Recent evidence shows that demographic characteristics such as age and income are related to specific investment strategies of pension funds (Gerber & Weber, 2007). At the same time, real estate and mortgage markets experienced structural changes which raised a number of challenges for national supervisors from a macroprudential perspective due to strong links with macroeconomic imbalances and the financial sector. This paper analyzes the investment behavior of Dutch pension funds with special attention to direct residential mortgage schemes. The author establishes a relationship between lender and mortgage specific characteristics and their mortgage portfolios. More specifically, Loan-To- Value (LTV) and Loan-To-Income (LTI) ratios are used to compare the risk exposure of Dutch pension funds in the landscape of the entire financial sector. The results are subsequently interpreted in the context of challenges to the resilience and solvency of Dutch pension funds as a consequence of demographic changes, financial regulations and low interest rates.
The first part of this research paper combines findings on the Dutch mortgage market, the role of pension funds, demographic and regulatory challenges, as well as lender characteristics. For this purpose, the
author employed descriptive data analyses and reviewed previous literature, revealing a high degree of heterogeneity across the Dutch mortgage market. Furthermore, the analysis of risk indicators identifies
a heterogeneous distribution of triggers such as NHG coverage, age, share of interest only mortgages, mortgage value and performance status. The obtained findings highlight the importance of macroprudenial policy instruments, such as LTV and LTI caps, amortization requirements and tax, in the mitigation of systemic risk and in increasing resilience against economic shocks.
The second part of this research paper is dedicated to the research question of whether borrower, lender or loan specific characteristics influence the investment behavior of pension funds related to direct
residential mortgage investments. Using multivariate OLS regression models, the author concludes that these characteristics have a statistically significant impact on the mortgage debt volumes issued by pension funds. Furthermore, this research paper analyzes whether loan and borrower specific characteristics have an impact on the funding ratio of pension funds, and confirms a significant link between mortgage default rates and the funding ratio of pension funds.
Finally, the data shows an increase in mortgages issued after the crisis, begging the question whether this phenomenon is caused by the Regulatory Framework (RF) that imposes recovery plans on underfunded institutions. Using balance-sheet data of pension funds that largely immunized themselves against drops in the funding ratio as a reference group, causality between the RF and mortgage investments can be proven. The regression model reveals that pension funds whose funding ratio dropped below the legally required minimum following the 2007/08 financial crisis held substantial investments in residential mortgages already before the crisis. After the crisis however the funds with recovery plans started investing more into mortgages relative to funds that went through the crisis without the need to recover. So, the aforementioned relationship describes a causal effect of the crisis and the RF on the solvency of pension funds with relatively large and risky mortgage portfolios.