Recent Posts

Wednesday, February 17, 2016

Difficulty Of Extending Universal State Pensions (Part 2)

With the coverage of defined benefit pensions falling, other private sector means of providing retirement income are tending to fall short. Although the financial products to provide a retirement income exist, they end up under-utilised in practice. The alternative is for the central government to step up its provision of retirement income replacement. However, the wide income disparity within the "middle class" means that it is difficult to provide wider coverage without creating structural changes to the economy.

This article builds on the previous, which outlines how Canadian retirement income provision is set out. I am using Canada as an example, but I believe that many other developed countries will have similar issues. (I was expecting to publish this article quite a bit earlier, but I ended up commenting on an incipient financial crisis, and faced some surprise home renovation work...)

What Is The Problem?

In the earlier past-war era, defined benefit pensions typically provided retirement benefits that "replaced" working age income (where replacement is typically thought of as 70% of working income, as one presumably has less expenses when retired). The decline of defined benefit pensions has made it extremely difficult for middle class households to arrange that income replacement.

Workers have been pushed to invest in funds that are under their control (defined contribution pensions; retirement investment accounts). The problems with these replacements include.
  • Historically, Canadian retail investment funds have had too high management fees. Since businesses took no responsibility for investment results, firms had limited incentive to push for more cost-effective fund management.
  • Investment allocation decisions were under the control of individuals, and some made costly mistakes.
  • It is unclear what savings rate is needed to hit retirement income targets, and there was a widespread tendency to under-contribute.

Is This A Problem?

In a country where most people appear to consider themselves "middle class," not taking middle class complaints seriously is a recipe for a short career in politics. However, the income distribution politics here are awkward.

The Canadian universal pension system and Old Age Security (OAS) provides income replacement for the working poor ($30,000 pre-tax annual income or less). There is a supplementary income programme for those with extremely low incomes, creating a de facto minimum guaranteed income that depends upon the household's situation. The system may not be perfect, but it appears to deal with the problem of absolute senior poverty. Providing such a safety net is politically popular, and the programmes are largely untouchable.

However, replacing the income of the "middle class" is no longer just keeping seniors' incomes in line with the working poor; they would easily be above those incomes. I will reproduce the household after-tax income distribution table (based on CANSIM table 206-0031) from the previous article. (Please see that article for some of the qualifications I had about this table; it does not perfectly fit the analytical needs here.)

Average after-tax income ($)
All deciles 66,600
Lowest decile 9,200
Second decile 21,500
Third decile 30,900
Fourth decile 39,600
Fifth decile 48,700
Sixth decile 59,200
Seventh decile 71,900
Eighth decile 88,300
Ninth decile 113,100
Highest decile 183,600

If someone earned $80,000 a year (pre-tax), why should the government guarantee a retirement benefit which is a multiple of those in the second/third income deciles? Undertaking such a guarantee implies a huge expansion of the programme, and it raises the question of why the government is entrenching inequality. This inequality makes the programme politically vulnerable should it prove to be costly.

The path of least resistance politically is to push people to increase contributions to defined contribution pension plans. As I discussed earlier (link #1, link #2), the private sector has no problem supplying defined contribution pensions and annuities to provide income replacement. Since the private sector is providing the level retirement income, the government can wash its hands of the unfairness associated with income inequality. The only real downside with this strategy is that there is no way to determine how much the government should push workers to save.

Potential Changes To Increase Retirement Income

There are two ways for the government to ensure that retirement incomes will be more satisfactory for retirees.
  1. The government can increase transfers to the elderly.
  2. The government can increase the size of the universal defined benefit pension plan.
The first option is discussed within the rest of this article; the second will be discussed in the next part.

Increased Transfers

If the safety net is seen as inadequate, there are two ways of increasing transfers to seniors to shore it up.
  1. Targeted (means tested) transfers. Increase the supplements for those households with very low incomes. The advantage of this method is that spending is aimed at those households with the greatest need, reducing the footprint of the programme. However, it requires a bureaucratic steps to determine that households qualify for the transfers.
  2. Increase The OAS. The OAS is essentially a basic income for seniors; it could be increased to generate an across-the-board improvement in senior income.  The fact that it is universal is presumed to make it more attractive politically. (I think that is wishful thinking, but many backers of basic incomes would disagree.)
The first option (targeted transfers) is straightforward, and would have a limited economic impact since the number of households covered is small (unless the housing market deteriorates very badly). As a result, I see few problems with such an extension. However, this option is purely an anti-poverty programme, and it does not help middle class households replace their working income, which is the subject of this article.

The second option would be a step towards income replacement for the middle class, but unfortunately the step is going to be too small. My guess is that $12,000 per year would be the upper limit for these transfers (up from around $7,000), unless an income guarantee was introduced for the entire population. Introducing an income guarantee for all ages would be a step that would require a rethink of the entire tax structure. It would also introduce a large structural change to the economy, and so we end up in a discussion that is far away from a discussion of pension reform.

If we keep the increase near my "feasible limit" ($12,000), it would help the situation for poor households, but would still be irrelevant for households in the upper half of the income distribution. In other words, it would improve the retirement income for some households, but it would not help the entire middle class. Broadly speaking, the difficulty of income replacement would still exist for most households.

It should be noted that although I believe an increase in the OAS to $12,000 would not actually be that radical a step, I believe that it is safe to say that I am in the minority. An immediate increase to that level in the current environment would create a political firestorm. Rating agencies, Bay Street economists, and bloggers would be in an uproar over the cost of the programme. It would take some deft political work to implement such a step.

Increasing transfers would have an economic impact that is not too troublesome in the current environment. Canada is facing weak final demand, and so some increased spending would help push the economy back towards its potential. Given that an increasing wave of citizens are hitting 65 year of age, this would create a structural steady increase in demand. This would eventually necessitate tax hikes to compensate for this demand. These taxes will probably be linked to the increased payments, reducing the popularity of the programme amongst younger Canadians.

Finally, one advantage of extending transfers to the elderly is that it can reduce the incidence of poverty amongst retirees, and people who are near the end of their working careers. Increased savings in a defined benefit (or a defined contribution) pension plan is not going to make a significant difference in the retirement income for someone who is going to retire in one year. With respect to Baby Boomers on the edge of retirement, the only plausible way to deal with increased elderly poverty is via increased transfers. The other major alternative solution (increasing universal defined benefit pensions) will only matter for people who have sufficient time ahead of retirement to build up increased benefits.

Expanding The Canada Pension Plan

The next article will discuss the issues associated with expanding the Canada Pension Plan (CPP), which is a universal defined benefit pension plan. The advantage of expanding the CPP over straight transfers (as discussed here) is that it is more sustainable from a political perspective, however, increasing the size of the programme creates other problems.

(c) Brian Romanchuk 2015


  1. "If we keep the increase near my "feasible limit" ($12,000)"

    The thing we have to understand is why direct transfer (with possible associated taxation) is 'unfeasible'. Yet the higher taxation of pension contributions plus payment by a private provider is feasible.

    Even though they are essentially the same thing - with the private provider (or public middleman like the UK NEST) being more expensive because of higher overall admin costs - and tying up real resources doing something that is essentially pointless.

    Then perhaps you can see a middle way where people 'contribute' to a public pension savings account where they can see their 'capital sum grow', and then when they retire they get a payment 'based on that capital sum'.

    Whereas of course the whole thing is a complete political fiction and the numbers in the 'capital sum' account are just numbers that grow ever year by 5% or 7% or whatever the growth rate is to enable the end transfer that was going to happen anyway.

    So you create a facsimile that looks like a pension fund, but it doesn't actually invest in anything real. Which would make it the same as all current pension funds :)

    1. What I am talking about in this section are transfers that are unrelated to working age contributions (in whatever form they take). Essentially, an income guarantee for oldsters.

      My "feasible limit" is my feeling of what would work politically in the long run. As an example, assume that the OAS is raised to $25,000 per adult, and somehow price and wage levels did not move. A couple who both turned 65, and somehow never managed to work, might be raking in more than a couple who were both working at low end retail jobs, and who were actually paying a fair amount of tax. This is not going to make the people who are working particularly happy. And once the programme looks vulnerable, things snowball against it - young people will quite rightfully view it as unsustainable, and that they will never see similar payments. The only way the politics works is if you embed the payments within a larger "basic income" programme; benefits for seniors may be higher, but not disproportionate.

      To be sustainable politically, the benefits need to be tied to working age contributions, as the benefits are no longer a gift; they are a right. That's what the CPP programme does, and I will be writing about it next. I wil then be trying to respond to your points. I do not exactly disagree, but the government is still somewhat forced to act like a private sector pension, and hoard financial assets, as a result of how the financial flows work.


Note: Posts are manually moderated, with a varying delay. Some disappear.

The comment section here is largely dead. My Substack or Twitter are better places to have a conversation.

Given that this is largely a backup way to reach me, I am going to reject posts that annoy me. Please post lengthy essays elsewhere.