Longevity risk transfer markets: market structure, growth drivers and impediments, and potential risks


Date de parution : 10/18/2013
Personne(s)-ressource(s) : Ethan Kohn

October 18, 2013

Joint Forum of Basel Committee on Banking Supervision
Secretariat of the Joint Forum (BCBS Secretariat),
Bank for International Settlements
CH-4002 Basel, Switzerland


Dear Sir/Madam:

The CLHIA appreciates the opportunity to comment on the Joint Forum of the Basel Committee on Banking Supervision paper entitled, "Longevity risk transfer markets: market structure, growth drivers and impediments, and potential risks".

Established in 1894, CLHIA is a voluntary trade association that represents the collective interests of its member life and health insurers and reinsurers which, together, account for 99% of the life and health insurance in force in Canada. Our members contribute to the financial well-being of millions of Canadians by providing a wide range of financial security products, including over $3.9 trillion of life insurance coverage. During 2012, life and health insurers made benefit payments of $66.4 billion, or roughly $1.3 billion a week, to policyholders and beneficiaries. As well, and of particular note for this consultation, in 2012, Canadian pension plans paid $1.1 billion in premiums to purchase annuities for their members.

We commend the Joint Forum for having drafted a consultative document which is comprehensive in nature and which canvasses the key considerations in this nascent market. The paper touches on important issues such as the current size of the market; the potential size of the market; the types of product by which longevity risk transfer ("LRT") can be effected; the risks associated with each such product; and, finally, the document sets out a series of recommendations to promote the development of a secure, fair and well-functioning market for
such products.

We support the recommendations put forward by the Joint Forum and, by a copy of this letter, strongly encourage Canadian regulators to move forward expeditiously in implementing them
with a view to protecting the retirement savings of Canadian pensioners, while promoting the solvency of financial institutions that will be taking on longevity risk.

In our comments, we provide some context for the Canadian marketplace, we discuss the long-term nature of the promise being made, we look at institutional capability in evaluating longevity risk, we consider issues involving a level playing field, we touch on issues surrounding systemic risk and, finally, we comment on the recommendations in the paper.

Background -- Canada

To provide a Canadian perspective, pension assets in Canada are estimated at over $2.4 trillion (at the end of 2011), with employer-based plans accounting for over half of this total at over $1.3 trillion.

Annuity buy-outs have been around in Canada for over 100 years, allowing plan sponsors to transfer investment and longevity risk to an insurance company. Annuity buy-outs have been the most common transaction historically, but buy-ins are growing in popularity and in the last five years there have been 15 annuity buy-ins covering about $370 million of pension liabilities. The largest annuity buy-out transaction in recent years was for over $400 million in September 2011 and the largest annuity buy-in transaction was for $150 million in June 2013. One reason for the growing popularity of buy-in transactions relative to buy-out transactions is that, in a low interest rate environment with many pension plans being underfunded, the buy-in does not require a large up-front lump sum from the pension plan to bring the plan into a fully-funded position.

To date, no longevity only risk transfer transactions have been completed in Canada, however there are signs showing growing interest in Canada from some pension plans wanting to transfer only their longevity risk.

It is estimated that Canada saw $1.1 billion in risk transfer deals in 2012, a significantly lower share of total pension plan assets than was realized in either the U.S. or U.K. There are currently at least eight insurers active in the Canadian pension risk transfer market and it appears that Canada is poised to follow the lead of these more developed markets. The factors at play include:

    - The full range of de-risking solutions offered around the globe is now available in Canada
    - All of the major pension consulting firms have de-risking teams
    - New accounting rules came into effect in 2011 and 2013 making pension deficits and pension risk more transparent, which increases the attractiveness of pension risk transfer transactions
    - The Towers Watson 2012 Survey of Pension Risk reports that 16% of private sector plans are thinking about buying annuity protection
    - 54% of plans are considering changing their investment strategy
    - With the recent introduction of proposed new mortality tables and longevity improvement scales for Canadian pension plans, there is a heightened awareness of longevity risk among Canadian plan sponsors
    - There are several large Canadian plans ($1 billion plus) that are currently seeking solutions and have issued Requests for Proposal in that regard.
For all of these reasons, we believe that the Joint Forum paper is timely and relevant for Canada, and the considerations and recommendations found in the paper should be of interest to Canadian policy makers.

Implications of the Long-Term Nature of Longevity Risk

The paper highlights the importance of institutional expertise in managing the risks associated with longevity risk protection. As experience has shown, underwriters must take special care to price appropriately products that carry long-term risk, because the long duration of these products means that the impact of inaccurate projections will be more pronounced. Of all financial institutions, insurance companies tend to have the most experience managing long-term assets and liabilities on their balance sheets.

Not only do insurers and reinsurers have the requisite knowledge, experience and systems in place to manage long-term risk, they also have a potential natural hedge for annuity exposure in the life (re)insurance which they write. We agree with the statement on page 9 that:

      So far most ultimate "buyers" of longevity risk have been life insurers and reinsurers for whom longevity risk may provide a partial hedge for their insurance exposure. This is because the two risks potentially offset each other -- life annuity liabilities increase when annuitants live longer whereas life insurance liabilities decrease.
On a related point, the paper explains that longevity swap transactions tend to mature prior to the closure of the underlying pension plan, which creates rollover risk. This gives rise to the concern that market participants may lack the wherewithal to effect a genuine risk transfer using a contract which lasts for a period of time which is significantly shorter than the length of the underlying contract. This rollover risk may not exist with insurance-based longevity transactions, as insurers may offer contracts that last for the lifetimes of the covered members.

Longevity Risk Assessment

The paper describes a number of potential ways in which longevity risk may be transferred, including buy-in transactions, buy-out transactions, longevity swaps and longevity insurance. Buy-in, buy-out and longevity insurance transactions all involve an assessment of the lifespan of a group of individuals, which assessment falls squarely within the unique skill set that insurers possess.

With respect to index-based longevity swaps, the difficulty, as the paper points out, is the tension between establishment of a standardized, liquid, broad-based index, on the one hand, and indexes tailored to the underlying population in a given transaction. A broad-based index would promote a liquid market, but would carry with it basis risk since the particular group being insured would not have identical characteristics to the index. A more tailored index would suffer from a lack of liquidity. In short, all LRT transactions will necessarily require an assessment of mortality risk of a defined cohort whether a buy-in, buy-out, longevity insurance or longevity swap transaction. The counterparty to the transaction, including under longevity swap, will be underwriting mortality risks.

The paper also points to certain attributes which suggest that insurance may be a preferred vehicle for offering longevity protection. As noted on page 7 of the paper:

      In swaps, the risk may be distributed more broadly, yet may return to the swap intermediary (which could be an investment bank) in case of a tail event (e.g., cure for cancer) …. There is reason to suspect that, because of more stringent regulation, insurance-based transactions lead to more complete risk transfer, as a result of lower counterparty risk.
The Joint Forum supports the development of more detailed and up-to-date mortality and longevity data to aid in measurement and management of longevity risk and to help reduce the basis risk created by transactions based on out-of-date or inappropriate mortality data or on standard indices. As the most successful LRT market, the UK has up-to-date, commonly-accepted mortality experience tables and provides regular updates to these tables. The CLHIA urges the Canadian Institute of Actuaries to follow through in providing pertinent pensioner mortality studies on a regular basis for pension plans and group annuities issued by insurance companies, as well as guidelines in projecting future mortality improvements to Canadian practitioners on a regular basis.

Consistency of Regulatory Treatment

The consultation paper delves into the reasons why certain forms of longevity risk transfer may be more attractive than others. From the perspective of a pension plan, longevity risk transfer can provide relief from the requirement to hold reserves and can reduce or eliminate undesirable risks such as longevity and investment, in much the same way that insurers reinsure risks which they underwrite. However, as long as pension plans will be subject to less stringent regulation than insurers (e.g., they can run temporary funding gaps, they face less stringent actuarial requirements, they are not required to hold capital, and they are allowed to use higher discount rates than insurers) transfer of risk will remain less attractive.

Systemic Risk

Another aspect touched upon by the consultation paper is the potential for systemic risk as the market expands and matures. As was demonstrated in 2008 with respect to credit risk transfer, the buildup of concentrated leveraged positions can be very dangerous, particularly where there is a lack of transparency. This can be the case with longevity risk transferred through index-based swaps. As transactions proliferate, there is the possibility that credit markets may obscure true risks being transferred, that interdependence of certain events (e.g., a cure for cancer) leave the entire market at risk of failure from one event, and that the stability of the financial system as a whole could be placed at risk. Finally, as articulated on page 17 of the paper:

      Due to inconsistent treatment of longevity risk across financial sectors, longevity risk may accumulate where it is least regulated or not regulated at all. There may also be a tendency that transfers take place from parties that have risk expertise to parties that have less risk expertise. Consequently, risks may accumulate where they are least understood and are monitored and managed less effectively.
Each of these concerns is highlighted in the paper, and the CLHIA shares these concerns expressed by the Joint Forum.

Policy Recommendations

The CLHIA supports the eight recommendations found in the consultation paper, and in this section, we would like to elaborate on each of these in turn:

    - As we have noted in the body of our submission, we agree with Recommendation #1--that supervisors should communicate and cooperate on LRT to reduce the potential for regulatory arbitrage. While the paper notes the potential for arbitrage due to differing rules governing pension plans compared to (re)insurers, the CLHIA also believes that all transferees of longevity risk should operate on a level playing field, particularly with respect to the collateral, reserves and capital that they require to support their obligations.

    - With respect to Recommendation #2, our members note that while there is still more ground to cover in terms of fine tuning the ability to predict more accurately mortality improvements, insurers and reinsurers have the most sophisticated understanding of longevity risk (even more than pension funds, which are often an adjunct to the main business of their sponsor and tend to be focused on investment performance of the fund).

    - We strongly support the recommendation that policymakers should review their explicit and implicit policies with regard to where longevity risk should reside to inform their policy towards LRT markets. We believe that this review should occur expeditiously because the market is evolving rapidly. Participants should be able, with some certainty, to enter into contracts which control longevity risk, and regulators ought not to be presented with a foregone conclusion where their only option would be to roll back arrangements that have been consummated.


      Policymakers should consider the relative size of ultimate demand for longevity protection (pension liabilities) and the supply of current providers of protection (life (re) insurance sector). Given this imbalance (demand could far exceed supply) there will come a point where longevity risk is passed from the (re)insurance sector to the broader capital market. Regulators and (re)insurers should support the development of the framework to support such a capital market. Furthermore, policymakers and regulators need to seriously consider the extent to which current pension valuations in North America understate pension liabilities. They should also consider the funded status of the implicit and explicit governmental support, if they are to craft sustainable policy to support the promises made to employees through their pension plans.

    - The CLHIA supports a review of rules and regulations pertaining to the measurement, management and disclosure of longevity risk with the objective of establishing high qualitative and quantitative standards.
    - We are in favour of the recommended policy of ensuring that institutions taking on longevity risk can endure increases in longevity, and we are also in favour of removing market distortions that result in an unlevel playing field.
      - We support a push for transparency in the LRT market in order to guard against unhealthy levels of systemic risk in the system.
        - We agree with Recommendation #7 that tail risk needs to be considered. The insurance and reinsurance markets are already regulated to reserve for tail risks and holds the risk as part of a range of diversifying and offsetting risks to cope with the "cure for cancer" scenario.
          - Finally, we strongly support the compilation and dissemination of more granular and up-to-date longevity and mortality data. The development of more refined longevity and mortality data will provide additional information to improve modeling of longevity risk which, in turn, will encourage further development of a longevity capital market. At present, Canadian mortality tables are very much out of date, masking the true cost of the pension promise that sponsors are making, and making the longevity risk less visible. New tables were recently proposed that are more current but it is important to encourage all market participants to continually investigate mortality trends and longevity improvements, looking to best practices around the globe. It should be noted that even with more refined data, basis risk will still exist unless a perfectly matched population can be found. Without an insurance indemnity solution, elimination of basis risk is unlikely.
        The CLHIA appreciates the opportunity to comment on the Joint Forum paper on longevity risk transfer, and, as the market expands and continues to evolve, we would be pleased to provide additional input to the Joint Forum and to Canadian insurance and pension regulators.

        Yours sincerely,

        "Original signed by"

        Frank Swedlove


        c.c. Mr. Stuart Wason, Senior Director
        Office of the Superintendent of Financial Institutions Canada

        Mr. Chris Fievoli, Resident Actuary
        Canadian Institute of Actuaries