Multi-Factor Investing: Combining Factors in Your Portfolio

Why Combine Factors?

Multi-factor investing combines multiple return drivers in a single portfolio. If individual factors offer return premiums, why not just pick the best one and concentrate there?

The answer is the same principle that drives all diversification: uncertainty. We don't know which factor will perform best next year, or even next decade. Combining factors spreads that uncertainty across multiple return drivers, applying diversification logic to sources of return, not just sources of risk.

The Case for Factor Diversification

Factors Have Different Cycles

Each factor goes through periods of outperformance and underperformance:

• Value struggled from 2010-2020, then partially recovered
• Momentum crashed spectacularly in 2009
• Quality held up well during COVID but lagged the recovery
• Small-cap has been inconsistent for two decades

If you knew which factor would win next, you'd concentrate there. But you don't—and neither does anyone else.

Factor Correlations Are Low

This is the key insight: factors don't move together.

| Factor Pair | Typical Correlation | |-------------|---------------------| | Value & Momentum | -0.2 to -0.4 (negative!) | | Quality & Value | -0.1 to 0.2 | | Low Vol & Momentum | -0.1 to 0.3 | | Size & Value | 0.2 to 0.4 |

The low (and sometimes negative) correlations mean combining factors can significantly reduce portfolio volatility while maintaining expected returns.

Smoother Returns

Single-factor portfolios can underperform for extended periods. A multi-factor approach smooths the journey:

Single-factor value investor (2010-2020):

• Decade of underperformance vs. the market
• Constant questioning of the strategy
• Behavioral pressure to abandon

Multi-factor investor (2010-2020):

• Value drag offset by momentum and quality gains
• More moderate tracking error
• Easier to stay the course

The premium you capture might be slightly lower than the best-performing factor, but you're more likely to actually capture it because you can stick with the strategy.

Implementation Approaches

There are two main ways to build a multi-factor portfolio:

Approach 1: Single-Factor ETFs ("Factor Sleeve" Approach)

Buy separate ETFs for each factor you want:

| Factor | Example ETF | |--------|-------------| | Value | VTV, IVE, RPV | | Quality | QUAL, SPHQ | | Momentum | MTUM, PDP | | Low Volatility | USMV, SPLV | | Small-Cap | IWM, VB |

Advantages:

• Full transparency—you know exactly what you own
• Control over factor weights
• Can rebalance tactically (though we'll discuss why this is hard)
• Lower expense ratios for individual ETFs

Disadvantages:

• Multiple positions to manage
• Potential overlap (the same stock might appear in multiple factor ETFs)
• Rebalancing complexity and costs
• More complex tax management

Approach 2: Integrated Multi-Factor ETFs

Buy a single product that combines multiple factors internally:

• LRGF (iShares U.S. Equity Factor): Value, quality, momentum, size
• GSLC (Goldman Sachs ActiveBeta): Value, momentum, quality, low volatility
• VFMF (Vanguard U.S. Multifactor): Value, momentum, quality

Advantages:

• Simplicity—one holding
• Professional integration of factors
• Internal rebalancing
• Often better at avoiding overlap

Disadvantages:

• Less control over factor weights
• "Black box" feel—harder to see exactly how factors are combined
• May include factors you don't want
• Typically higher expense ratios than individual factor ETFs

Which Approach Is Better?

Neither is universally superior. Consider:

• Portfolio size: Smaller portfolios benefit from multi-factor ETF simplicity
• Time/interest: Factor sleeves require more attention
• Control preference: Do you want to adjust factor weights?
• Tax situation: Factor sleeves offer more tax-loss harvesting opportunities

Factor Timing: The Tempting Trap

If factors have cycles, why not time them? Go heavy on value when it's "due" to recover; shift to momentum when it's working.

This is extremely difficult and most investors should avoid trying.

Why Factor Timing Fails

Factors don't mean-revert reliably: Just because value has underperformed for a decade doesn't mean it will outperform the next decade. Factor cycles don't follow predictable patterns.

Costs compound: Shifting between factors generates trading costs and potential tax consequences. Active factor timing needs to be very right to overcome these frictions.

Behavioral traps: Most factor timing is actually performance chasing in disguise. Investors shift toward whatever worked recently, buying high and selling low at the factor level.

The evidence is weak: Academic studies of factor timing show mixed results at best. Some signals have modest predictive power, but not enough to reliably profit after costs.

What Might Work (Slightly)

Some research suggests modest factor timing signals:

• Valuation spreads: When value stocks are unusually cheap relative to growth, the value premium may be larger going forward
• Momentum crash indicators: High market volatility may signal momentum crash risk
• Crowding measures: Extremely crowded factors might have lower future premiums

But the effects are small, noisy, and easy to over-fit. Most investors are better off with static allocations.

Practical Considerations for DIY Investors

How Much Factor Tilt Is Enough?

This is the key question. A portfolio that's 100% factor ETFs has maximum factor exposure but also maximum tracking error versus the market.

Conservative approach:

• Core position: 60-80% broad market (like VTI)
• Factor tilt: 20-40% in factor ETFs
• Expected tracking error: 1-3% annually

Aggressive approach:

• All equity allocation in factor ETFs
• No "market" position—factors are your equity exposure
• Expected tracking error: 4-8% annually

More tilt = more potential premium = more risk of extended underperformance.

Which Factors to Include?

Not all factors are equally robust or easy to implement. A reasonable shortlist:

High conviction:

• Market (beta)
• Value (despite recent struggles, evidence is strong)
• Profitability/Quality
• Momentum

Moderate conviction:

• Low Volatility
• Size (weakening evidence)

Avoid:

• Exotic factors with limited evidence
• Factors based on single studies
• "Factors" that are really just sector bets

Costs Matter

Factor ETFs charge more than plain index funds. A typical expense ratio comparison:

| Fund Type | Typical Expense Ratio | |-----------|----------------------| | Total market (VTI) | 0.03% | | Single-factor ETF | 0.15-0.25% | | Multi-factor ETF | 0.20-0.35% |

If factor premiums are 2-3% and costs are 0.2%, you're keeping most of the premium. But if premiums are actually lower going forward, costs become more significant.

Rebalancing

Factor exposures drift over time as different factors perform differently. Annual rebalancing is generally sufficient—more frequent rebalancing adds costs without clear benefit.

For factor sleeves, set target weights and rebalance when positions drift more than 5% from targets, or annually, whichever comes first.

A Sample Multi-Factor Portfolio

Here's one reasonable approach for a DIY investor:

| Holding | Weight | Purpose | |---------|--------|---------| | VTI (Total US Market) | 40% | Core market exposure | | VTV (Vanguard Value) | 15% | Value factor | | QUAL (iShares Quality) | 15% | Profitability/quality factor | | MTUM (iShares Momentum) | 15% | Momentum factor | | VB (Vanguard Small-Cap) | 15% | Size factor |

This provides:

• Broad market foundation
• Diversified factor tilts
• Reasonable costs
• Manageable complexity

Alternatives might use international factor ETFs, substitute USMV for MTUM if you want lower volatility, or use an integrated multi-factor ETF instead of individual factors.

What to Expect

Multi-factor investing isn't a get-rich-quick strategy. Realistic expectations:

Expected premium over market: 1-2% annually (if factors continue to work)

Tracking error: 2-5% annually (you will underperform the market in some years)

Time horizon: 10+ years minimum to have confidence in capturing premiums

Behavioral challenge: Watching simple index funds beat you for extended periods is psychologically difficult

The payoff is subtle: over decades, 1-2% additional annual return compounds dramatically. But the path there is bumpy.

Key Takeaways

• Combine factors for diversification: Individual factors have unpredictable cycles; combining them smooths returns
• Factor correlations are low: Value/momentum negative correlation is especially valuable
• Two approaches: Factor sleeves (more control, more complexity) vs. multi-factor ETFs (simpler, less control)
• Avoid factor timing: Static allocations beat most timing strategies
• Costs matter: Factor premiums may be modest; don't erode them with expensive funds
• Set realistic expectations: 1-2% premium, 10+ year horizon, significant tracking error
• Start simple: A multi-factor ETF or 3-4 factor sleeve is plenty

Multi-factor investing applies diversification logic to why you earn returns, not just what you own. It's a sophisticated approach, but one that DIY investors can implement with basic tools. The challenge isn't complexity—it's patience.

---

This is the fifth article in our Factor Investing series. Conclude with The Case Against Factor Investing: Criticisms and Debates for an honest look at the skeptics' arguments.