Fixed income: is geopolitics now trumping rates as the dominant force in markets?

In this edition of our fixed-income quarterly, Alphorum, we focus on the structurally less cohesive and more uncertain macroeconomic environment – and what this complexity means for global bond investors. We also explain our conviction in Europe as a decarbonisation leader and share evidence of the effectiveness of machine-learning techniques in credit strategies.

Portfolio positioning: lower rates, steeper curves – but rising uncertainty

Recent events have accelerated the shift from globalisation, trade liberalisation and international cooperation to reshoring, protectionism and increased competition. At the same time, the geopolitical situation has precipitated both the European Union’s ReArm Europe initiative and the release of Germany’s ‘debt brake’ to allow heavy investment in defence spending and infrastructure.

Heightened uncertainty is likely to bring more frequent market volatility and weigh on economic activity, while tariffs threaten to further disrupt already-mutating trade flows and spur inflation. Real GDP figures and central-bank projections hint at a slowing macroeconomic backdrop, while the US administration’s mercurial approach to domestic and foreign policy discourages capex, threatening to slow growth further.

Although non-farm payroll figures and jobless claims suggest some weakening, it would be hard to argue the US economy is actually contracting. US short-term nowcaster indicators continue to show solid readings, while in Europe, data show recent improvements in manufacturing activity and a normalising labour market. Overall, corporate fundamentals show only mild deterioration.

Most central banks are reaching the end of their cutting cycles, although rates vary significantly. In an uncertain environment, the calibration of monetary policy presents a challenge for central banks, making it tricky for market participants to anticipate future rate moves.

Read also: America first to safety first – finding opportunity and shelter in fixed income

Debt sustainability has resurfaced as a concern, driven by factors including a deterioration in countries’ debt-to-GDP ratios and fiscal deficits, elevated central-bank holdings of government bonds and overall leverage. Concerns about the rising supply of government debt have caused a repricing of the term premium and put pressure on long-dated real yields. While we see no imminent danger, we prefer the shorter end of the curve, which will be less impacted by supply concerns.

Lower rates and steeper curves mean fixed income looks increasingly attractive versus cash. However, multiple sources of risk make it vital to be nimble and ready to adapt. With volatility likely to remain high, we prefer to seek visibility and liquidity through a move up in credit quality, while we continue to like US Treasury Inflation-Protected Securities (TIPS) for their attractively priced real yield and ability to defend against inflation.

Lower rates and steeper curves mean fixed income looks increasingly attractive versus cash.

Government bonds: a complex backdrop requires agile positioning

The fundamental outlook for global sovereign fixed income continues to depend on a complex mix of regional dynamics and macroeconomic forces. An unsettled mood prevails, with ongoing concerns over tariffs joined by a new focus on fiscal policy. Increased spending on defence in Europe and the cost of Trump’s ‘Big Beautiful Bill’ in the US are likely to put fiscal policy into expansionary territory, with the latter almost certain to add to the US budget deficit even after tariff revenues are taken into account.

Central banks around the world have responded to rising downside risks with further rate cuts. While short-term sovereign bonds benefited from this dovish stance, steepening pressures led once again to modest underperformance versus cash. Notably, neither the US dollar nor US Treasuries acted as safe havens during the sharp risk-off move in April, raising questions about the dollar as a reserve currency and supporting the case for higher term premia in Treasuries.

Increased government spending in Europe and the US is likely to keep supply elevated for years. The UK also faces challenges relating to fiscal headroom and high issuance, while Japan has struggled with weak demand for its 30-year bonds due to investor concerns about fiscal easing and political uncertainty. In light of these technical factors, we prefer to seek duration exposure in the belly of the curve or through curve-steepening positions.

Looking ahead, we expect the full impact of tariff increases to become visible in economic data during H2 2025, potentially prompting further easing from major central banks. Meanwhile, markets currently assign low probability to a significant shock, a situation that we believe creates asymmetric upside for duration. Valuations generally remain attractive across sovereign bond markets globally, with regional divergence presenting opportunities – given ongoing uncertainty we like markets with manageable debt levels and favourable supply dynamics, especially those that are early in their easing cycles.

Markets currently assign low probability to a significant shock, a situation that we believe creates asymmetric upside for duration.

Corporate credit: is the TACO trade still on the menu?

Credit markets have rallied since the volatility following the ‘Liberation Day’ announcements and new issuance is healthy – but the question remains: can this continue as tariff levels solidify?

So far, the impacts of the US administration’s radical policies on markets are following a pattern set by other seismic market events (and their aftershocks): a typically massive initial reaction is followed by adaptation, and then a faster recovery from each reoccurrence. This trend towards normalisation has been reflected in the so-called ‘TACO’ (‘Trump Always Chickens Out’) trade in recent months. Our own perception is that US political risk is likely to remain high for the foreseeable future – and the resultant potential for market volatility shouldn’t be ignored.

Read also: What Liberation Day’s reciprocal tariffs mean for investors

For companies, an unpredictable environment results in lower efficiency and higher costs – a particular problem for vulnerable sectors such as autos and steel that are in the frontline of the tariff war. Uncertainty is already being reflected in the lower confidence of many firms, although it is likely to be Q3 earnings calls that show the greatest impact from tariffs. The hope for these companies is that by then, announced tax cuts should boost consumer confidence and support demand.

For credit, the biggest potential issue would be if a softening US economy leads to consumer credit challenges, leaving banks less able to lend to businesses. However, regulations mean banks are generally far better capitalised and well run than in the past, minimising this risk.

Overall, despite tight spreads, credit still offers the opportunity to lock in decent yields, and investors can reasonably expect outperformance as interest rates reduce over time. In a challenging international trade environment, we continue to prefer domestic names, particularly those less focused on the US. Amid ongoing uncertainty, real estate continues to offer a relatively safe harbour.

Sustainability: Europe’s net-zero advantage

In a world where climate imperatives and geopolitical uncertainty compete for attention, Europe stands out as a stable, policy-driven and industrially aligned environment for net-zero investing.

Having peaked back in 1990, European Union (EU) emissions had fallen 37% by the end of 2023, while GDP rose by more than 68% in the same period. Yet unlike the US, the EU still has a strong industrial sector: more than 50% of the bloc’s GDP comes from manufacturing, construction, energy and transport – sectors undergoing the most profound transformation in the net-zero transition.

Broad public and cross-party consensus is supported by an arsenal of policy commitments and regulation. These include the European Green Deal, the Net-Zero Industry Act (NZIA), vehicle emissions legislation, a well-developed Emissions Trading System (ETS) and the Carbon Border Adjustment Mechanism.

FIG 1. European emissions are in structural decline¹

The bloc invests more than ten dollars in clean energy for every dollar in fossil fuels, derisking a climate-aligned focus for long-term investors. USD 110 billion went into renewable generation in 2023, while investment into energy storage and demand-side flexibility is also strong. Streamlined permitting is a further support for net-zero projects. This supportive environment is reflected in a high proportion of energy from clean sources and the continued phaseout of coal.

Meanwhile, the NZIA provides policy certainty and demand visibility for a wide range of clean technology, including photovoltaic and thermal solar, onshore and offshore wind, energy storage, carbon capture and storage (CCS), grid technologies and heat pumps, among others. It also seeks to secure access to critical raw materials and CO2 storage capacity.

In summary, for investors seeking long-term, policy-aligned exposure to the net-zero transition grounded in an industrially focused economy, Europe offers a compelling hunting ground. The continent’s regulatory clarity, public and political support, and industrial strategy should make it an ideal part of any serious net-zero investment strategy, in our view.

Read also: The economics of decarbonisation make the net-zero transition Trump-proof

Systematic research: applying machine learning to credit investing

The rise of artificial intelligence (AI) has sparked excitement across industries – and finance is no exception. But much of what is advertised as AI in investing is actually machine learning (ML). Still, this does not diminish its potential value as a tool to improve investment outcomes.

We have applied ML to credit investing for nearly half a decade, using it to enhance and refine existing models – an extension to them, rather than engineer radical departures. Applying ML in financial datasets poses unique challenges, since the data is prone to cause the two main sources of model error, namely bias and variance.

The future of systematic credit isn’t artificial, but rather intelligently constructed, rigorously validated, and subject to human judgement.

We use an ‘ensemble’ ML method, training a group of smaller models and combining their weak outputs into a single stronger one. Our preferred approach is ‘boosting’, which builds models sequentially, with each one learning from the errors of the last. This makes it particularly effective at refining a baseline model – in our case a fairly simple cross-sectional relative-value signal. The aim is to reduce the loss between the realised spread-adjusted excess return and that predicted by our base model.

To measure the effectiveness of our model, which we call ‘RVSBoost’, we benchmark it for US investment-grade and high-yield bonds against existing systematic signals including relative value (RVS), equity momentum (EMOM) and a combination of the two. These signals are well known from the literature and to credit investors, and are highly significant inputs into the model itself. The results are shown in Figure 2 below.

FIG 2: Simulated performance of systematic long-short quintile corporate-bond strategies²

In practice, we look at a composite of both ML-derived and more traditional signals for our investment decisions. Furthermore, we employ a ‘quantamental’ approach, using input from our fundamental credit analysts to challenge the signals where necessary. 

Our experience with RVSBoost underscores the value of combining intuitive, interpretable models with robust ML techniques and fundamental oversight. We believe that in the relatively illiquid and heterogenous world of credit, success comes from building transparent, stable models grounded in an understanding of the nuances of fixed income. The future of systematic credit isn’t artificial, but rather intelligently constructed, rigorously validated, and subject to human judgement.