19 Jul 2023 4 min read

Bond volatility: how should DC savers react?

By John Southall , Detian Chen

DC savers and trustees need to understand the risks of bonds but may wish to tread carefully before disinvesting.


Last month, the Pensions Regulator (TPR) published a blog reminding DC trustees not to lose focus on guarding savers from economic volatility. They commented on the gilts crisis of last year and how bonds have continued to suffer in 2023 in a climate of rising interest rates and high inflation.

Understandably, TPR wants to ensure DC trustees don’t sleepwalk into a crisis for savers approaching retirement, and called on them to use their guidance to protect savers.

Bonds have come under fire recently on two fronts. First, following a more than 3% rise in yields over the past 18 months (bond values fell sharply), bonds are seen as riskier. Second, we may potentially have moved to a regime in which the stock-bond correlation is positive. This means bonds may no longer help hedge falls in equity values.

These are valid concerns, but should be understood carefully in the context of savers meeting their objectives. As TPR says, trustees should primarily focus on outcomes.

Bonds: less risky than they look?

Many people understand that bonds help to insulate savers from changes in annuity rates in the run-up to retirement. Relatively few annuities have been bought since freedom and choice, but a key point is that bonds can potentially be beneficial even if the saver doesn’t plan to buy an annuity.

To illustrate, we consider a 65-year-old investor about to start income drawdown from a diversified multi-asset growth strategy. They start with a £100,000 pot and plan to withdraw £9,000 per year in retirement. The graph below shows how our simulated chance of success (where ‘success’ means not running out of money before they die) varies with gilt yields if the initial pot remains at £100,000 i.e. is not sensitive to yields.


The higher yields are, the greater the chance of success. To stabilise the chance of success, the saver therefore actually needs the £100,000 pot to be ‘yield sensitive’ i.e. grow when yields tumble and fall when yields rise. This is what bonds do. It’s particularly important to get this effect when yields are volatile, as they have been recently.

The model makes the standard assumption that a higher risk-free rate leads to a higher expected return across all assets. We appreciate some people may be sceptical that bond yields have anything to do with the expected returns on other assets. But if that were the case, the recent rise in bond yields would mean they now find bonds more attractive compared with other assets.

Stock-bond correlation

It’s impossible to be sure, but we may have moved into a regime where the stock-bond correlation has flipped from negative to positive. The concern is that if bonds no longer hedge falls in equities, savers lose the incentive to hold them. There is some truth to this, but bonds aren’t only about hedging falls in equities.

To illustrate we’ve set up a simple ‘optimisation’ model across equity, gilts and cash that reflects that DC savers should be ‘outcome-driven’. For most savers it isn’t short-term risk and return that matters but the ability to meet future cashflow objectives, such as a lump sum at retirement or an income in retirement.

Assuming a target cashflow in eight years[1] we optimised for two different stock-bond correlations: -0.4 and +0.4. For comparison we also performed corresponding ‘asset-only’ optimisations that focus only on the short-term risk of the assets:


Under asset-only optimisation, switching to a positive stock-bond correlation disposes of all gilts (50% drops to zero) but, when objective-driven, a sizeable gilts allocation remains (68% drops to 37%).

The above also assumes that the equity risk premium (ERP) doesn’t change. But if investors dump bonds, owing to them being a worse equity hedge, then all else equal their yield should rise, compressing the ERP. This could offset or even cancel the 20% shift towards equity (32% to 52%). In general, it’s important not to consider possible changes in parameters in isolation. The market is also aware of the change and is likely to have already priced it into valuations.

A saver near retirement who plans to splurge their entire pot on a cruise is ill served by medium- or long-dated bonds. But for those seeking to generate a pension, an allocation to bonds is likely to remain important, in our view. Despite their recent volatility they can stabilise and improve outcomes, regardless of the stock-bond correlation.


[1] In practice, most DC savers tend to have objectives linked to inflation. Similar arguments and examples in this section and the previous can be made by replacing nominal objectives with inflation-linked ones, nominal yields with real yields, and nominal gilts with index-linked gilts.

John Southall

Head of Solutions Research

John works on financial modelling, investment strategy development and thought leadership. He also gets involved in bespoke strategy work. John used to work as a pensions consultant before joining LGIM in 2011. He has a PhD in dynamical systems and is a qualified actuary.

John Southall

Detian Chen

Quantitative Associate, Solutions Group

Detian currently works as a Quantitative Associate in LGIM's Solutions modelling group. Detian is involved in projects detailing the research and development of financial models on asset and liability management (ALM) and investment strategies for pension schemes.

Detian joined LGIM in October 2022 from Deutsche Bank, where he worked in the product validation group and was a lead in data infrastructure governance and the assessment of model risks in equity and commodity derivatives. He holds a master's degree in Quantitative Methods in Risk Management from the London School of Economics.

Detian Chen