Roland van den Brink: Is the management of interest rate derivatives like a leaky tap?

Roland van den Brink: Is the management of interest rate derivatives like a leaky tap?

Interest Rates Pension system Pensionfunds Derivatives
Roland van den Brink

This column was originally written in Dutch. This is an English translation.

By Roland van den Brink, Founder of TrigNum

Pension funds use interest rate derivatives to reduce the interest rate sensitivity between pension liabilities and investments. But is their management cost-efficient in practice? Below are some tips that can help you ask specific questions.

The investments of a pension fund can be divided into a return portfolio and a portfolio that serves to manage interest rate sensitivity. In technical terms, the latter is called 'LDI portfolio' or 'matching portfolio'. Part of the interest rate hedging takes place through investing in bonds and mortgages. However, interest rate derivatives (interest rate swaps) are used for the most part.

The LDI portfolio often has the greatest influence on returns. For example, interest rate derivatives at pension funds lost €146 billion in value by 2022 due to sharply increased interest rates. In light of the Future Pensions Act (Wtp), various funds have further reduced their interest rate sensitivity during 2023, often with the help of interest rate derivatives. That is why I consider the implementation of interest rate hedging. If it is suboptimal, you have a leaking tap.

No complaints?

Many large parties from other countries, especially the US and England, manage the LDI portfolios. From marketing we know slogans such as 'customer first'. Here we see something striking. In many European countries, pension funds have complained about extremely low interest rates, causing the valuation of pension liabilities to skyrocket.

You would expect that the asset managers would also complain. However, I have not found an article or interview showing that these parties saw low interest rates as a problem. Could it be that the financial sector has an interest in interest rates being low and/or fluctuating considerably? The management fee is often a percentage of the value of the portfolio, which makes a high valuation welcome. Continually adjusting the interest rate hedge leads to purchases and sales that also generate revenue.

Theory vs practice

Let me be clear: suboptimal management lies primarily with the funds and not with the asset manager. The latter group does what the customer tells them to do. Where things sometimes go less well is in the translation of policy into guidelines for the LDI manager. There are several examples. With some funds you see that there is hardly any deviation from the (exact) objective. The board has decided that the interest rate hedge is x% and this assignment is then closely monitored, within tight bandwidths, through frequent periodic transactions.

The fact that the board did not intend such a strict interpretation, or is not aware of it, has escaped attention. One of the causes is that, based on 'good governance', funds have introduced a strict separation between policymakers and executives, such as the many (small) transactions that are not necessary for the purpose, namely controlling the interest rate risk. This suboptimal management often increases when working with runtime segments – in technical terms called the bucket approach.

In other words, the cash flows of the liabilities are divided into maturity segments. Think of 0-1 years, 1-3 years, 3-7 years and so on. The idea is that the interest rate can be different in each of these segments. The interest rate sensitivity for each segment is then hedged with appropriate interest rate swaps. With a strict mandate, the division leads to more (unnecessary) transactions.

What can you pay attention to?

Are the numbers large? There are (large) pension funds with 1,200 swap positions and funds with more than 250 swap transactions per year. How relevant are these numbers for managing interest rate risk? Some funds use interest funds from providers and are not aware of these numbers. The reason is simple: these investment funds stick to their prospectus, not to the intended purpose of the funds.

Is there an alternative? Pension expert Keith Ambachtsheer's most relevant statement is: 'What is measured, is managed.' I also experience in practice 'not measured, not managed'. Various pension funds do not use any benchmark for efficient implementation of interest rate hedging. Industry funds in particular use the credo 'actual = benchmark' for their benchmark, in other words: we do not test the implementation against an alternative approach. That paves the way for a leaky tap.

A proven solution

My message is that with seven or eight interest rate swaps you can approximate the value development of each cash flow profile to be hedged very well. This means that a handful of interest rate swaps are sufficient to manage interest rate sensitivity. Such a (theoretical) benchmark portfolio can serve as an excellent benchmark for cost-efficient implementation.

I refer to the 2005 IPE article 'Devising an investable index' by Arun Muralidhar and Jan Willem van Stuijvenberg. The fit is so accurate that it is usually sufficient to recalibrate the position once a year. In other words: a few transactions per year is enough. If the interest rate changes sharply, such as in 2021, an interim adjustment may be considered. Another reason for recalibrating the benchmark may be an assigned indexation.

In short, it is wise to use the questions described and the application of a benchmark to determine whether the tap is leaking. Don't expect administrators to be eager to cooperate, which is another sign on the wall.

For the expert

So why does the handful of swaps approach work so well? The European Central Bank (ECB) has a website where it shows the yield curve of European government bonds every day. To determine the interest rate for each term, the ECB uses an interest rate model with six parameters. In principle, six different bonds or swaps are sufficient to determine the six parameters.

In practice, it is better to use some additional swaps of different maturities. The (theoretical) swap portfolio can be composed in such a way that the first derivatives of these parameters are the same as the forecast pension cash flows. This means that the value of the two portfolios will only differ if there are very large interest rate movements, something that in practice has only occurred once every 20 years.

This article contains a personal opinion from Roland van den Brink