A macro demographic analysis of pensions investing
Amlan Roy argues for a broader understanding of the factors that drive pension fund risk and asset allocation
The objective of this article is to appeal for an integrated, holistic and transparent approach to pensions. It represents a synthesis of research in the areas of macroeconomics, demographics, portfolio theory and pensions, and is based on the premise that macroeconomics and demographics interact, influencing fundamentals that affect pensions at the individual, institutional as well as national levels.
It is also an attempt to highlight Robert Lucas’s ‘micro foundations of macro’ and George Akerlof and Robert Shiller’s ‘behavioural aspects of macroeconomics’ as essential for factoring the macro-environment where pensions are. Economics, demographics, politics and geography all influence pensions at an individual and aggregate level.
The starting point is Peter Drucker’s statement that “we do not pay attention to demographics and when we do, we miss the point”. Drucker is a management leader, who has focused on the unsustainability of US corporate pensions.
In my view, demographics is about people and their characteristics, mainly at a consumer and worker level. Consumers and workers affect income statements and balance sheets at both individual and aggregate levels. The conventional understanding of demographics as just age or people numbers is misplaced and the cause of many misinterpretations.
An error that is pervasive and made by all has been to classify the population into three age groups; 0-14, 15-64 and 65-plus. The last age group of homogenous retirees was relevant in the 1960s and 1970s but, given today’s longevity advances, is outdated. The 80-plus age group grew 394% globally between 1970 and 2015, in contrast to overall population, which grew 100% over the same period.
This fast growing age group of 80-plus years is creating a different fiscal burden for countries based on health, pensions and long-term care expenditure, accounting for 20% of GDP, which is unaffordable and unsustainable. Another misconception is that countries with large, youthful populations have good demographics. This is not automatic and I challenge the fact that a preponderance of young workers does not guarantee higher growth or better income distribution, especially if they are in poor health, uneducated and unskilled.
A growth-accounting framework that I use decomposes GDP growth into working age population growth, labour productivity growth and labour utilisation growth. In the G6, growth from 2005-14 has been much lower than that the preceding decade, mainly due to a drop in labour productivity. It is easy for these advanced countries to grow again in terms of labour productivity by employing more young workers and more women. This would alleviate youth unemployment and gender inequity pressures. On the fiscal side, in the European Union nearly 80% of the total social benefits allocated are due to old age, sickness/healthcare and disability. This needs tackling through preventive care, education, training and efficient budgeting. No country has the resources to pay these costs based on current fiscal policy. An important factor is that politics reflects the power of older groups, who vote more actively and influence national policies that are often skewed in their favour.
The current pattern of increased longevity is different from past rises in longevity and creates a systematic effect where pooling of risks does not help1 . As per S Haberman “the survival curve was moving north-east but now it is moving east”. Reductions in mortality are now occurring at much older than younger ages, leading to increases in lifespan inequality. Experts have underestimated longevity and there is a need to combine forecasting techniques to improve estimates because the consequences of getting longevity wrong feed through to pensions and insurance products as well as the balance sheets of insurers and pension funds.
A point often ignored is that the demographics of advanced countries – in terms of fertility, longevity or population increases – are not homogenous. (Significant differences exist across all countries – see figure.)
A recent demographic study by longevity experts Gillespie, Trotter and Tuljapurkar (2016) finds that divergence in age patterns of mortality change drives international divergence in lifespan inequality. Using international data on mortality and longevity changes, they derive a link between age-specific mortality and lifespan inequality.
In addition to living longer, we are also living through uncertain political and economic times, with low interest rates and low inflation dominating the economic landscape. Central bank policy has been unconventional, resorting to unprecedented policy measures with arguably limited effect as they are operating close to the zero lower bound, reducing policy effectiveness.
Understanding, macroeconomics and demographics is integral for pension funds in managing their assets and liabilities strategically. There needs to be a focus on the macro-fundamentals that affect both liabilities and assets. This is essential, else the industry runs the risk of focusing too much on either liabilities (risks) and too little on assets (returns). A holistic understanding and application of finance requires an understanding that is summarised by Eugene Fama2 when he said “risk and return are joined at the hip”. To some of us, liability-driven investment and asset-liability management are one and the same. Good strategic asset-liability management requires assessment of the dynamic evolution of both assets and liabilities. This entails articulating the objectives of pension funds and their competing constraints and prioritising them before applying average variance optimisation techniques or taking liabilities as given, rather than looking at the uncertainty captured by changing distributions. It is important to understand these complex issues, but they require simplification to yield useful output and decision metrics.
Both companies and the public sector need to incorporate their pension balance sheets to get a holistic view of their true aggregate financial positions3. This is essential in view of the growing role of these long-term investors.
Finally, on the application of financial and investment approaches that are prevalent and used by banks and short-term investors – these do not directly apply to the objectives of long-term institutional investors. The return and risk measures and analyses used by pension funds must differ from those used by banks, insurance companies or endowments. The pensions industry, investment profession and regulators need to advance multi-asset-class investing in a rigorous, consistent and transparent fashion by considering how to incorporate new asset classes with small data histories.
We also need to acknowledge, as Franco Modigliani did in Rethinking Pension Reform, that neither defined denefit nor defined contribution pensions outperform or dominate under all conditions. He argued for hybrid pension plans with the best of both worlds.
In conclusion, we need to develop a ‘back to basics’ multi-asset investing model alongside dynamic liability management with a technical and investment focus. Applying dated single-period investment models developed across two or three asset classes will lead to problems. Let us be braver in asking questions, and understand constraints and limitations alongside objectives for a Brave New World in the spirit of Aldous Huxley.
1 S Haberman (2016) keynote presentations on longevity at CS Pensions conference.
2 Nobel Laureate and one of the Founders of applying CAPM to the markets and developing factors that explain equity returns. He expressed this in a Chicago interview alongside his colleague Richard Thaler, a leader in the field of behavioural finance.
3 L.J. Kotlikoff, The Coming Generational Storm, Franzoni & Marin (2006) and Jin, Merton and Bodie (2006) are studies that highlight improvements to pensions accounting and finance at corporate and national levels.
Dr Amlan Roy is a global macro-demgraphic researcher who is research associate at the LSE Systemic Risk Centre and guest finance professor at London Business School. This article is based on a keynote speech delivered at the IPE Conference in December 2016