What are you defending against

For decades, the answer to one of the most important questions in portfolio construction appeared straightforward. Equities provided growth, and bonds provided defence. The classic 60/40 construct became the mainstay of investors’ portfolios. But what do we mean by defence? Bonds have been expected to perform several roles at once: to provide income, reduce overall portfolio volatility, preserve capital and crucially, to provide an offset when equities fall.

For the four decades that followed the inflationary peak of the early 1980s, government bonds performed these functions incredibly well, to the point that investors conflated the function with the asset. It was a period of unusually favourable conditions for this type of balanced portfolio. Inflation and interest rates fell, which meant that not only did bonds provide a decent source of income, they also bestowed capital gains and excellent diversification benefits. When recessions threatened company profits, bonds often thrived, thereby cushioning equity losses.

As impressive as this marriage of bonds and equities proved to be, it was not the all-weather solution that many thought it was. As the global economy emerged from the COVID lockdowns, investors were provided a painful reminder of that in 2022. Having trended gradually lower for decades, inflation suddenly became the main economic risk; interest rates rose sharply and bonds and equities fell together. Defence and offence both suffered the same fate; portfolios experienced significant drawdowns, particularly in real terms.

But that bad experience does not mean bonds are no longer useful. Yields are now considerably higher than they were before 2022, and long-duration government bonds could perform exceptionally well in the event of a severe recession or renewed deflationary shock. The main point is that bonds provide protection against particular risks. They do not protect against risk in general.

As we know, inflation quickly subsided shortly after COVID, bond markets settled down – although notably, they only retraced a fraction of the preceding losses – and investors have gone back to building balanced portfolios in the same old way.

This may have something to do with the fact that investors retain a touching faith in central banks and their ability to keep inflation under control. Take, for instance, the US ten-year breakeven inflation rate – a market-based assessment of expected inflation derived from conventional and inflation-linked bonds – which remains remarkably subdued despite: (a) major conflicts in Ukraine and the Gulf; (b) inflation having remained above the Federal Reserve's 2% target for several years; and (c) government debt levels which significantly increase the risk of more inflationary policy responses in the years ahead.

At Westminster, we recognise the limitations of bonds as a defensive asset, and we think that inflation will be more of a problem going forward. As a result, we have adopted a slightly different approach to the defensive part of portfolios.

Rather than maintaining a large allocation to duration, we have chosen to minimise interest-rate sensitivity, favouring short-duration corporate bonds – an area in which we specialise – alongside alternative strategies including managed futures, commodities and hedge funds.

History provides some interesting perspective. We modelled the hypothetical performance of two simple, notional portfolios using monthly backtested data since 2002. The conventional portfolio comprised 60% global equities, 20% US Treasuries and 20% US corporate bonds. The alternative portfolio contained the same 60% global equity allocation but replaced the defensive 40% with 20% short-duration corporate bonds and 20% managed futures. Both portfolios were rebalanced monthly to their target weights.

Across the full period, the alternative portfolio returned 6% annualised against 5.6% for the conventional portfolio, with slightly lower volatility (10% versus 10.2%) and a smaller maximum drawdown (24.8% versus 28.2%). More interesting is why the portfolios behaved differently.

For that, we looked at inflation. We divided the sample simply according to whether our inflation indicator was in rising or falling mode. The simulated results for the two portfolios were revealing.

When inflation was falling, the conventional portfolio performed considerably better, but when inflation was rising, the relationship reversed. That is broadly what was expected. Falling inflation tends to be favourable for duration. It creates greater scope for interest rates and bond yields to decline, generating capital gains for bondholders. In those circumstances, reducing duration has an opportunity cost.

Rising inflation creates a different problem. Interest rates and bond yields are more likely to rise, undermining precisely the assets intended to provide portfolio stability. Short-duration bonds reduce that sensitivity. Managed futures, meanwhile, have the ability to take positions across bonds, equities, currencies and commodities and can potentially benefit from the sustained market trends that major changes in inflation and monetary policy often create.

It would be tempting to conclude from these results that investors should simply replace bonds with alternatives. That would be the wrong lesson. Our own analysis demonstrates why. During periods of falling inflation, the conventional portfolio performed better. Reducing duration meant giving up some of the returns available from falling yields.

Managed futures have their own vulnerability. They can experience long periods of disappointing performance, particularly when markets are rangebound. Commodities are volatile and generate no conventional income. Hedge funds introduce questions around fees, liquidity and manager selection. Short-duration corporate bonds retain credit risk and can perform poorly during a severe recession.

There is no perfect defensive asset because there is no single economic environment against which investors need to defend themselves. The question is therefore not whether bonds or alternatives are better. It is whether the portfolio contains sufficiently different sources of defence.

There is also a danger in taking that argument too far. If investors attempt to protect themselves against every conceivable economic outcome, they can construct a portfolio that is wonderfully diversified, remarkably stable and unlikely to generate particularly exciting returns.

Duration for deflation. Commodities for inflation. Managed futures for sustained market trends. Hedge funds for uncorrelated returns. Cash for liquidity. Equities for growth. Keep adding protection and eventually you diversify away your return.

Every defensive allocation has an opportunity cost. Capital allocated to an asset designed to perform during one economic environment cannot simultaneously be invested in whatever performs best in another.

Portfolio construction is therefore not an exercise in eliminating risk. It is an exercise in deciding which risks are worth taking and which risks need to be diversified away. The objective is not maximum diversification. It is having enough diversification to survive being wrong, without diversifying away the consequences of being right.

This suggests a subtly different way of approaching portfolio construction. For an investor primarily concerned about recession and deflation, duration may deserve an important place in the portfolio. For an investor concerned about persistent inflation, fiscal deterioration or monetary instability, relying upon duration as the principal source of portfolio defence may be considerably less attractive.

Neither approach is inherently correct. They are protecting against different risks. Our analysis illustrates the trade-off. The conventional portfolio performed better when inflation was falling. The alternative portfolio performed better when inflation was rising.

Perhaps that is the central weakness in the way we use the word "defensive" in investment management. It implies that some assets are simply safe and others risky, but markets do not work like that. Every asset protects against some outcomes while remaining vulnerable to others. So, the real task of portfolio construction is not to find an asset that is defensive. It is to decide what you are defending against.

* Table: Notional Annualised returns since 2002. Portfolios rebalanced monthly. Conventional: 60% global equities, 20% US Treasuries, 20% US corporate bonds. Alternative: 60% global equities, 20% short-duration corporate bonds, 20% managed futures. (Source: Bloomberg, Westminster Asset Management). The portfolio performance results presented are hypothetical, simulated, and backtested. They are derived from the retroactive application of a model designed with the benefit of hindsight. Notional/hypothetical performance does not represent actual trading, real cash accounts, or the impact of real-world market liquidity, transaction costs, and management fees. Actual investor results may vary significantly from the simulated data presented. Past performance—whether actual or hypothetical—is no guarantee of future results.