The financial crisis of 2007-2009 has been characterised by many researchers and commentators as a systemic crisis. In the wake of the crisis, interest in how individual financial institutions contribute to the risks of the financial system as a whole has increased enormously. Both regulators/supervisors and academics have started to ask such questions as “What is systemic risk?” and “How can systemic risk be measured?”
Moreover, we have seen determined policy responses by regulators and supervisors. In the US, the Dodd-Frank act was enacted in 2010, creating the Financial Stability Oversight Council (FSOC), whose role it is to identify risks to financial stability and respond to emerging threats. In Europe, also in 2010, the European Systemic Risk Board was installed through a similar mandate. Finally, in 2011, the mandate of the Financial Stability Board (created in 2009 as a global platform for monitoring financial stability) was expanded and its role in monitoring and promoting financial stability was strengthened.
There is a broad consensus that banks and insurance companies may contribute to systemic risk in the financial system. However, for other financial market institutions this is less clear-cut. In 2015 the FSB and IOSCO launched a consultation on systemic risk (FSB and IOSC, 2015), looking for ways to identify “non-bank non-insurer globally systemically important financial institutions” that through distress or disorderly failure, because of their size, complexity, or systemic interconnectedness, could cause significant disruption to the wider economic and financial system.
In this column, we study the role of pension funds, which are an important category of end-users of financial instruments. What are the characteristics of pension funds, how do they operate on financial markets, are they a stabilising factor on financial markets, and what are the systemic risks, if any, potentially associated with them?
Systemic risk and financial stability
What is systemic risk, and how can it be measured? Although a large literature on systemic risk has grown in the last eight years, a single, agreed-upon definition of systemic risk does not exist. Intuition tells us that systemic risk is about the functioning of the financial system as a whole, and the possibility that a single (combination of) event(s) leads to a chain reaction that threatens the functioning of the financial system in its entirety.
Bisias et al. (2012) provide a broad survey on existing definitions of systemic risk, and of different ways that have been proposed to measure systemic risk. Definitions of systemic risk range from very wide to very specific. An example of the former is that offered by the ECB (2010): “risk of financial instability so widespread that it impairs the functioning of a financial system to the point where economic growth and welfare suffer materially”. Other definitions focus on one aspect of systemic risk. These include mechanisms such as financial imbalances, correlated exposures, spillovers to the real economy, information disruptions, feedback behaviour, asset bubbles, contagion, and negative externalities (Bisias et al. 2012). This list shows that the concept of systemic risk is particularly multifaceted and complex, and that measuring or assessing it is not easy. Bisias et al. (2012) go on to provide an extensive overview of the proposed measures for systemic risk, grouped according to theme and data complexity.
A particularly interesting application is provided by Giglio et al. (2015), who combine 19 individual measures of systemic risk into an aggregate ‘systemic risk index’. Individually, the measures do not perform particularly well in predicting a financial crisis, but combined into the index, their predictive performance becomes quite strong. This avenue of research seems to be promising for gaining more insight into the phenomenon of systemic risk and for providing reasonable predictions of where pressures on financial stability might show up.
Pension funds’ role in financial stability
Pension funds have a number of characteristics that render them different from purely financial institutions. First, pension funds do not operate on a leveraged basis. They are legally restricted from borrowing, except for short-term liquidity needs. This means that pension funds only invest own funds (i.e. the contributions paid into the pension fund by members and sponsor), so that there are no multiplier effects in profits or losses that pension funds generate. Second, the average duration of the liabilities of pension funds is between 15 and 20 years. This makes pension funds natural long-term investors. Third, pension funds can be technically insolvent, but they cannot go into actual bankruptcy. Instead, a situation of technical insolvency will be resolved by using the policy instruments that pension funds have at their disposal, such as raising contributions, receiving additional financial support from their sponsor, limiting the indexation of entitlements to inflation, and –as a measure of last resort – marking down the entitlements of their participants. By combining these policy instruments, pension funds can recover in an orderly way from their position of underfunding over a number of years, instead of ending up in a much more disruptive bankruptcy procedure. Fourth, pension funds are only allowed to use derivatives that reduce their risk exposure, typically interest rate swaps (or swaptions) to reduce interest rate risk, and currency swaps to reduce exchange rate risk associated with their investments denominated in foreign currency. The combination of these four characteristics of pension funds means that pension funds by construction are very stable financial market participants.
In 2015, the European Insurance and Occupation Pensions Authority (EIOPA) conducted a stress test for pension funds (or ‘IORPs’ in European legal terminology). Its purpose was to assess the resilience of pension funds to severe shocks, as well as to assess their systemic impact. EIOPA recently published the results of the stress test (EIOPA 2016). The test looked at the IORPs’ investment policy responses to the financial crisis as a proxy for determining whether IORPs add systemic risk or not. The test was not designed to explore the potential knock on effects of an individual large fund getting into financial distress. Unlike banks, direct linkages of pension funds to other financial institutions are limited.
Based on the stress test results, EIOPA concludes that individual pension funds in a number of countries seem vulnerable to financial market shocks, but that for the financial system as a whole the impact of financial distress of pension funds is likely very low. In fact, EIOPA finds some evidence that pension funds tend to have a stabilising influence on volatile financial markets. The reason is that many pension funds have an investment policy of rebalancing. They typically split up their total investment portfolio into a matching part (high grade fixed-income assets and interest rate derivatives) and a return part (equity, lower grade fixed-income assets and real estate). Both parts of the portfolio are assigned a strategic weight.1 If a shock pushes a portfolio component below its strategic weight, then the fund will start buying assets from the class that was hit by the shock to restore its strategic portfolio mix. The increased demand for the assets that were initially hit by the negative shock may help in breaking a vicious cycle of asset price declines. Indeed, EIOPA found some supporting evidence for this hypothesis during the financial crisis of 2008-2009 (EIOPA 2016).
In a recent study using monthly data for almost 40 large Dutch pension funds over the period 2009-2014, Broeders et al. (2016) find strong evidence that these funds rebalance their asset allocations in the short run. Specifically, using econometric regressions they find evidence that of a passive portfolio weight increase resulting purely from market returns, 20% is corrected within the same month through active rebalancing in the case of equity and about a quarter is corrected within the same month in the case of bonds.
The discussion of pension fund characteristics and the results from the stress test conducted by EIOPA suggest that there is no reason to treat pension funds as systematically important institutions. However, we make one qualifying remark. There is a trend of imposing similar risk-based solvency requirements on all financial players (Basel III, CRD IV, Solvency II, and potentially Solvency II-like rules for pension funds). This makes sense from a micro perspective, but may carry the danger that the behaviour of all players on the financial markets becomes more similar – especially in reaction to sizable shocks – which may amplify these shocks and have a destabilising effect on financial markets. This way microprudential supervision may create system risks through the backdoor.
Bisias, D, M Flood, A W Lo and S Valavanis (2012) “A survey of systemic risk analytics”, US Department of Treasury, Office of Financial Research, Working Paper No 1.
Broeders, D W G A, D H J Chen, P A Minderhoud and C J W Schudel (2016) “Pension funds’ herding”, Mimeo, Dutch Central Bank and University of Amsterdam.
ECB (2010) “Financial networks and financial stability”, Financial Stability Review, 155-160.
EIOPA (2016) IORPs Stress Test Report 2015, 26 January.
FSB and IOSCO (2015) Consultative Document (2nd) Assessment Methodologies for Identifying Non-Bank Non-Insurer Global Systemically Important Financial Institutions: Proposed High-Level Framework and Specific Methodologies, 4 March.
Giglio, S, B Kelly, S Pruitt and X Qiao (2015) “Systemic risk and the macro economy: An empirical evaluation”, NBER, Working Paper 20963.
1 Examples of typical investment strategies are a defensive strategy of 90% matching and 10% return, or a more aggressive strategy of 50% matching and 50% return.