Linking pensions to longevity

The OECD has issued an interesting working paper Life-Expectancy Risk and Pensions: Who Bears the Burden?, which looks into a number of OECD countries' relatively recent policy changes to share part of the longevity risk with pensioners. Given the proportions of the risk and the massive inter-generational skew in cost/benefit, this is perfectly reasonable and should be adopted universally.

For some undisclosed reason, Switzerland is virtually omitted from the scope of the analysis, even though there clearly is no linkage between longevity risk and pensions whatsoever. In the Swiss three pillar system, longevity risk is borne in the first pillar by the tax payer, by employers in the second, and by individuals in the third pillar.

Victims IV: The real crisis

The Economist deserves praise for having featured a special report on the impact of ageing populations in this time of crisis, which overrides longer term requirements with its fiscal profligacy. The majority of industrialised countries were already on an unsustainable fiscal path before the crisis struck. It is difficult to see how government finances will ever be able to return to a trajectory that is stable longer-term.

It goes without saying that the foreseeable instability of public finances has a dramatic impact on capital funded retirement systems. At this juncture, the jury is still out on the prefix of instability, i.e. whether we will see inflation or deflation. Either way, the contradictory demands on the investment strategy of individual funds are anything but trivial and may need to be implemented consistently in very short order once the dust settles. Scenario analysis and preparation is the name of the game. I look forward to a workshop producing a Shell/Oxford-method scenario analysis on Switzerland 2030, to which I have been invited by the federal crisis management education unit.

Impact of accounting and prudential regulation on pensions

"A long-term view involves short-term risk, whereas a short-sighted strategy involves increased risk over the long term."

EDHEC just released its impressive report Impact of Regulation on the ALM of European Pension Funds. Even though we disagree in some instances, we think this is mandatory reading for anyone in the pensions investment space because it highlights those areas of regulation which will be of increasing consequence for pension funds' investment strategies in the near future, as we have continued to stress over the recent past.

At the core of the report is the development of an asset allocation model in the presence of liability constraints. The solution involves the components cash, risky assets and the liability hedging portfolio. The state of the art model takes inflation and longevity risk management into account as well.

There is not enough space nor time for an in-depth review of this valuable piece. Nevertheless, I would like to mention two issues that have slightly moderated my enthusiasm for the report:

• There seem to be a few at least implicit factual inaccuracies in the parts describing the regulatory environment. The most glaring of which may be the assumption that the EU pensions directive is applicable in Switzerland - it is not.
• Accounting standards seem to be understood to effectively determine investment action. While it is not unheard of that managements structure transactions in such ways as to optimise their reporting, this clearly goes one step too far. We are well aware of the interdependence between perception (qua accounting standards) and (economic) reality, but at least in an academic report, the latter needs to retain some vestige of predominance over the former. Remember: pension funds' long-term time horizon, as accounting standards can and do change.

Longevity in Switzerland

It is hardly a coincidence that the Federal Office of Statistics publishes a new study about the Future of Longevity in Switzerland (German, French) today. There is an upcoming referendum to decide about the proposed reduction of the transformation rate with which accumulated pensions capital will be transformed into annuities. Longevity expectations are an important factor in that hotly contended issue. The study expects an additional 5 to 9 years of life expectancy gains over the next 20 years with a continuation of morbidity compression.�

The tyranny of the present

In a recent issue of its flagship publication Sigma, SwissRe described Scenario analysis in insurance, identifying scenario analysis as a key tool to analyse fat-tail risks and their impact on profitability and competitive position of insurers. One of the pioneering sources of scenario analysis is the approach developed by Shell. Whereas scenario analysis is referred to as a key tool for both strategic planning as well as enterprise risk management of insurance, Sigma reports with some degree of astonishment that banks do not use it to assess their total enterprise risk exposure.

Applying the concept of scenario analysis to pension funds should be self-evident, not least if you think of a pension fund as the insurance subsidiary of your firm. It faces a set of opportunities, threats and parameters quite similar to those of an insurance, yet scenario analysis is not common in the pensions industry. Nevertheless, a number of pensions-specific scenarios easily come to mind: a jump in longevity due to unexpected medical progress, prolonged negative real interest rates, a pandemic (as explained in Sigma), regulatory changes to the competitive landscape ...

The Economist Intelligence Unit has just come up with its own bleak exercise in scenario analysis (hat tip Global Guerrillas). Its central forecast of stabilisation is assigned a probability of just 60%, whereas the more disruptive instability scenarios are assigned 30% (de-globalisation) and 10% (collapse in USD) respectively. Scenario analysis has been posted as a means to escape the tyranny of the present, but being where we are today, we are not so sure this is a good thing.

Mortality-linked securities

The Pensions Institute has an excellent new paper on�Mortality linked Securities and Derivatives. The paper describes the problem (longevity risk) and what conclusions can be drawn from present experience in pensions buyouts and securitisation transactions. They also discuss the pricing of longevity risk in the absence of a liquid mortality-linked capital market. For a complete picture, we'd be interested in the fallout of the present turmoil in the asset-backet securities and credit derivatives space on mortality-linked securities ... �

Redefining Old

Nomura has an excellent piece of research out that goes well beyond what that genre usually entails in the brokerage space. The Business of Ageing is an extensive discussion of the key risk of the pensions industry that is longevity, its implications for the real economy, financial markets and the major industries. I particularly value the section about longevity with its discussion of the technophysio approach which, in combination with longevity convergence across countries, is posited as leading to rapid longevity growth. Where official UN projections arrive at an average life expectancy of ca 85 years in 2050, Nomura models predict ca 90 years.�

Redefining Old refers to another interesting aspect of the paper: Whereas a social security definition in terms of years lived will lead to an increasing share of the "old" cohort burdening social security, the authors argue that with increasing healthy life expectancy due to morbidity compression, it will be reasonable (i.e. necessary) to expect people to work (much) longer. The authors pinpoint that age at about 80, which would be suicidal for any politician to ask for. Note that this blog has argued for the same number before.

Virtually unseen in brokerage research is the extensive, up-to-date scientific apparatus provided. The label useful is fully deserved.

As a side comment: In spite of Switzerland's claim of having an exemplary retirement system that is the envy of the world, her only (and favourable) appearance in the paper is in a table about obesity ...

Death and taxes ...

This is an excellent presentation by Governor Jens Thomsen of the National Bank of Denmark, on how to hedge and invest in an environment where average life expectancy rises by over 5 hours every day. He proposes that governments should issue more ultra-long term bonds to create a hedging substrate for that time horizon.�

What Thomsen does not address, however, is the challenge to such instruments arising from an investment environment with massively higher government debt, as it is foreseeable in many countries. The temptation to apply the inflation tax to reduce such debt may be overwhelming, which is why such ultra-long bonds should be issued with an inflation protection.

Innovative ways of financing retirement

In this day and age where financial innovation is (wrongly!) blamed for the end of capitalism and the world as we know it lock, stock and barrel, the title of SwissRe's latest edition of Sigma is courageous. Nevertheless, it is a comprehensive assessment of the growing role that insurance will have to play in the provision of old age retirement funding, where insurance is to be understood in a functional rather than an institutional sense.�

An important section of the study is dedicated to managing retirement product risks. Longevity risk is identified as prominent among them, but its management is limited by the rather shallow depth and breadth of longevity risk markets. Notably absent from that section, however, are considerations on valuation techniques. If there is one thing that we can learn from the current crisis, then it is how crucially important it is to value & stress test innovative products properly over their entire life cycle.

Swiss management of longevity

Influential Swiss newspaper NZZ has a good article (in German) about how Swiss Pensionskassen manage their longevity risks, i.e. not very much. The article claims that due to the usage of backward looking mortality tables, the longevity of members is systematically underestimated in a world of continuously rising longevity. Unfortunately the article does not take into consideration the experience in more advanced countries such as the UK. This is a valid issue which receives too little attention because of a rigid regulatory environment in Switzerland where many important parameters are politically determined with little consideration to factual developments.

CEIOPS State of Pensions Report

CEIOPS'�Report on financial conditions and financial stability in the European Insurance and Occupational pension fund sectors has a good section (starting p. 22) about recent developments in the European pension funds market, giving insights into last year's changes in a number of countries and a statistical overview, based on Eurostat.

The nemesis of pensions?

Is this man the nemesis of funded retirement systems? At any rate, Aubrey de Grey's work on life extension by regarding ageing as a curable disease gains increasing attention and traction, as witnessed by a veritable burst of recent media coverage such as Wired's. If his claim that life expectancy of about 120 years may be achievable to currently living generations, this would indeed pose a major challenge to retirement systems which rely on predictable mortality tables that shift slowly. Hence this space is certainly worth while watching for black swan risks.

Longevity concerns

The Economist brings us up to scratch on�Abolishing Ageing, highlighting those areas of medicine which directly address ageing itself. For obvious reasons, this is an area of science that the retirement industry needs to watch closely, even though the current trend towards widespread obesity would not suggest that the most immediately promising approach of caloric restriction�could win a popularity contest. But it could, once the side effects of caloric restriction (i.e. near starvation) are removed. Even so, the UK industry prepares for even higher longevity - some mortality tables expecting the average 65-year-old to live to 100 and beyond by 2050.�

It is therefore apprpriate that the finance industry provides tradable longevity indices, such as Goldman Sachs' QxX family, or JPMorgan's Lifemetrics toolkit. Credit Suisse's Longevity Index does not appear to be directly accessible on the web.�

Global Graying Report

Standard & Poor's has an interesting update of its Global Graying Report, which projects the current fiscal policy stance in conjunction with forecast demographic expenditures to arrive at hypothetical future ratings. It is a pity that Switzerland is missing from the simulations - France for instance goes from AAA today to Speculative in 2040.

Actuaries hedging

Pension liabilities heavily depend on forecasts of the mortality of the pension plan beneficiaries. These forecasts are produced by actuaries. Until quite recently, the actuarial science was deemed to be capable of coming up with a reliable point estimate (if you can speak of point estimates in the context of large cohort mortality tables) of future mortalities. This expectation does not survive under closer scrutiny of course. That is not a bad thing, because point forecasts of the future are necessarily inaccurate, therefore such expectations were never realistic.

The British actuarial profession is leading the pack again with its publication of a draft library of mortality projections, indicating that it may be reasonable to utilise a number of scenarios in mortality projections. But more closely to home, and indicative of imminent changes to current mortality assumptions and thus liability valuations, they warn that currently employed mortality tables may considerably underestimate improvements in future mortality. We understand this to imply that more realistic tables will assume higher average life expectancies and, consequently, higher pension liabilities.

Ask the economist

The OECD has published an interesting protocol of an online debate with OECD economists about the pensions challenge. There's plenty of materials and food for thought there, especially in the background readings listed, or the annuity debate; one wonders whether the lack of private annuity products may not to some extent be due to an effective crowding out by mispriced public policy alternatives - but unfortunately that question is not touched upon.

Institutional Life Markets Association

IPE has an interesting story about the recent foundation of Institutional Life Markets Association (ILMA), a non-profit trade association to ?encourage best practices and growth of the mortality and longevity related marketplace?. Founding members of this potentially highly relevant association are Bear Stearns, Credit Suisse, Goldman Sachs, Mizuho International, UBS and West LB AG. The association does not appear to have a website, yet.

Annuities: a private solution to longevity risk

Its title may appear a bit facetious, but the latest issue of SwissRe's Sigma is anything but. It contains a comprehensive overview of the challenges to capital based retirement provision arising from increasing longevity. The prime focus of the publication is on insurers and insurance products, naturally, but most of its considerations and precepts are directly applicable to non-insurance pensions providers. A very worthwhile read for everyone in the retirement business!

Mortality assumptions

Today, the 7th Annual Conference of the IEBA has been brought to my attention because of its several presentations about pan-European pension funds (scroll down to Past Events).

While the presentations themselves did not strike me as out of the ordinary, I noticed the slides of an earlier event about mortality assumptions used in the EU. The slides present a good overview of this important paper. Most notable is the slide shown on the side. It displays the difference between observed and assumed future life expectancy of pension scheme members across a number of countries. The difference varies widely with Denmark, Switzerland & Germany showing the slightest differences whereas France, Spain and Ireland have the most "longevity reserves" built into their mortality assumptions: a plan using French mortality assumptions will show considerably higher liabilities than the same population of beneficiaries subjected to Danish assumptions.

Mortality bonds

Our earlier story seems to have been very topical. epn also has an interesting piece on the securitisation of longevity and recent developments therein. All this co?ncides (probably not randomly) with an extreme mortality securities for some mortality risks in France, Japan and the US being successfully issued by SwissRe in November of last year.

Securitising mortality

SwissRe's latest issue of Sigma is dedicated to Securitisation - new opportunities for insurers and investors. It presents a good overview of structures, instruments and recent developments in the market for insurance linked securities (ILS), which has been booming in recent years, and is expected to continue to do so because of its more favourable regulatory treatment in the context of Solvency II and a dearth in insurance capacity otherwise.

The hitherto virtually untapped market segment of extreme mortality might prove to be exceptionally interesting for pension funds because of its potentially long duration, its size and its natural hedging capacity for pension funds: a mortality bond's payout is negatively related to longevity, which is one of the principal risk factors affecting pension funds' liabilities. To be an effective hedge, the populations covered need to overlap significantly, though. SwissRe estimates a current market size of USD 5'500 bio, which it expects to grow to USD 7'000 bio by 2010.

A word of caution is in order, though (which is notably absent from the SwissRe paper). Insurance capacity is a highly cyclical asset with short cycles and high variance. Therefore it is likely that broader capital markets will be tapped when capacity is relatively scarce & expensive. Insurers will thus tend to offer relatively less favourable terms in order to take advantage of the broader market's limited sophistication & knowledge of insurance cycles.

How long do we want to live?

In its 50th anniversary issue, the New Scientist has forecasts for the next 50 years from over 70 of the world's most brilliant scientists. Three of those forecasts deal with ageing. The common theme is as extension of life expectancy "by about 40%". It is possible that a half-century from now, the most urgent question facing our society will not be "How long can humans live?" but "How long do we want to live?" This would amount to a considerable acceleration in the long running 2.5 years per decade growth rate of life expentancy, as discussed in an earlier post.

Taboo topic retirement age [CH]

Credit Suisse weighs on the debate about the retirement age in Switzerland by challenging the uncharacteristically disconnected political consensus that a retirement age of 65 is sustainable in the present demographic situation. In its recent study, it argues that only an increase of the working life can return social security and retirement provision systems to sustainability and addresses widely held contrary beliefs, especially concerning higher flexibility in retirement as long as it does not lead to an increase of retirement age on average.

ECB discusses ageing

In its Occasional Paper 51 - Macroeconomic implications of demographic developments in the Euro area, the European Central Bank looks at the impact of ageing on long term growth rates, labour market policies, financial markets and public finance. The thrust of measures proposed is not surprising:

to increase labour participation by closing a gender gap, raising average hours worked and raising the effective retirement age,

to prepare for a more important role of financial intermediation in retirement provision, taking into account an expected further decline in the real equilibrium interest rate,

to take measures against public expenditures rising above 3% of GDP in most countries in the wake of increased pensions and healthcare costs, especially by increasing the importance of (partially) funded retirement systems,

to monitor the impact of ageing on monetary policy via the aggregate savings rate and real interest rates.

Disquietingly, the ECB does not fully discount the theory that ageing will lead to a decline in asset prices due to increased unsaving of retirees - it only provides several technical caveats. Food for thought indeed!

Demographic leporello [CH]

Swiss think-tank Avenir Suisse has published a comprehensive leporello (a.k.a. leaflet) about the demographic challenges in Switzerland and Europe. It is also available in French.

Rising retirement age north & south [DK] [GR]

The political debate about the sustainability of static retirement age in the face of rising life expectancy in European countries is beginning to bear fruit. While Alpha Bank of Greece expects (according to this article in Handelsblatt) that the retirement age for recent job starters will have to rise to 75 from the current average of about 60, the local politicians apparently do not grasp the seriousness of the situation quite yet.

Meanwhile, up north in Denmark, a broad consensus among the most important political parties has resulted in an agreement to reform the country's social security system. According to a recent article in NZZ which is not available online, the average productive period of a Danish person went from 40.5 years in 1979 down to 38.5 years in 2005, while at the same time, life expectancy rose by 2.25 years, resulting in the prolongation of the average pension duration from 19.25 years to 21.5 years.

The pragmatic Danes have resolved to raise the retirement age to 67 (from 65) by about 2025. Thereafter, the retirement age will be adapted dynamically to the average life expectancy, which is probably the most sustainable approach.

Superlongevity

As we know, static old age retirement systems are under severe stress due to increased longevity (ca. 2.5 years per decade). But that's not all: the World Future Society has an article about superlongevity. The proponents of superlongevity expect that the trend of increasing longevity will not remain stable, but will actually accelerate due to technological progress. Michael Zey reckons that life expectancy will rise to ca. 125 years by 2075 from the current ca. 80 years. This averages to rate of increase of about 6.4 years per decade. With that kind of perspective, the absolute necessity for more flexible and, most of all, delayed retirement should be painfully clear.

Ageing & financial stability

This week's Third Conference of the Monetary Stability Foundation in Frankfurt has been dedicated to the challenges to the financial system arising from ageing and low growth. Prof. Weber, President of Deutsche Bundesbank, has given a rather thought provoking opening address, sketching the key transmission factors by which ageing impinges on financial stability.

Ageing hits politics [NL]

Demographic ageing is no longer approaching, but effectively starting. The Dutch Ministry of Finance receives recommendations of a fiscal policy study group charged with assessing the impact of ageing on Dutch government finance. The recommendations are tough: The next Cabinet should reduce expenditures by some EUR 15 bio (3% of GDP) in order to remain sustainable. Nonetheless, addressing the issue head on is exemplary.

Reproductive technologies to address demographic challenge?

Rand Corporation has looked into utilising Assisted Reproductive Technologies (ART) as a component of population policy to mitigate falling fertility rates and thus unfavourable demographic developments across Europe. To do that, the researchers transposed parameters from Denmark, where ART are more widely available than in the UK. They found ART to be three to four times more cost effective than child benefits for instance (via BBC).

Demographics to hurt banks / insurance? [D]

The German branch of consultancy Booz Allen Hamilton projects a draw of up to 25% on revenues of banks and insurances by 2030, caused a shift in the German age structure which will reduce savings rates and related demand. The German financial services industry is considered to be ill prepared for changing demand (via FTD).

Transparent longevity assumptions

Interesting! Following a January call of an informal group of London's leading investment bank and fund management analysts (the "Corporate Reporting Users Forum"), a number of companies have started publishing and thus exposing to be challenged their longevity assumptions used for the valuation of pensions schemes, which are obviously critically important (via FT). Virtually at the same time, an assessment of pensions liabilities of 26 Swiss SMI-component firms has been published (via Vorsorgeforum). The average discount rate applied, while being in line with Swiss legal requirements, is an unsustainably high 4.25% (down from 4.61%). I suspect that longevity assumptions are not published.

This is precisely the way ahead with full fair value valuations as well, especially in those cases where price information is not directly attainable. There needs to be a critical dialogue between users and preparers of financial statements concerning assumptions used. This dialogue obviously checks preparers' position of power, which is probably why they are often reluctant to participate in that dialogue.

Live long & prosperously

It is a truism that increasing longevity (as documented strikingly here) will put static retirement provision systems under severe stress.

It is all the more interesting to compare different institutions' approaches to the issue, as represented by UBS' recent research focus Demographics: a coming of age and Ageing and Pension System Reform: Implications for Financial Markets and Economic Policies, a November 2005 supplement to the OECD Financial Market Trends publication.

UBS evaluates ageing from a macroeconomic model point of view, apparently employing steady state equilibrium models. With these, it is attempted to establish the impact of ageing on several countries' economies as a whole (production, productivity, income, consumption), on financial markets (asset allocation, emerging markets, real estate) and on different industries (consumer, technology, health care and financials). From the magnitude of the task and the nature of the tools used, it is hardly surprising that the results of their work barely scratch the surface of the obvious. Given the study's ready understandability and pedagogic outreach to the general public, this would not be so bad, if only public policy implications were less fuzzy. There are no useful items with regards to pan-European pensions.

The OECD paper has a very different approach, given its G10 Executive target audience. Where UBS works with theoretical models, the OECD looks at empirical data and deploys microeconomic concepts to come to relevant policy conclusions. There are many interesting considerations of the current supply situation of fixed interest instruments given the ongoing shift in funds' asset management due to liability matching objectives and up to date cross country comparisons. Comparing the paper's general thrust to the Pensions Directive, it appears that the Directive is in line with state of the art best practice, especially as far as the Prudent Person rule is concerned, which it introduces to EU legislation and many member countries which have hitherto used quantitative restrictions (as are still in effect in Switzerland). This paper is an interesting read for the expert.