Most investors believe their portfolios are diversified – but Toby Watson argues that the correlation trap catches even experienced allocators off guard, particularly when market conditions shift suddenly and unexpectedly.
Diversification is the closest thing finance has to a free lunch – in theory. In practice, the benefits of holding multiple asset classes tend to diminish precisely when they are most needed. When markets come under stress, correlations between assets that appeared distinct under normal conditions converge, stripping portfolios of the protection they were designed to provide. Toby Watson, whose career spans decades of navigating complex market environments, has developed a framework for identifying and avoiding this trap that goes well beyond conventional asset allocation thinking.
The correlation trap is not a new phenomenon. It was observed during the 1998 Long-Term Capital Management crisis, during the 2008 financial crisis and again during the sharp market dislocations of March 2020. In each case, assets that had demonstrated low or negative correlations under normal conditions moved sharply in the same direction simultaneously, leaving investors with far less protection than their portfolio construction models had suggested. Toby Watson has engaged with this challenge throughout his career and brings a clear and practical perspective to building portfolios that are genuinely resilient to this dynamic rather than merely appearing diversified on paper.
The Mechanics of the Correlation Trap
Correlation is a statistical measure of how two assets move in relation to each other over a given period. Most diversification frameworks rely on combining assets with low or negative correlations to reduce overall portfolio volatility – the logic being that when one asset falls, another will hold steady or rise, smoothing returns over time.
The problem is that correlations are not stable. During benign periods, different asset classes often behave somewhat independently, reinforcing the impression that the portfolio is well diversified. But during periods of acute stress, the behavioural dynamics of markets tend to overwhelm the fundamental differences between asset classes. Investors selling to meet margin calls, reduce risk or simply respond to fear do not discriminate carefully between asset types – they sell what they can, and the resulting price movements push correlations towards one across the board.
Toby Watson has pointed to this dynamic as one of the most important and underappreciated structural features of modern markets. Understanding it has direct implications for how portfolios should be constructed and stress-tested.
Why Do Diversification Models Fail to Capture This Risk?
The core problem is that most diversification models are built on historical correlation data, which by definition reflects past market behaviour rather than future stress. Toby Watson, who spent nearly two decades at Goldman Sachs working across multi-asset portfolios, principal funding and structured credit, has noted that models calibrated to normal market conditions systematically underestimate the degree of correlation convergence that occurs during crises. The solution is not to abandon quantitative frameworks, but to supplement them with scenario analysis that explicitly models stress conditions – including scenarios where correlations move to levels that historical data would consider improbable.
How Toby Watson Approaches the Correlation Problem
Toby Watson’s approach to the correlation trap begins with a fundamental shift in perspective: rather than asking how different asset classes have historically correlated, the more useful question is why they might be expected to behave differently under stress – and whether those reasons are likely to hold in the specific scenarios most relevant to the current environment.
This means moving from a statistical framework to an economic one. Instead of relying on correlation coefficients, the analysis focuses on the underlying drivers of each asset’s return – the cash flow dynamics, the investor base, the liquidity characteristics and the macro sensitivities. Two assets may have low historical correlation simply because they have not yet been subject to the same stress simultaneously. That is a very different proposition from two assets whose return drivers are genuinely independent in ways that will persist under pressure.
At Rampart Capital, where Toby Watson serves as partner, this distinction shapes how diversification is assessed and how portfolio resilience is stress-tested. The emphasis is on understanding the economic logic of diversification, rather than relying on statistical patterns that may not survive contact with a serious market dislocation.
Factor Exposure as a More Reliable Diversification Framework
One of the more robust approaches to avoiding the correlation trap is to analyse portfolios in terms of their underlying factor exposures rather than their asset class labels. Toby Watson’s years at Goldman Sachs reinforced the value of this approach across multiple market cycles. Two instruments from entirely different asset classes may share the same sensitivity to credit spreads, to liquidity conditions or to risk appetite – meaning that they will behave similarly under stress despite appearing different in conventional classification frameworks. Identifying and managing these shared factor exposures is a more reliable basis for genuine diversification than asset class labels alone. Toby Watson has applied this thinking consistently across his career, from his time in institutional markets through to his current work in private wealth management.
Building Portfolios That Resist Correlation Convergence
Constructing a portfolio that genuinely resists correlation convergence requires deliberate attention to several dimensions:
- Liquidity differentiation – holding some assets that are genuinely liquid under stress conditions so that the portfolio can be managed actively when correlations converge, rather than being forced into passive acceptance of losses. Toby Watson has consistently argued that liquidity is a structural component of resilience, not simply a convenience.
- Return driver independence – ensuring that the portfolio contains genuine sources of return that are economically independent, not just statistically uncorrelated. This typically means including assets whose cash flows are driven by contracted revenues, regulatory frameworks or specific credit dynamics that are insulated from the sentiment-driven selling that drives correlation convergence in public markets.
Toby Watson has noted that this kind of portfolio construction is more demanding than conventional diversification, both analytically and in terms of the discipline required to maintain it through periods when some positions underperform.
Toby Watson on Resilience as the True Goal
The ultimate objective of addressing the correlation trap is not to eliminate all correlation – that is neither possible nor necessary. It is to ensure that the portfolio retains meaningful protective properties under stress, rather than discovering at the worst possible moment that the diversification it appeared to offer was largely illusory.
Toby Watson’s perspective, shaped by his experience at Goldman Sachs and refined through his current work at Rampart Capital, is that genuine portfolio resilience requires constant analytical vigilance. Correlation structures shift, market dynamics evolve, and the stress scenarios most relevant to a portfolio change over time. Staying ahead of those changes – rather than relying on historical patterns that may no longer hold – is the defining discipline of serious long-term portfolio management.
