4 Nov 2025
20 minutes
A quick overview of the key themes in this research, showing how the reinforcing forces that sustain dollar cycles also shape the relative performance of US and non-US equities.
The relative performance of US and non-US equities aligns with long-term dollar cycles.
The sustained outperformance of US equities over 14 years1 has materially reshaped global allocation decisions. It has attracted significant foreign capital into US markets, contributing to dollar strength, while also raising home bias among US investors, amplified by substantial allocations to US-focused private market investments.
This period of US equity leadership has unfolded alongside a remarkably persistent dollar cycle. Now 23 years since its last major peak, the current cycle is far longer than typical business, credit, or equity cycles. Its duration led us to research the drivers of such prolonged currency cycles and how they eventually reverse. As we outlined in our research paper, ‘The unstoppable dollar meets the immovable Mr Trump’, these cycles endure due to four reinforcing forces:
There have only been six turning points in dollar cycles since 1970. These have occurred when each of these structural forces flips in a relatively short period.
This paper extends our research by examining US versus non-US equity performance. According to the Macrohistory database2, from 1870 to 2020, both US and Global ex-US equity markets delivered average nominal returns of 9% per year. However, US equities exhibited higher volatility — 18% annually versus 14% for Global exUS — implying that comparable long-term returns have historically been achieved with lower risk outside the US.
When viewed through this historical lens, it becomes clear that leadership between US and non-US equity markets tends to shift over multi-year periods. These cycles of relative performance frequently align with broader dollar cycles: periods of dollar strength have often coincided with US equity outperformance, while weaker dollar phases have supported international markets. While the relationship is not perfectly mirrored, the directional alignment is notable and reinforces the importance of understanding underlying macro dynamics.
Figure 1: The performance of US versus non-US equities aligns with dollar cycles
Source: Bloomberg, MSCI and Ninety One, July 2025.
While it might be tempting to attribute currency valuations and relative equity returns primarily to capital flows, doing so would underestimate the complexity of equity markets and the four reinforcing forces that underpin dollar cycles.
We have also seen periods of divergence. In the 1980s, Japan’s booming economy muted the relative upward trend in US equities, though the US dollar surged with much higher interest rates than Japan and Europe. More recently, we observed shorter episodes of divergence in 2017 and 2020, when softer US growth expectations pushed the dollar down. Despite this, US equity markets continued to outperform non-US equities as investors focused on the resilience and profitability of US companies, particularly large technology firms, while weak economic conditions elsewhere continued to draw capital into the US. However, history shows these cycles eventually realign.
Therefore, the key question we seek to answer is not whether capital flows drive or reflect dollar cycles; they are one part of the equation. Rather, if we enter a multi-year downcycle in the dollar, as we believe we are, will we see an inflection in equity market leadership and sustained non-US equity outperformance?
If our analysis is correct, and all four reinforcing forces turn, investors will soon need to navigate a market environment distinctly different from the past decade and a half.
In this new cycle:
1 Morningstar, Bloomberg, June 30, 2025.
2 macrohistory.net
Dollar cycles last longer than business or equity cycles because four forces create considerable inertia. These same forces shape the relative performance of US and non-US equities.
Evidence for a new cycle lies in the decisions taken by policymakers and investors worldwide who act on the four forces and appear to be in the process of reversing them.
We analyse each in turn.
Discomfort around the unique role of the dollar and US assets in the global financial system is not new, but recent US policy action has made it more pressing. US policy measures, notably using secondary sanctions and freezing Russia’s central bank assets, accelerated efforts toward de-dollarisation. The Trump administration has added to this impetus, contemplating additional taxes on US capital account transactions and creating unease around institutional stability, policymaking predictability, and longer-term fiscal credibility.
While these factors are less directly tied to equity market performance, even modest shifts in regional allocations driven by such dynamics could have meaningful impacts over time. Importantly, this does not imply large-scale selling of existing US assets by global investors. Instead, the key issue is how incremental savings and reserve allocations will evolve.
Figure 2: Official sector survey shows 73% of central banks expect to hold fewer dollars over the next five years
Source: World Gold Council, Central Bank Gold Reserves Survey 2025, Perspectives on gold reserves. Available at: gold.org
Throughout the last cycle, the US ran an expansionary fiscal policy. It delivered relatively strong growth, supported by interest rates that remained higher than in other regions, even at neutral levels, where they neither expand nor contract the economy. European economies, by contrast, were held back by austerity, delivered very limited growth, and interest rates became stuck at or below zero. With the notable exception of China, emerging economies were much more constrained in their ability to deploy fiscal and monetary stimulus.
The landscape is now shifting:
Since the global financial crisis, capital flows have surged into the US economy and have reached levels with no parallel in history. By the end of 2024, non-US investors held US assets worth US$62 trillion, more than double the size of the US economy. After subtracting non-US assets held by US investors, the US Net International Investment Position (NIIP) was US$26 trillion, approximately 90% of US GDP. The primary driver has been portfolio investment, with global investors attracted by higher yields and stronger growth prospects in US debt and equities. Given the scale of these holdings, even a modest slowdown or reversal of incremental capital flows into the US could have major ripple effects across global markets.
Figure 4: The scale of international ownership of US assets has no parallel in history
Source: Bloomberg, July 2025.
The success of US technology companies during the internet era played a significant part in driving these inflows. Internet platforms benefited from powerful network effects and economies of scale, enabling new digital business models to emerge8. Companies specialising in online advertising, software-as-a-service, and infrastructure-as-a-service grew rapidly, delivering exceptional profits. The result has been extreme market concentration, currently at its highest level since 2001, at the peak of the dot-com bubble, and before that, during the aftermath of the Great Depression.
Historically, periods of extreme market concentration are associated with weaker subsequent returns.
To measure concentration in the US equity market, we compare the average market cap of the largest 10% of US-listed companies to the market cap of the median US company. As of 30 June 2025, the median company had a market capitalisation of US$4.4bn – the size of the Avis Budget Group or the retailer Abercrombie & Fitch – and the largest decile had an average market capitalisation of US$240bn – the size of Goldman Sachs or McDonald’s. The median company was therefore valued at less than 2% of the average large company, in line with the lowest that this metric has been over the last 100 years.
Figure 5: Periods of intense concentration have typically led to weaker 10-year returns
Source: Ninety One, Bloomberg, July 2025.
A broadening of market leadership in the next innovation cycle appears increasingly likely, driven by rapid AI advances and widespread adoption. In contrast to the internet era, the benefits of AI are expected to be distributed more broadly. Internet innovation in recent decades led to the growth of new platform business models and saw online revenue growth concentrating among a few winners. AI appears to be a general-purpose technology with the capability of automating a wide range of tasks currently undertaken by humans, leading to higher productivity growth and possibly profound changes to labor markets9. If this is right, then AI should deliver returns across many industries through cost savings and greater operational efficiency with smaller companies able to benefit from these advances in a similar fashion to the largest businesses.
Major US technology companies currently lead in AI model development and have significant resources to maintain their advantage. However, examples like DeepSeek or Z.ai show that other companies can quickly compete with much smaller budgets. Moreover, the AI-as-a-service model differs from the internet business models of the last cycle, with higher upfront costs and different scalability dynamics.
Successfully navigating this environment will require careful selection. The winners are unlikely to be simply the earliest adopters or those with the deepest pockets. Instead, investors should adopt a selective, bottom-up approach, focusing on companies that are intentional and strategic in their use of AI, able to convert incremental technological advances into meaningful, lasting performance.
The Nixon shock (1971), Plaza Accord (1985), and launch of the Euro (2002) were important triggers for the prior downcycles in the US dollar and coinciding periods of non-US equity outperformance. Recent speculation around a potential ‘Mar-a-Lago accord’ echoes these historical examples, although formal intervention may not be required as the threat of policy action is already influencing behavior among policymakers and investors. One clear example is the Taiwan dollar’s 10% appreciation in the second quarter of 2025, sparked by anticipation of trade negotiations with the US.
Currency market interventions have a greater chance of success when they correct significant imbalances rather than exacerbate them. The US dollar has declined from its peak but remains around 15% overvalued relative to the long-run average real effective exchange rate, or 30% above the last time it reached a low point in 2011.
Figure 6: The US dollar sits 15% above its long-term average
Source: Bloomberg, July 2025.
Furthermore, US equities are trading at 26x trend earnings versus 13x for ACWI ex-US equities. While the US equity market deserves a premium due to its quality and sector composition, this divergence has limits.
The high relative valuation of US equities appears closely linked to disproportionately strong capital inflows into US markets. Although there is some circularity, as rising equity valuations feed into the NIIP calculation, equities constitute less than 20% of the total NIIP, suggesting this is not the primary driver. Instead, it appears the same forces driving broader US capital inflows also support elevated US equity valuations.
Figure 7: The premium investors pay for US equities overlaps with international flows
Source: Bloomberg, July 2025.
A slowdown in the growth of US net liabilities could therefore put downward pressure on the valuation premium US equities currently enjoy.
3 The World Bank estimates the value added by the Chinese manufacturing sector was $4.7trn in 2024. The equivalent figures for the US and European Union were below $3trn each.
4 CATL’s Ningde (Fuding) facility spans about 1.5 million m² (~16 million ft², ~278 football fields) [EnergyTrend]. It produces one lithium-ion cell per second [InsideEVs], operates at around 95% automation [PR Newswire], and according to industry reports, uses around 1,000 robots on its lines [World Economic Forum, Global Lighthouse Network]
5 https://nistep.repo.nii.ac.jp/record/2000116/files/NISTEP-RM341-SummaryE.pdf, IP Facts and Figures.
6 https://www.nature.com/nature-index/.
7 Over the 20 years to end 2023 (the latest data available), the World Bank estimates that Chinese household consumption grew by $11.8trn or 11% per annum, whilst US household consumption increased by $11.1trn or 4.5% per annum in PPP terms. In real USD terms, the aggregate growth in Chinese household consumption in this period was slightly lower than the US ($5.9trn vs $6.5trn).
8 See Autor et al - The Fall of the Labor Share and the Rise of Superstar Firms and Emery - Market dominance in the digital age.
9 Significant uncertainty remains on the size of the productivity impacts of generative AI. For a range of different views see work by Goldman Sachs, McKinsey and Daron Acemoglu.
The AI era and changing macro landscape mean the next cycle of equity leadership is likely to be broader, requiring a rethink of how portfolios are regionally allocated.
If today’s market extremes and the emergence of the AI era mark the beginning of a new investment cycle, investors may need to rethink how they approach regional exposure. Relying on market-cap-weighted benchmarks is, in essence, anchoring to past returns, returns that have repeatedly proven unreliable guides to future outcomes. Despite its unmatched scale and liquidity, the nearly 70% dominance of the US equity market within global benchmarks creates a distorted lens, undermining genuine geographic diversification.
If historical returns are unreliable for determining regional allocations, what other options exist? GDP weighting is not appropriate, as GDP includes economic activities unrelated to listed companies, and multinational firms do not confine their operations within national boundaries.
A more intuitive approach is to focus on fundamental measures of listed companies. Based on metrics such as revenues, earnings, cash flows, and dividends, the US currently represents between 40% and 50% of the global total. Developed ex-US markets contribute around 35% to 40%, and emerging markets approximately 15% to 20%. A further advantage of fundamental measures is stability. Over the last 30 years, the fundamental share of the US market ranged from 36% to 48%, whereas market-cap weightings have fluctuated widely from 33% to 67%.
Figure 8: US share of company fundamentals has operated in a tighter band
Source: Bloomberg, July 2025.
Figure 9: Regional weights based on a range of different metrics
| US | DM ex US | EM | |
|---|---|---|---|
| Market cap | 67% | 23% | 10% |
| GDP (USD) | 26% | 32% | 41% |
| GDP (PPP) | 15% | 25% | 60% |
| Revenue | 48% | 37% | 15% |
| Earnings | 52% | 33% | 14% |
| Cash flow | 46% | 34% | 20% |
| Dividends | 46% | 40% | 14% |
| Book value | 41% | 41% | 18% |
| Range | 41% - 52% | 33% - 41% | 14% - 20% |
Source: Ninety One as of 31 December 2024.
Our Capital Market Assumptions, which forecast income, growth and valuation over the next decade, suggest a clear divergence in return potential:
| 3.7% US equities | 6.3% Developed ex‑US equities | 7.2% Emerging market equities |
Returns are expressed in US dollar unhedged terms; absolute return expectations vary for investors with other base currencies but relative returns do not. Higher income yields and the potential for currency appreciation relative to the US dollar underpin the stronger expected returns outside the US.
While forecasts inevitably carry uncertainty, they provide a structured basis for setting strategic allocations. Using these projections, we constructed a constrained global equity efficient frontier based on fundamental ranges. Given the lower expected risk-adjusted returns in the US market, the analysis points towards a more balanced global allocation with greater scope for opportunities across developed and emerging markets.
The global equity portfolio offering the highest risk-adjusted return comprises:
| 50% US equities | 30% Developed ex‑US equities | 20% Emerging market equities |
This allocation delivers a projected annual return of 5.2%, above the 4.7% forecast for the MSCI AC World Index.
This analysis excludes domestic equity allocations, which form a meaningful share of portfolios for most investors. The question now is how to balance that domestic component with global exposure in a way that reflects both fundamentals and the outlook for returns. For example, alongside a fixed 30% domestic allocation these global equity weights rebalance to 35% US, 21% developed ex-US and 14% emerging markets.
In practice, regional equity allocations will evolve over time as benchmarks and market capitalisations rebalance, reinforcing the value of maintaining flexibility within long-term strategic ranges.
Rebuilding balance across regions may prove central to capturing the next phase of global equity leadership.
The great rebalancing shows that equity leadership moves in cycles, aligned with the dollar, and today the evidence suggests a new cycle is beginning.
Regional performance cycles have remarkable staying power, but once clear evidence emerges that a cycle is shifting, investors can reposition portfolios for sustained new trends. Today, the convergence of geopolitical realignment, divergent policy paths, shifting capital flows, and coordinated efforts among global policymakers signals the onset of a new multi-year investment cycle.
Market-cap-weighted benchmarks anchor portfolios in the past, reflecting yesterday’s winners rather than tomorrow’s opportunities. The current dominance of US mega-cap stocks, powered by internet-driven platforms, appears increasingly vulnerable as valuation extremes collide with a weakening dollar and the transformative impact of AI. Investors should look beyond the familiar to identify new growth engines emerging across regions, sectors, and market segments. Successfully navigating this next cycle will require forward-looking diversification and careful, bottom-up selection to capitalise on a broader global opportunity set shaped by fundamental trends rather than index inertia.
For years, capital has gravitated to the US. A one-way trade powered by tech dominance and economic heft. But as the world tilts on its axis, the next cycle is unlikely to resemble the last, and the investment map is beginning to redraw itself.
A shrinking supply of shares has flattered US equity returns for years. The AI issuance boom is putting that into reverse.
Ninety One’s Capital Market Assumptions framework focuses on the key drivers of long-term performance. We do this to better understand possible future returns, enriching discussions with our clients.
Easy globalisation is over: multipolarity, bottlenecks and public dissatisfaction are reshaping the world. For investors, that means old assumptions are less reliable and resilience matters more.
War in the Middle East has brought one of the market’s long-standing geopolitical fault lines into sharp focus, particularly the risk of disruption in the Strait of Hormuz. Against an already fragile backdrop, the key question is whether this episode remains contained or escalates into a shock with global economic consequences.
Sahil Mahtani, Director of Ninety One’s Investment Institute, and Paul Gooden, Portfolio Manager for Global Natural Resources, discuss Venezuela and broader energy themes following the Goldman Sachs Global Energy Conference in Miami.
Sahil Mahtani, Director of Ninety One’s Investment Institute, and Nicolas Jaquier, EM Fixed Income Portfolio Manager, are joined by Phil Gunson, Senior Analyst at the International Crisis Group, who is based in Caracas for an on-the-ground perspective on the rapidly evolving situation in Venezuela.
Maduro’s exit raises the chance of change, but power is likely to remain with the security state as the US opts for pressure and negotiation over regime overhaul. Markets appear ahead of reality, with any improvement in oil and debt outcomes likely to be slow and uneven.
A major headwind to emerging market equity performance is fading. Read how this structural shift could lift future returns and reshape the case for EM allocations.
Ninety One’s Capital Market Assumptions framework focuses on the key drivers of long-term performance. We do this to better understand possible future returns, enriching discussions with our clients.
Our research offers three reasons why the dollar’s upside is limited, while the balance of risks increasingly point lower.
Dollar cycles are longer than others because four self-reinforcing forces create inertia. They rarely reverse unless all four turn at once. Trump-era policies, fiscal strain and shifting global capital could trigger such a convergence.
Shifting from reducing financed emissions to financing reduced emissions. Using practical examples, this paper sets out how we can evolve the approach to net-zero investing to achieve the dual objectives of delivering decarbonisation in the real economy while optimising returns for clients and beneficiaries.
Ninety One’s Capital Market Assumptions framework focuses on the key drivers of long-term performance. We do this to better understand possible future returns, enriching discussions with our clients.
Trump’s sweeping trade reset marks the largest US tariff escalation in nearly a century. Ninety One’s Investment Institute unpacks the policy shift, outlines scenario-based outcomes, and explores what it means for markets.
Monetary, fiscal, migration. How the US recession was averted, 2022-2023.
US equity markets reacted favourably to the US election result. However, Philip Saunders, Director of Ninety One’s Investment Institute reminds us that regardless of the party in power, long-term market movements are driven by fundamentals – growth matters.
Ninety One’s Capital Market Assumptions framework focuses on the key drivers of long-term performance. We do this to better understand possible future returns, enriching discussions with our clients.
Ninety One’s Capital Market Assumptions framework focuses on the key drivers of long-term performance. We do this to better understand possible future returns, enriching discussions with our clients.
Over the next decade a combination of the energy transition, nearshoring, geopolitics, demographics, technology, and public investment spending could trigger a capex supercycle that would see capex growing 2-3 percentage points faster than global GDP.
Market and portfolio insights, webinars & events curated from across our investment teams to help you steer through changing investment landscapes.
Important Information
This communication is provided for general information only should not be construed as advice.
All the information in is believed to be reliable but may be inaccurate or incomplete. The views are those of the contributor at the time of publication and do not necessary reflect those of Ninety One.
Any opinions stated are honestly held but are not guaranteed and should not be relied upon.
All rights reserved. Issued by Ninety One.