ETF Analysis and Gold’s Rally

Major ETF Performance and Correlations

• QQQ has been the top performer in terms of cumulative returns, while TLT (treasury bond market ETF) has been the worst, especially in a rising interest rate cycle.
• Gold has shown high volatility and has been catching up in performance.
• Gold’s daily returns are negatively correlated with the stock market (S&P 500), suggesting it could appreciate when the market declines, acting as protection against sell-offs.
• The correlation coefficient of TLT with the market is near zero, indicating no significant correlation, but this can vary with changes in Fed funds rates.
• High-yield bonds (like HYG) are positively correlated with the S&P 500, making them sensitive to stock market sell-offs.

Gold’s Hedging Properties

• Gold provides a high alpha (around 4% return) with almost neutral beta exposure to market volatility, outperforming NASDAQ/tech ETFs in this aspect.
• Gold can act as a defensive asset in low liquidity environments, providing protection against market distress and potentially positive returns.
• Gold provides positive returns even during periods of shrinking market liquidity and market distress, such as in Q4 2022 or Q1 2025.
• Gold has outperformed the S&P 500 ETF 50% of the time over the past 1,253 trading days, making it a theoretically good asset for portfolio allocation.
• Gold is potentially uncorrelated with equities, with most ETFs showing minimal correlation to gold, except for silver.

Here is a Venn diagram illustrating gold’s multiple hedging roles:

Portfolio Optimisation Metrics

• The Sharpe Ratio measures risk-adjusted return, optimising for higher annualised returns relative to contained annualised volatility.
• The Sortino Ratio is a proxy for the Sharpe Ratio but focuses on downside risk rather than overall standard deviation.
• The Calmar Ratio and Return-to-Volatility Ratio are other measures that assess return relative to peak-to-trough drawdowns or standard deviation, respectively.
• The tangency portfolio is considered optimal for risk-adjusted returns and has the best Sharpe Ratio in a CAPM model.

Risk and Volatility Measures

• Annualised volatility (standard deviation) measures the fluctuation of returns.
• Skewness indicates the asymmetry of return distribution; most analysed assets show negative skewness, meaning returns are slightly skewed to the negative side.
• Kurtosis measures the “tailedness” of the return distribution; most assets, especially bonds, exhibit kurtosis greater than three, indicating “fat tails” (higher probability of extreme returns).
• Value at Risk (VAR 95/99) measures the maximum potential loss over a specific period at a given confidence level (e.g., 5% probability of a 3.05% loss or more in one day for silver).
  • VAR parameters tend to underestimate true downside risk, as actual losses can exceed VAR levels, especially during prolonged market distress.
• Expected Shortfall is a more conservative risk measure than VAR, estimating the average loss beyond the VAR threshold.
• Maximum Drawdown refers to the largest peak-to-trough decline in an investment.
  • TLT experienced a -40% drawdown over approximately three years due to rising interest rates.
  • Silver is identified as the most volatile asset with the largest conditional volatility parameter (1.75) and historical VAR 95.
  • Gold (GLD) and utilities (XLU) have the lowest conditional volatility parameters.

Portfolio Construction Strategies

• Equal Weight Portfolio: Each asset receives the same allocation. It is easy to implement but often suboptimal and may not be truly diversified if underlying assets are similar.
• Risk Parity Portfolio: Allocates assets to ensure each contributes equally to the portfolio’s overall risk. This strategy tends to give larger allocations to lower-volatility assets like gold and bonds.
• Tangency Portfolio: Maximises the Sharpe Ratio, representing the optimal portfolio on the Markowitz efficient frontier.
• Minimum Variance Portfolio: Aims to minimise overall portfolio volatility for a given level of return.
• Maximum Diversification Portfolio: Focuses on uncorrelated assets to reduce overall portfolio risk.
• Modern Strategies: Include Black-Litterman, Core-Satellite, and Factor-Based portfolios, which often incorporate investor views, strategic core holdings with tactical satellite investments, or sensitivities to macro factors.
• Constant Proportion Portfolio Insurance (CPPI): A dynamic allocation strategy that shifts between risky and safe assets based on market conditions.

Here is a mind map of portfolio construction strategies:

Investor Behaviour and Macro Factors

• Investor behaviour in recent years has shown a strong preference for tech stocks, large-cap growth, and small-cap growth companies, driven by trends like AI.
• Many ETFs, particularly in energy (XLE) and financials (XLF), exhibit a very high beta (sensitivity) to inflation break-even rates.
• The 10-year Treasury yield is a significant macro factor, with many ETFs showing beta sensitivity to it. TLT, for example, has a negative beta coefficient to the 10-year Treasury yield.

Risk-Adjusted vs. Beta-Adjusted Portfolio Strategies

• Risk-adjusted portfolios (using measures like Sharpe Ratio, Sortino Ratio) often find that Vanguard technology ETFs (VGT, XLK) and communication services (XLC) perform better than QQQ on a risk-adjusted basis. Real estate and energy ETFs tend to carry higher risk with lower returns.
• For higher expected returns, risk-adjusted portfolios often require dynamic management and periodic rebalancing.
• Beta-adjusted portfolios incorporate the asset’s beta (market sensitivity) into the allocation.
  • Assets with high beta include financials, energy, the Russell 2000, emerging markets, and technology.
  • Assets with low beta (close to zero) include bonds (TLT) and gold, which help isolate from market risk.
  • A maximum Sharpe Ratio beta-adjusted portfolio tends to maximise expected return (e.g., 16.3% expected return) and typically allocates to tech stocks, growth stocks, and potentially some gold or S&P 500.
  • Beta-adjusted portfolios generally offer better expected returns and Sharpe Ratios compared to risk-adjusted portfolios, which often yield lower expected returns for traditional strategies like tangency or risk parity.