The iM Seasonal Multi-Sector Investment Strategy

• The iM Seasonal Multi-Sector Investment Strategy capitalizes on a seasonality-based approach to investing, leveraging the well-known “Sell in May and Go Away” phenomenon.
• During the winter period (end of October to the end of April) the model invests in the five highest-ranked U.S. Sector ETFs equally weighted.
• Selection is based on the performance of 30 sector ETFs during the previous one- and two-year winter periods, and not by selecting arbitrarily cyclical- and defensive categories for the winter- and summer periods.
• In the summer period the model allocates funds equally between the iShares 20+ Year Treasury Bond ETF (TLT) and Invesco QQQ.
• By combining sector rotation and hedging mechanisms for inflationary periods, this strategy offers a systematic framework for maximizing returns and managing risks.

This model capitalizes on seasonal effects in equities, particularly the “Sell in May and Go Away” strategy. This approach posits that markets, especially cyclical sectors tend to perform better from November to April than from May to October.

The “Sell in May” phenomenon is illustrated with SPY’s seasonal performance (Appendix-1, Figures 1 & 2). The effect is strongly evident, as the performance disparity makes a compelling case for avoiding SPY during Summer months and focusing on Winter periods. This insight validates seasonal investment strategies, particularly those that leverage sector rotation or alternative assets during Summer to enhance returns and mitigate drawdowns.

Winter Investment Strategy

During the winter months, from the end of October to the end of April, the model invests in the five highest-ranked U.S. Sector ETFs equally weighted. These ETFs are selected from the offerings of SPDR, iShares, and Vanguard, which collectively provide 30 Sector ETFs (excluding Real Estate).

The ranking system prioritizes ETFs based on their performance during the winter months one and two years prior, with higher past performance being favored. The model starts in March 1999 to utilize some historical data for ranking, initially focusing on the first sector ETFs from SPDR whose inception was in December 1998.

Summer Investment Strategy

In the summer months, from the end of April to the end of October, the model allocates funds equally between the iShares 20+ Year Treasury Bond ETF (TLT) and Invesco QQQ, the ETF tracking the Nasdaq-100 index. However, if inflation expectations, as indicated by the University of Michigan Inflation Expectation Index, exceed 3.5%, the model shifts investment to the Invesco DB US Dollar Index Bullish Fund (UUP), which typically strengthens during periods of high U.S. interest rates.

Trading and Rebalancing Rules

Trades occur only at the end of April and October, though rebalancing may take place during the investment periods. Positions are usually held for six months. However, a position showing a loss exceeding 15% is sold and not replaced.

Backtesting Results

A backtest performed on Portfolio123 from March 1999 to December 2024 reveals an annualized return of 18.1%, significantly outperforming the benchmark S&P 500 ETF (SPY), which achieved 8.1% over the same period. The model also demonstrated a maximum drawdown of -33%, compared to -55% for SPY (Figure-3).

The backtest involved 179 completed trades, with 84% yielding positive returns, excluding dividends. A starting investment of $100,000 would have grown to $7,400,000, accounting also for total trading costs of $42,000 due to an assumed slippage rate of 0.1% for each trade.

(click to enlarge)

For annual performance see Appendix-2, Winter performance see Appendix-3, Summer performance see Appendix-4, and Sector Representation & List of Trades see Appendix-5.

Strengths and Considerations

Strengths

• Efficient Allocation: High win rate (84%) demonstrates strong sector rotation and defensive shifts.
• Dynamic Risk Management: The stop-loss mechanism effectively curtails losses.
• Macro Alignment: Inflation hedging with UUP addresses economic variability.

Considerations

• Market Conditions: The strategy depends on the persistence of seasonal trends and inflation-dollar dynamics.
• Selection Bias: Reliance on past performance may not always predict future outcomes.
• Tax Implications: Short-term gains could reduce net returns in taxable accounts.
• Limited Trading Frequency: Reduces transaction costs but may miss intermediate opportunities or market shifts. However, the strategy is simple to execute, particularly for smaller or more passive investors.

Following the model

Weekly updates are available to subscribers of the Bronze tier on our website, iMarketSignals. Subscribers gain exclusive access to ongoing insights, performance updates, and detailed analyses of this model and others.

Appendix-1

The “Sell in May” phenomenon is illustrated with SPY’s performance in Fig-1 and Fig-2.

(click to enlarge)

The data highlights the stark seasonal disparity in SPY’s performance between the Winter and Summer periods. Here are the key takeaways:

1. Winter Period Dominance (Fig-1):
   • Approximately 75% of SPY’s total return since 1999 was generated during the Winter periods (end of October to end of April).
   • Out of 26 Winter periods, 23 were positive, demonstrating a high probability of positive returns during these months.
   • Over the 26-year period from 1999 onward, the total return for Winter period investments in SPY was 478%, meaning an initial $100 investment grew to $578, or inflation adjusted to $305.
2. Summer Period Underperformance (Fig-2):
   • Only 15 of the 26 Summer periods (end of April to end of October) had positive returns, a significantly lower success rate.
   • Over the 26-year period from 1999 onward, the total return for Summer period investments in SPY was just 31%, meaning an initial $100 investment grew to $131. Inflation adjustment reveals a negative real return: the purchasing power of the $131 dropped to $69, underscoring the inefficiency of holding SPY during these months

Appendix-2

The annual performance of this strategy is demonstrated in Fig-4, listing year-by-year returns.

(click to enlarge)

Over the 26-year backtest period since 1999, the seasonal investment strategy produced positive returns in all but one calendar year. In 2018, it recorded a negative return of -5%, similar to SPY. During this time, SPY had six calendar years with negative returns. Additionally, the strategy outperformed SPY in 17 of those years.

Appendix-3

Fig-5 shows the 1999–2024 performance when investing in appropriate sector ETFs during the winter periods only and going to cash without interest during the summer periods.

(click to enlarge)

This analysis highlights the compelling advantages of a seasonal investment strategy focusing on high-performing sector ETFs during the winter periods while avoiding market exposure during higher-risk summer months. By achieving an annualized return of 10.1% and limiting the maximum drawdown to -32%, this approach demonstrates a superior risk-adjusted performance compared to a continuous SPY investment, which returned 8.2% annualized with a much deeper drawdown of -55%.

This stark difference underscores the benefits of strategically timing market exposure to capitalize on seasonal trends and reduce vulnerability during periods historically associated with higher market volatility and lower returns. It also suggests that a well-structured seasonal strategy can provide significant upside potential without the extreme downside risk often seen in traditional buy-and-hold strategies.

Appendix-4

Fig-6: Performance of the strategy for Summer Periods only

Fig-6 shows the 1999–2024 performance when investing in QQQ, TLT, and UUP during the summer periods only and going to cash without interest during the winter periods.

(click to enlarge)

The 1999–2024 performance when using the summer allocation strategy (QQQ & TLT) and going to cash during the winter demonstrates notable robustness. The annualized return of 7.3% with a maximum drawdown of -16% is a strong result, particularly when compared to SPY under the same approach. SPY delivered a much lower annualized return of 0.8% and suffered a far steeper drawdown of -50% (Fig-2).

This comparison highlights the effectiveness of the summer period allocation in capitalizing on the defensive and growth characteristics of TLT and QQQ, respectively, while also hedging against inflation effects with UUP. The stability and drawdown control offered by this strategy stand in sharp contrast to the vulnerability of SPY during the same periods.

Key insights include:

• Significant Outperformance in Returns: The summer allocation strategy achieves a nearly 9x improvement in annualized return over SPY under the same conditions.

• Improved Drawdown Management: The maximum drawdown of -16% for the summer allocation is markedly less severe than SPY’s -50%, reflecting better risk mitigation.

Appendix-5

Sector Representation & List of Trades

This data highlights a few interesting aspects of the seasonal strategy’s sector selection during Winter Periods:

1. Diverse Representation: The model selects a wide range of sectors, with 28 out of 30 eligible ETFs being chosen. This counters the assumption that only cyclical sectors like Industrials and Technology dominate in favorable Winter Period investment periods.
2. Top Performers: Materials (24 times), Energy (21 times), Technology (20 times), and Consumer Discretionary (20 times) were the most frequently selected, suggesting these sectors have historically delivered robust performance during the Winter Period.
3. Less Represented Sectors: Healthcare (15 times), Industrials (10 times), and Financials (8 times) also see significant representation, but sectors like Utilities (7 times), Telecommunications (6 times), and Consumer Staples (4 times) appear less often, which might be attributed to their generally lower growth potential or defensive nature.
4. The model’s current allocation suggests that the ranking system has identified a concentrated opportunity in the Industrials and Technology.  This focused selection implies a degree of confidence in these sectors’ near-term performance, based on the underlying metrics driving the ranking system. The absence of diversification across other sectors supports the assertion that the system is effective at identifying pockets of strength rather than randomly distributing investments in 5 of all the available 30 ETFs.
5. The sector concentration becomes evident from October 2001 onward after the inception date of the iShares Sector ETFs in May 2000, and is even more pronounced after the inception date of the Vanguard Sector ETFs in January 2004.

This distribution underlines that while cyclical sectors may generally perform better, the model is adaptive and diversified, capable of identifying opportunities across a broad range of sectors. The strong presence of Materials and Energy also suggests that commodity-related market drivers might significantly influence Winter Period performance.

Performance of UUP:

The performance of UUP as an inflation hedge is evident from its strong returns during some of the model’s summer periods. Here’s a breakdown of the performance shown:

Observations:

1. Correlation with Macroeconomic Factors: UUP tends to perform well in periods of rising interest rates, inflation concerns, or global financial uncertainty.
2. Consistency in Volatile Periods: These examples show that UUP can provide a hedge during times when equities may face headwinds due to inflation or market instability.

This performance justifies its inclusion, complementing the seasonal strategy’s goal of mitigating risk while seeking returns.

List of Trades since 1999