- Consulting with Canadians -

Towers Perrin's Submission in Response to Finance Canada's Regulatory Framework for Federally Regulated Defined Benefit Pension Plans consultation:

September 21, 2005

Our Ref: /N

Ms. Diane Lafleur
Financial Sector Policy Branch
Department of Finance
L'Esplanade Laurier
20th Floor, East Tower
140 O'Connor Street
Ottawa, Canada
K1A 0G5

Dear Ms. Lafleur:

Towers Perrin is pleased to provide the Department of Finance with our comments on the Consultation Paper "Strengthening the Legislative and Regulatory Framework for Defined Benefit Pension Plans Registered under the Pension Benefits Standard Act, 1985" released in May 2005. We appreciate the opportunity to comment and trust that our input will be helpful in the formulation of further recommendations.

Towers Perrin is an international management and human resources consulting firm, with Canadian offices in Montreal, Toronto, Calgary and Vancouver. We advise organizations on human resource strategy and management, organizational effectiveness, compensation, benefits and communication. We provide actuarial services, asset consulting, administration and consulting services in respect of defined benefit and defined contribution retirement plans to clients in both the private and public sector. As such we are cognizant of the various challenges facing defined benefit pension plans in Canada.

Uniform Pension Legislation

Towers Perrin is an active advocate of uniformity of pension legislation among Canadian jurisdictions.

We note that some provinces (e.g., Quebec and New Brunswick) have already made temporary changes to their pension legislation to deal with some of the immediate pension funding issues mentioned in the Consultation Paper. Quebec, the Canadian Association of Pension Supervisory Authorities (CAPSA) and your department have issued consultation papers addressing permanent changes to funding requirements. We believe that coordination between the various jurisdictions in implementing any future changes would be most efficient and would result in better legislation

Context

In May 2004, Towers Perrin released a White Paper proposing a framework for building a 21^st century private pension system. The paper addresses several issues (regulatory system, funding, accounting, investment and plan design) that prevent defined benefit plans from evolving to keep pace with changes in our society. A copy of this paper is enclosed for your convenience. PDF Version

Our views on possible changes to the legislative and regulatory framework for defined benefit pension plans are guided by the following general principles:

• Pension promises should be clearly articulated and honoured by plan sponsors.
• Plan members should be provided with information that allows them to understand the nature of the "deal" with the plan sponsor. Disclosure requirements should be reasonable, and result in effective disclosure to plan members.
• Legislation should support and promote the fulfillment of these promises.
• Regulation cannot guarantee that pension promises are met, but it should create a framework that maximizes this possibility at a supportable cost.
• Legislation should not unduly discourage promotion of new plans and enhancement of existing plans.
• The proper balance must be maintained between the added value of additional regulation and the incremental cost of compliance.
• The costs of all aspects of plan governance should be maintained at manageable and supportable levels.

Our comments on the Consultation Paper are offered within the context of these general principles.

Our specific comments are set out below. Our comments follow the structure of the Consultation Paper.

A. Surplus

We believe that the current uncertainty surrounding surplus ownership has discouraged plan sponsors from contributing more than the minimum funding.

Pension surplus has long been a troubling issue for plan sponsors and employee groups, and related issues continue to grow in complexity. In most cases, plan sponsors are responsible for funding any plan deficit but often they must share surplus with plan members in the event of a plan termination or a surplus withdrawal while the plan is ongoing. This surplus/deficit asymmetry has led plan sponsors to adopt minimum funding policies. As a result, pension plans are less stable, with lower levels of funding.

We believe that resolving the ongoing issue of surplus ownership would improve the security of pension plan benefits and would help ensure the viability of defined benefit pension plans. We acknowledge that there are diverging views on this issue. Therefore it is crucial that it should be addressed within the legislative and regulatory framework. Any solution should recognize the voluntary nature of the private pension system and should strike a balance between the interests of the various stakeholders. We believe that there would be a better chance of developing a solution acceptable to all stakeholders if all issues affecting pension plans (not only funding and surplus) were addressed, similar to the approach we used in our White Paper.

The Dispute Settlement Mechanism for Surplus Distribution

The 2/3^rd consent level for each category of beneficiaries may not always be appropriate (e.g., a situation in which there are few active members or few former members). We suggest that the consent level be applied to all plan beneficiaries as a group rather than be applied separately to each category of beneficiaries.

The positive consent model is not the only model that could be used. The Government of Canada should consider a model in which a proposal is deemed to be accepted unless a specified percentage of members object to the proposal, such as that used in Quebec. This model has the advantage of being simpler to administer.

Since we contend that surplus is not ascertainable while the plan is ongoing, we support the use of a buffer with a view to ensuring continued benefit security for plan members. However, the requirement to maintain a buffer of 25% of the solvency liabilities while the plan is ongoing is too large and is not in any way related to the asset/liability mismatch risk specific to a particular plan. In some situations, such requirement would be an incentive for a plan sponsor to terminate its plan instead of withdrawing surplus. We recommend that the Government of Canada consult with the Canadian Institute of Actuaries on the appropriate level of such reserve.

We do not believe that an improvement in the dispute settlement mechanism for surplus distribution would create, by itself, a material incentive for plan sponsors to fund more than the minimum requirements as it does not recognize the nature of the "deal" between the plan members and the plan sponsor. The plan sponsor commits to provide a defined benefit and in almost all situations takes on the investment risk associated with that commitment. If the plan sponsor accepts the responsibility to fund any shortfall, the plan sponsor should be able to benefit from excess funding which arises.

The question of surplus distribution on partial termination is very controversial. The problem created by the Monsanto Decision is one of the disincentives to plan sponsors to fund plans beyond the minimum requirements. In our view, there are many inequities among plan members with a surplus distribution on a partial termination, including:

• Reduced benefit security for remaining members.
• Significantly different amount of surplus distributed on a partial termination than on a full termination, if any, as the existence of a surplus depends on several factors including interest rates and fund return.

Also, the criteria for declaring a partial termination are vague. There will always be situations where it is not obvious whether or not a partial termination should be declared and, if so, who is included in the affected group. This leads to disputes and leave plan sponsors open to challenge by affected members. This uncertainty is another disincentive for employers to maintain pension plans.

The concept of partial termination creates two standards: one for "regular" terminations and a more generous one for members affected by a partial termination. It is difficult to justify why a member who is terminated due to an event leading to a partial termination (e.g. plant closure, significant downsizing) should receive higher benefits (e.g., full vesting) than a member who is involuntarily terminated under other circumstances. Even a differentiation in benefits between voluntary terminations and involuntary terminations creates concerns around fairness and administrative issues. Any such distinction should be a matter of plan design and should not be imposed by pension legislation.

For these reasons, we recommend the elimination of partial terminations as a mandatory legislative concept. This would simplify the administration of defined benefit plans as it would eliminate the issues discussed above as well as the administrative burden of a partial termination. We note that Quebec has eliminated the concept of partial termination from its pension legislation since January 1, 2001; all stakeholders in Quebec seem satisfied with the results of this progressive change.

We support the use of letters of credit as an additional financing option because they provide additional flexibility for plan sponsors without affecting benefit security. Also, the use of letters of credit would somewhat alleviate but not eliminate the problems related to surplus/deficit asymmetry.

However, it must be noted that the potential triggering of a letter of credit can complicate the turnaround of a troubled company. The negotiation of restructuring with employees, creditors and shareholders must be conducted with a "sword of Damocles" dangling overhead. Also, the triggering of the letter of credit, if it occurs, affects the various stakeholders in different ways. For example, younger employees may lose their jobs while older employees get their pensions secured; the issuer of the letter of credit may be obliged to make a non-recoverable payment, while secured creditors may be unaffected. These sorts of differences can drive wedges between stakeholder groups and make the negotiation of restructuring more difficult.

The use of letters of credit should be optional only and should not be imposed on plan sponsors as there are a number of practical considerations (fees, governance, impact on credit line, etc.) that make their use unappealing for many plan sponsors. Because letters of credit do not represent a practical option for many sponsors, it would be inappropriate for pension regulators to impose higher funding of pension plans on the basis that the use of letters of credit is allowed.

We recommend that a letter of credit be permitted only to cover solvency amortization payments and should remain in effect until the amortization payments covered by the letter of credit have been remitted into the plan or a solvency surplus has emerged. If the plan sponsor does not remit the required contributions or does not renew the letter of credit, the letter of credit must be called by the trustee unless it is no longer required due to the presence of a solvency surplus.

A letter of credit should be recognized as an asset in the solvency valuation but not for the going concern valuation. The Government of Canada may wish to consider a limit on the use of letters of credit issued by any one institution similar to other investments in a pension fund.

The current five-year amortization period is considered too strict by a majority of plan sponsors as it results in excessive volatility in their contribution levels. We believe that a longer amortization period (e.g., 10 years) would be appropriate subject to certain conditions. These conditions for allowing an extended amortization period might include one or more of the following:

• The market value of assets is used to determine the plan’s solvency position.
• A limit on the amount that could be amortized over the longer period (e.g., 20% of the solvency liability).
• A solvency margin based on the plan’s degree of asset/liability mismatch is used in the determination of the solvency liability.

The last condition would require several changes. First, the pension regulator would have to consider minimum standards for determining the solvency margin, ideally in collaboration with the Canadian Institute of Actuaries. Second, the Income Tax Act would need to provide greater flexibility for plan sponsors who wish to build a solvency margin into their pension funds; plan sponsors should be able to continue to contribute, within limits, even if the plan is already in a surplus position. Finally, the asymmetry on the surplus/deficit issues would need to be addressed to encourage plan sponsors to build margins into their pension funds.

The conditions could be more restrictive for plan sponsors under CCAA or BIA – e.g. requiring consent from plan members or the unions representing them.

We would prefer a resolution to the surplus/deficit asymmetry issue. If the asymmetry issue is not completely resolved then we agree that alternative approaches be considered. These alternative approaches should provide plan sponsors with additional funding flexibility without reducing the level of benefit security in the plan. However, conventional funding coupled with the elimination of the surplus/deficit asymmetry is a better alternative.

One of these alternatives could be a separate fund that is tax-sheltered but different from a trust and is segregated from both the pension fund and the general assets of the plan sponsor. The pension fund would have a priority claim on this separate fund over other creditors in the event of the plan sponsor’s bankruptcy. Also, for the purposes of the actuarial valuations, the pension fund assets would include the asset value of the separate fund. A plan sponsor would contribute to the separate fund the solvency amortization payments and any contributions in excess of a specified minimum funding requirement. When combined with the pension fund, amounts in the separate fund that are in excess of a minimum cushion to protect against adverse events could be refunded to the plan sponsor or used for contribution holidays via a transfer from the separate fund to the pension fund. Assets in the separate fund that are not necessary to cover any plan obligation would revert to the plan sponsor on full plan termination, even if there is a provision in the plan text that surplus reverts to the plan members.

Another alternative would be a notional account that would work similarly to a separate fund described above except that the monies would be remitted to the pension fund and accounted for separately.

Amendments to the Income Tax Act (Canada) and the Pension Benefits Standards Act would be required to accommodate these alternatives.

We agree that disclosure should be encouraged so that plan members have a better understanding of what their benefits are, how the deal is intended to work and the degree to which their benefits are or are not subject to risk. However, such additional information should not result in unreasonable administrative expenses. Annual pension statements, bulletins or displays on the plan sponsor’s website could be used to communicate additional information. We note that some member education will be involved to make such disclosure more meaningful.

We disagree with the establishment of a solvency threshold for plan improvements. We have consistently taken this position in the past. We maintain that the combination of three requirements should suffice to protect plan members:

• Full funding on plan wind-up
• Retroactive voiding of unfunded improvements upon wind-up in a chronological order or in an approach similar to Quebec, and
• Clear disclosure to plan members of the funded position of the plan and the inherent risk of a solvency deficit.

If the concept of prescribed solvency ratio is adopted, pension plans should be permitted to make plan improvements provided that offsetting funding is provided at the time that the improvement comes into effect to maintain the plan’s solvency ratio at the prescribed level (or to fund 100% of the additional solvency liability, if less).

We note that the "void amendment" concept appears to be inconsistent with the treatment of newly established plans that provide retroactively past service to plan members.

We also note that the "void amendment" concept imposes restrictions on the collective bargaining process. The negotiation process could last several months during which time the solvency ratio of a plan could change drastically. The "void amendment" concept would affect the ability of the plan sponsor to budget for the negotiated plan improvements. To alleviate the uncertainties created by the "void amendment" rule, we suggest to simply base the test on the last-filed valuation report at the time the benefits are bargained rather than when the benefit improvement comes into force.

We support full funding on plan termination; however, we believe that the Government of Canada needs to take into consideration the fact that, if the plan sponsor is required to pay the cost to fully fund all benefits on plan termination, then the plan sponsor should also have entitlement to any surplus arising on plan termination.

Furthermore, we strongly recommend that the government consider amending the legislation to grant a statutory right to take contribution holidays in conjunction with granting a statutory right to surplus arising on plan termination. These financial risk/reward concepts should be enshrined in the legislation.

We do not believe that the requirement of full funding should be applied differently for financially vulnerable plan sponsors. However, there may be situations where a modified requirement would enhance the security of the plan members’ pension benefits. For example, the government could consider extending the funding of a deficit over 10 years for a financially vulnerable plan sponsor that terminates its plan while remaining in business. This modified requirement could be subject to various conditions including the consent from plan members or the union representing them.

Note that multi-employer defined benefit pension plans with negotiated contribution rates should be exempt from the requirement for full funding on plan termination. Contributions to such plans are limited to the negotiated amounts.

The Government of Canada is seeking views on the viability of a federal pension guarantee fund including any comments on its possible design, operation, and powers.

Experiences in the U.S. and in Ontario suggest that these funds do not work well in practice. Also, if these funds are operated in the same way as the pension funds that they are supposed to protect, they will be facing the same long term risks of failure. To be viable, a pension guarantee fund must be set up and operated the same way as an insurance company and be subject to the same regulations as insurance companies (e.g., strict matching investment policy). This means charging a risk-based premium for the protection which would reflect the main risk factors (i.e., the current deficit of the plan, the asset allocation of the plan relative to the liabilities, and the probability of default of the plan sponsor).

We believe that it would be difficult for practical and political reasons to set up and operate a pension guarantee fund the same way as an insurance company. It is most likely that employers with low risks will subsidize those with high risks. It has been suggested already that some US industries that compete globally (e.g., steel and airlines) are subsidized by the PBGC. There is also the threat that the World Trade Organization would consider a pension guarantee fund illegal if it provides uncompetitive advantages through subsidies.

We believe that pension legislation that deals with the surplus/deficit asymmetry and encourages proper funding would be more efficient than a pension guarantee fund.

Towers Perrin supports the government’s commitment to improving the security of pension plan benefits and ensuring the viability of defined benefit pension plans. We greatly appreciate the opportunity to comment on the Consultation Paper. We reiterate our encouragement to consider a more global approach to resolving major issues affecting defined benefit pension plans, as proposed in our White Paper.

We would welcome the opportunity to address any questions you may have concerning our comments.

Simon East Jacques Lafrance

Principal Principal

cc: Pierre Lemelin, Montreal

Mark Campbell, Calgary