While attending last week’s Pension Policy Forum in Toronto, it became clear that the focus of pension reform in this country has gradually been shifting.

Seven years ago, the focus was on Pillar 3, in particular on how to promote workplace pension plans. Then we eventually wrote off Pillar 3 as an effective way to improve coverage and instead pinned our hopes on Pillar 2 improvements, meaning an expansion of the Canada Pension Plan (CPP). While CPP expansion should not be written off just yet, more attention is now being paid to Pillar 1: old age security (OAS) and the Guaranteed Income Supplement (GIS). Pillar 1 plays a more pivotal, as well as a more controversial, role than we thought. In fact, GIS is the problem child within our retirement income system.

The one reason that Canada has a much lower poverty rate among seniors than most of the developed world is the additional income that lower-income seniors derive from the GIS. Most Canadians in the lowest-income quintile can expect to have higher disposable income after retirement than they ever had before retirement, even if they never save a cent in an RRSP or pension plan.

This represents both a strength and a weakness of our retirement income system. On the one hand, we should be happy to see lower-income Canadians achieve some retirement security, but are we overdoing it? Statistics Canada data indicate that 85% of recent retirees in the lowest-income quintile saw their disposable income rise by at least 15% in retirement, and in a large number of those cases, it rose by 25% or more.

The cost of OAS and GIS will climb over the next 10 years as baby boomers retire in large numbers. And, as rising healthcare costs squeeze all other government spending, the high replacement rates due to GIS will come under greater scrutiny. Regardless of our political leanings, it is difficult to reconcile net replacement rates of 125% or more as being appropriate. For those who believe that low-income retirees will need every cent of this money, it begs the question why they didn’t need more when they were still working.

These high net replacement rates also make it challenging, to say the least, to design earnings-related programs such as an expansion to the CPP, the proposed Ontario Retirement Pension Plan (ORPP), or auto-enrolled pooled registered pension plans. We know there is a retirement income gap in this country, though it has only recently become clear that the gap is concentrated on middle- to upper middle-income households. How do we close that gap for middle-income earners without exacerbating the problem for the lowest-income group?

For instance, the proposed ORPP would require Ontarians to contribute 1.9% of earnings up to $90,000. To date, no lower-income threshold on contributions has been proposed. If the Ontario government implements an ORPP without such a threshold, it will be doing three things:

  • making low-income working Ontarians even poorer;
  • increasing the number of retirees with net income replacement rates of 125% or more; and
  • inadvertently reducing the federal government’s payments toward the GIS.

It is a fair bet that none of these outcomes is intentional or desirable. On the other hand, if an income threshold of say $30,000 is imposed, it raises its own problems. Perhaps one worker in 10 will change jobs in a given year and hence have income split between two employers; a threshold would mean that 10% of workers wouldn’t have coverage in a given year for that reason. Similarly, workers holding down two or more part-time jobs would see most or all of their income exempted from contributions. A threshold also creates other problems.

  • Many workers who are barely above the contribution threshold would be contributing tiny amounts that may cost more to collect and administer than it is worth.
  • Workers who spend half their career with incomes hovering around the contribution threshold before rising above it would have suboptimal coverage.
  • Then there is the problem of optics: it’s hard for a government to introduce a new pension program that benefits just higher-income workers.

This is equally a problem for an expansion of the CPP. As long as these thorny problems exist, the pension reform debate still has some distance to go before we reach closure.

Fred Vettese is chief actuary of Morneau Shepell. These are the views of the author and not necessarily those of Morneau Shepell or Benefits Canada.
Copyright © 2017 Transcontinental Media G.P. Originally published on benefitscanada.com

Benefits Canada Newsletter

For the latest industry news and opinions, sign up for our daily newsletter.

See all comments Recent Comments

Retail Investor:

Quite correct but there is also another wrinkle being ignored. Any forced savings in either an expanded CPP or ORPP, or PRPP, etc will come at the expense of the low-income-earner’s contributions into a TFSA.

Most people are getting the message that in the bottom tax bracket, outcomes are better when saving in a TFSA (vs an RRSP). But all these new proposals have the same tax impacts as an RRSP. The income in retirement from these plans will be clawed back at a 50% tax rate (additional to the statutory rate).

These plans force people out of their better option (TFSA) and into a worse deferred-tax plan. See what a typical $40,000 wage earner income looks like in retirement (assumes 20% CPP benefit).

For this person the next $10,000 of pension-type income will be taxed at 72.5%. If allowed to use the TFSA there would be 0% tax. Lower income people would have smaller CPPs and more of their expanded CPP and ORPP and PRPP income will be taxed at that 72.5%.

This problem could be fixed if contr and w/drawals to these new plans bypassed the Tax Return. Have one tax rate for everyone. The account administrator deals directly with the government to both fund the ‘tax refund’ and collect on withdrawal.

Friday, October 17 at 10:13 am | Reply

Add a comment

Have your say on this topic! Comments that are thought to be disrespectful or offensive may be removed by our Benefits Canada admins. Thanks!

* These fields are required.
Field required
Field required
Field required