Recent Posts

Sunday, March 9, 2014

Book Review: The Third Rail

In "The Third Rail: Confronting Our Pension Failures" (2013), Jim Leech and Jacquie McNish discuss the problems with the Canadian pension system, as well as offer some solutions. The background material is international in scope, but the discussion of solutions is Canada-centric. The book could possibly be interesting for readers outside Canada who are interested in the political aspects of pension reform, but the discussion of solutions would be less relevant, as the gritty details vary from country-to-country. In this review, I try to keep my discussion relevant for an international audience.

Book Description

Jim Leech was the CEO of the Ontario Teachers' Pension Plan (known as "Teachers") when the book was written, but he retired in November 2013. Teachers is one of the largest asset managers within Canada (about C$130 billion in assets at the end of 2012). Jacquie McNish is a senior writer at the Globe and Mail, and the author of other best-selling books. Accordingly, the book is slick, and offers a good view of the pension situation from an establishment perspective.

The book consists of 5 chapters:

  1. Canada's Pension Promise - a history of pensions. It explains how Canada was a pioneer in the initial developments of pension plans.
  2. New Brunswick - the story of pension reforms in New Brunswick (a Canadian Province on the Atlantic coast, for those of you who are not experts on Canadian geography).
  3. Rhode Island - government pension reform in the state of Rhode Island, in the United States. Includes a history of the development of the state.
  4. The Netherlands - the story of the Dutch pension system since its development after World War II.
  5. Solving Canada's Pension Crisis. The book also mentions reforms in some other countries in this chapter, as models for Canadian reforms.

The book is largely non-technical in its discussions. So the discussion of the excitement around liability discounting - which roiled the euro area swap market - is only mentioned in passing. This probably makes good commercial sense - a book that covered the details of euro area swap trading would likely sell literally dozens of copies. What the book mainly provides the reader is a discussion of political strategies around pension reform in the first four chapters, as well as a journalistic history of pension development in a few countries.

The book is aimed at a general audience of Canadian readers. For someone outside of Canada (the bulk of my readers), the book could be a useful resource if you have already developed an interest in pension reform. The book provides a quick overview of strategies that have been used in a number of countries. However, it does not give you the tools to analyse the details of reforms, so if you are already a specialist in the area, you may find the treatment too simplified.

I do not have much knowledge of the details of the histories presented in the book (other than what was going on in euro swap markets at various points), so I will not comment on those parts. I will instead look at the discussion of solutions.

How Big Is The Problem?

The problem with writing for a general audience is that if you suppress all the messy details of the analysis, you cannot gauge how big the problem really is. In a previous article, I gave a summary of why I have ignored the alleged demographic fiscal time bomb that faces the developed countries - the expectation is that fiscal policy will be gradually tightened in response to increased transfer payments to the aged, as it makes no sense to tighten policy now. Some of the pension problems identified by the authors can be lumped in with this general "demographic" analysis. I will be writing out my own arguments elsewhere, so I will not discuss this point in detail within this review.

My concern with the analysis within the book is that it is written from the point of view of a pension plan officer or an actuary (which is not surprising, given that one of the authors presumably plowed through hundreds of detailed actuarial reports during his career [note: I updated the phrasing here since the previous text was not clear.]). In fact, that is a general defect of almost all analysis of pension plans, as they are typically written by actuaries. Throughout the book, the critical importance of funding ratios is assumed. However, funding ratios only matter if there is a risk that the sponsoring entity (employer) will go bankrupt; if we assume that the employer is a going concern, the funding ratio loses some of its significance.

If we assume the employer is a going concern (will not go bankrupt), a defined benefit pension plan can be split into two components:
  1. a pay-as-you-go plan, where the inflows from the employer and current employees are used to make payments to retirees; and
  2. a pool of assets which is being accumulated, which are meant to generate cash flows to meet future deficits on the pay-as-you-go portion of the plan.
The first pillar of the Canadian pension system is a government-provided pension and old age supplement, which can replace most of the income of the poorest-paid workers (up to C$30,000 annual salary). Outside of Québec, the Canadian Federal Government runs the program. Since Canadian dollars are liabilities of the Canadian Federal Government, it makes little sense to worry about the Canadian Federal Government running out of them, and so it should be analysed as a going concern. (The Government of Québec is not in that position, and so it needs to follow the same strategies that corporate plans do, which means that financial assets need to be accumulated within a plan.)

Properly analysed, the "funding" of a pension fund by the Canadian Federal Government is really just an emission of debt to purchase private sector financial assets. In other words, it is attempting to arbitrage the capital markets; it is assumed that private sector assets will return more than the debt it issues. The question is: does it make sense for the government to issue debt to buy up x% of corporate securities outstanding (in order to get a junior claim on x% of the earnings of public corporations) - or to just increase the corporate tax rate by y% (in order to get a senior claim on the first y% of earnings of all Canadian corporations)? This is a question of public policy objectives, and is a judgement call by economists, politicians and citizens. Actuarial analysis gives us little insight into these trade-offs. (I explain these assertions  in more detail in this later article ("Government Pensions Are Always Pay-As-You-Go").)

As a result, the discussions of the first pillar - the basic income provided by the government - actually ends up outside of the scope of the book. If there are reasons to be concerned about those demographic trends - which I grant is possible - it requires a deep analysis of the economy, with macroeconomic models that actually work (which is not a property of almost all the models being used by various bodies to do the analysis). 

Supplementing Government Transfers

The real debate is around supplementing the retirement income for those with annual incomes greater than $30,000. Even for this, the solutions discussed within the book are somewhat light on detail.

For example, it is stated that the voluntary retirement plans (tax shelters, known as RRSP's and TFSA's in Canada) have failed. I agree with this assessment, but more detail to back this up this presumably controversial view would have been nice. The reason the programmes appear to be a failure is that not everyone has taken advantage of these plans, and so a good number of citizens face rather catastrophic drops in income after retirement. How is society going to react to these citizens' woes is not obvious, to say the least. 

The book discusses the use of compulsory and/or automatic enrolment in defined-contribution pension plans. (In a defined contribution plan, you own a specific pool of assets, and your pension income is derived from your asset returns. In a defined benefit plan, the plan sponsor guarantees a certain income based on your working salary, and years with the plan. The sponsor takes the investment risk.) Although I am in favour of people saving more for retirement, I dislike some of the implications of the government forcing people to buy particular private sector services. For example, although I think that we bloggers provide extremely valuable services to society, it would make little public policy sense for the government to force people to divert 1% of their incomes towards us. Similarly, one has to be concerned about the government forcing people to hand money over to favoured private sector financial "partners".

In addition to this concern, government-mandated savings brings with it an implicit government guarantee. What happens if a good number of people stick all of their assets into a Treasury Bill fund, and the central bank follows some crazy macroeconomic theory that tells it to keep short rates near 0% for a couple of decades? In the end, the government is probably going to have to bail out people whose funds have underperformed on humanitarian grounds. In any event, such a plan is too late to do anything about the bulge of the population that is near retirement already.

Additionally, if the government creates a forced pool of savings that cannot be touched until retirement, it is not useful for dealing with personal financial crises before retirement. Such earmarked savings plans are actually argued against in the book "The Two Income Trap" (which I recently reviewed). If people treat such forced savings as being equivalent to voluntary savings that you can tap into, the financial situation of many households may be worse as a result of the reform, not better.

The authors repeatedly stress how much better run the large pension funds are than Canadian funds marketed to retail clients. I will note that I am biased - I was recently a low-level functionary at another large Canadian pension asset manager. I agree that the cost of many Canadian retail mutual funds is too high. However, even I think the authors laid it on a bit thick, for example when they complain about the fragmentation of having millions of individual funds. The reality is that once the back office infrastructure is in place, a mutual fund manager does not care whether it has one client or millions. The only cost of increasing the number of clients is on the support side, which is also faced by the large pension funds.

The book talks glowingly about the use of annuities, about which I have reservations as a general solution. The advantage of a defined benefit plan is that longevity risk is spread out amongst plan members; people who keel over soon after retiring consume very little of plan assets, allowing more benefits to be paid to long-lived members. Someone with a defined contribution plan has to build into their spending plans an assumption of living long, which means that many will die with a lot of assets. This is good for heirs, but it means that retirement income is well below what it would have been if you knew your life expectancy in advance. (Teachers' had a particular problem with longevity risk; most of their plan members are sensible women, whose life expectancies were well beyond earlier projections.) In an annuity, a sponsoring organisation (e.g., an insurance company), pools people together so that retirement assets are matched better to average life spans. The idea is that you can buy annuities with your defined contribution plan assets, giving you a guaranteed income for life (like a defined benefit plan).

The catch is that annuities are expensive. Insurance companies are forced by regulators (and shareholders) to price longevity risk conservatively, and so the annuity providers catch a lot of the upside gained by pooling mortality risk. I would argue that annuities can be a good product for individuals, depending on circumstances. However, forcing everyone to subscribe to an inherently expensive financial product is a spectacular transfer of resources to the financial services industry that would need to be carefully analysed.

Corporate Pensions

I also have some reservations with the discussion of corporate pensions within the book. Once again, the problem is thinking like an actuary or financial analyst, and not considering a "going concern" point of view.

Corporate defined benefit pensions made a lot of sense when:
  1. Corporations employed a lot of people.
  2. Corporations followed conservative financial plans, with fortress balance sheets and AAA credit ratings.
  3. Corporations actually paid income tax.
A pension fund does not pay tax, and so it allows corporations to accumulate financial assets in a tax-advantaged vehicle. Even if the sponsoring corporation could not directly touch the pension plan assets, any surplus in the plan provides it with scope to enhance compensation, giving it a competitive advantage. If the above conditions are true, the calculation of the cost of a pension fund needs to take into account the advantage of pre-tax returns of sheltered investments over the after-tax returns on non-sheltered investments. Once this adjustment is made, the cost of a pension plan is a lot smaller than it appears than when the plan is analysed a stand alone entity. However, the problem is that none of the above assumptions hold any more. 

Once this is taken into account, the authors' complaints about the Canada Revenue Agency blocking reforms or allowing plans to accumulate huge surpluses ring hollow. Corporations were never going to be allowed to accumulate huge war chests of tax-sheltered assets, even if that makes the plan managers' jobs easier. 

Given these changes in corporate business practices, private defined benefit pension plans make limited sense. Creating pooled managers for many sponsoring organisations is one possibility suggested by the authors. However, it will be necessary to think very carefully about the national interest, such as the value of the implicit or explicit government backstop those pooled plans will end up with, and the real resource cost of running those funds.


Although the book is interesting, it needed to place the concerns about pensions within the overall economic situation facing the country. It may be that their suggested reforms make sense, but it would still be necessary to look at other analysis to draw that conclusion.

I have spent very little time looking at the trends in pensions, but I expect that I will increasingly look at demographics over time. For more analysis of pensions, I recommend reading The Pension Pulse, a blog written by a hard-working ex-colleague of mine, Leo Kolivakis. He discusses the trends in asset management, as well as the problems associated with reforms, for the North American pension industry (with an accent on Canada).

Final Note: The book is available at: The Third Rail: Confronting Our Pension Failures ( affiliate link).

(c) Brian Romanchuk 2014

No comments:

Post a Comment

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.