan Nellshen, chief financial officer of pharmaceutical giant Bayer’s defined benefit pension fund that, like the rest of the participating schemes, took the test in the summer of 2015. “A harmonized . . . assessment cannot be a reasonable approach since it cannot take national specifics sufficiently into account,” he says.

But the different national frameworks are precisely why Europe needs a uniform methodology, says Gabriel Bernardino, chairman of the Frankfurt-based European Insurance and Occupational Pensions Authority.

“If you have different mechanisms in the different member states on how to value pension funds’ assets and liabilities, then you cannot compare them,” he says. “If you want to understand it on a higher level to make some comparisons, you need to have a more common methodology.”

More than 60 defined contribution pension plans and 140 defined benefit and hybrid pension schemes from 17 European Union countries took the pension authority’s test. The chosen plans were meant to be a representative sample of Europe’s pension industry. They had to make two assessments within the test: one under the common methodology and another one under their national rules.

Read: European Commission tells France to reform pensions

Unlike the national rules, the common methodology is more consistent with the market, says Bernardino. “You take the market value of the assets and a more realistic value of liabilities, discounting with a risk-free zero rate.”

Under the common methodology, before applying any shocks to defined benefit and hybrid portfolios, total liabilities exceeded assets by 428 billion euros.

After applying a sudden decline in assets under the common methodology, that deficit ballooned to 755 billion euros.

And with an increase in inflation and an abrupt decline in interest rates under the uniform assessment, the 428-billioneuro gap jumped to 773 billion euros.

But the common rule showed that defined benefit plans are relatively more resilient to a permanent decrease of 20 percent in mortality rates in comparison to the other two shocks. “While this is a big factor of stress for pension funds, they will have sufficient time to cope, provided that they incorporate these longevity scenarios in their own calculations,” says Bernardino.

Read: 3 pension risk transfer strategies

On the defined contribution side, the watchdog applied similar shocks to the projected pension income of three member groups: those who are 35, 20 and five years from retirement.

The results showed the oldest plan members have the highest pension savings but the least time to recover from declines in asset prices. The group most affected by low asset returns in the long term is young plan members. And low interest rates make it more expensive to convert the money they’ve accumulated into annuities.

Canadians can learn

Just as many Europeans dislike the idea of a standardized test for all jurisdictions, Ken Choi, Toronto-based director of investment consulting at Willis Towers Watson, doesn’t think it would find support in Canada, either.

“I’m not so sure I would support a regulatory or other means of enforcing it. The appropriate stress test for [one] pension plan may not be appropriate for another pension plan,” says Choi, whose firm conducts pension stress tests.

The Canadian Institute of Actuaries requires defined benefit pension plans to conduct only a basic stress test in which they measure the effect of a one-per-cent decline in the discount rate on their funding status on a going-concern and solvency basis for their actuarial valuation reports. Canada’s defined contribution plans don’t have to do any stress testing.

Read: DC lessons from across the globe

“While having more robust risk management is a laudable goal, we do have to balance that against the practical implications. If regulation becomes too burdensome, then we run the danger that we’re going to push more defined benefit plans to close and convert to defined contribution,” says Choi.

Still, he says Canadian pension plans can take a page from the European book and broaden the scope of their voluntary stress tests by including more extreme scenarios, such as a prolonged period of low interest rates and permanent increases in longevity.

“We’ve seen more activity in that area in recent years where pension plans are doing more than the traditional asset-liability modelling,” says Choi, noting the practice still isn’t widespread in Canada.

Results already known

Besides disliking the idea of a mandatory, standardized exercise, many Europeans also point out the pension authority’s test didn’t reveal anything new.

“I don’t want to be cynical — it’s a serious exercise. But to a large extent, this does tell us what we already knew, which is that if the economy is under stress, that also puts pension schemes under stress, ” says Walsh.

Pension plans across Europe have always known their top risks include interest rates, inflation, asset performance and longevity, says Alex Waite, partner at British law firm Lane Clark & Peacock LLP. “We’ve seen a lot of movement in certain countries — the U.K. being one of them — to try and reduce exposure to most of those risks, most notably, performance of assets and interest rate risk.”

Read: Kingfisher Group agrees medically underwritten buy-in

But Bernardino insists the results are useful. “It’s one thing to say, ‘I’m vulnerable to low interest rates.’ It’s another thing to understand if these low interest rates continue for the next 10 to 15 years, what is the amount of liabilities that I have extra and what is possibly the [extra] amount of contributions I as a sponsor will need to put in the fund,” he says. “That’s very valuable from a policy, supervisory and management perspective.”

A costly exercise

The stress test was also expensive, says Nellshen.

European extremes

In Belgium, Denmark, Germany, Finland, Norway, Slovenia and Sweden, 100 per cent of workplace pension plans are defined benefit.

In Bulgaria, Latvia, Poland, Romania and Slovakia, 100 per cent of workplace pension plans are defined contribution.

Source: European Insurance and Occupational Pensions Authority

The extra cost for pension funds came mainly from doing the math under the common methodology, which was “far too complex, too technical,” says Niels Kortleve, innovation manager at PGGM, which provides administration and asset management services to Dutch pension funds.

A lot of defined benefit pension plans across Europe, particularly smaller ones, had to hire consultants to help them with the calculations because they lacked internal resources. That led to an additional cost of anywhere from two to eight euros per member, says Kortleve.

“The costs were quite high and the benefits — if you get insights that you already know — are relatively small.”

Read: How multinationals are rising to global benefits governance challenge

But Bernardino doubts the common methodology calculations were much of a stretch.

“In most countries, on the assets side, there’s no issue because they already use market values or close-to-market values, so this is information that you have. And then, the valuation of the liabilities, instead of using what in many countries is used, what we’ve done is to [ask them to] discount liabilities according to a risk-free rate. That’s the only effort that, at the end of the day, pension funds had to do.”

Those calculations could be a burden for the smallest pension funds, but they didn’t have to participate, adds Bernardino.

Loss of autonomy

Another reason for the negative attitude towards the test is pension funds’ resistance to taking orders from European Union institutions.

“People just expect European rules to be unhelpful and bureaucratic,” says Waite.

The rules chip away at autonomy, adds Rob Bauer, a professor of finance at Maastricht University in the Netherlands. “The Dutch and many other countries want to keep their own pension system. Any European influence on that, like doing a stress test, will take a little bit of autonomy away. So it’s highly political.”

Read: Top global pension fund assets exceed US$15 trillion

Bauer notes many pensions funds are already subject to stringent regulations at the national level.

“We have a very tough regulator in the Netherlands, so the way we discount our liabilities [with a rate between one and three per cent] and what we promise is very tough, very market-driven, so we don’t need those occupational pension stress tests from Europe.”

Bauer adds it’s difficult to get a complete view of the issue without also testing the health of the various government pension plans, an area that’s even more politically charged for Europeans.

“Many countries see that as their political freedom,” he says. “Germany can decide to raise taxes… or they can decide to give their pensions less retirement income.”

A look at solvency and discount rates in workplace pension plans

Even if the European pension community disagrees with the European Union watchdog’s common methodology and the results it has produced, the question remains: Are European workplace pension plans healthy?

It’s a difficult question to answer. “Europe is a bunch of countries, and in every country in Europe, the pension system is different. And it’s not just the second occupational layer but also the first pillar and the third pillar,” says Rob Bauer, a professor of finance at Maastricht University in the Netherlands.

“In the last 15 years, many funds have gotten into some underfunding,” he adds.

“On the one hand, you have those countries that are in a terrible state with respect to pensions . . . and you have a bunch of countries like the Netherlands, to some extent Denmark, maybe some U.K. funds, in which the situation is good.”

For example, Bauer says Dutch defined benefit pensions are in better shape than their U.S. counterparts, which use a discount rate of about eight per cent to calculate liabilities. “It looks like the liabilities are really small when you do that,” he says. “So they totally overstate the funding, and even with this total overstating of the funding, they’re heavily underfunded. In the Netherlands, we use a discount rate of not even three per cent on average, between one per cent and three per cent, and we’re still funded.”

Compared to Dutch pension funds, some of Canada’s largest workplace defined benefit pension plans also use higher discount rates:

Ontario Teachers’ Pension Plan:

  • 2014 discount rate: 4.95%
  • Funded status in 2014: 105%

Ontario Municipal Employees Retirement System:

  • 2014 discount rate: 6.5%
  • Funded status in 2014: 90.8%

Healthcare of Ontario Pension Plan:

  • 2015 discount rate: 5.6%
  • Funded status in 2015: 122%

Get a PDF of this article.

Yaldaz Sadakova is associate editor of Benefits Canada: yaldaz.sadakova@rci.rogers.com.