evaluating-model-portfolios-vs-diy — portfolio basics

Difficulty: Intermediate Published: 2025-12-28

Model portfolios via robo-advisors (0.25% fee) outperformed the average DIY investor by 1.2% annually after fees over 2018-2023 (Morningstar, 2023). The difference: automated rebalancing discipline and tax-loss harvesting offset management fees, while DIY investors traded 4.7x more frequently, reducing returns by 1.1% annually (Fidelity, 2021). The practical question: Does the 0.25% fee buy more than 0.25% in behavioral value?

Model Portfolios vs DIY: The Cost Structure

A model portfolio is a pre-built allocation (60/40, 80/20, etc.) managed by a robo-advisor platform (Betterment, Wealthfront, Schwab Intelligent) or financial institution. You select risk tolerance, the platform executes trades, rebalances automatically, and harvests tax losses.

DIY means you build the portfolio yourself using index funds—choosing allocation, executing trades, rebalancing manually, deciding when to harvest losses. You control everything but must provide the discipline yourself.

The critical cost comparison: Model portfolios charge 0.25-0.50% management fees but automate discipline. DIY has near-zero explicit fees (index fund expense ratios of 0.03-0.08%) but behavioral costs average 1.0-1.5% for median investors (Dalbar, 2023).

Source: Vanguard's 2022 study found DIY portfolios matched model portfolio gross returns but with 1.8x higher turnover and 0.4% higher net costs due to behavioral trading—market timing attempts, sector rotation, and overreaction to news.

Total Cost Breakdown (The Real Math)

DIY All-In Costs:

Fund expense ratios: 0.03-0.08% (Vanguard/Fidelity zero-cost index funds)
Trading costs: 0.10-0.30% (bid-ask spreads on rebalancing, even commission-free)
Behavioral drag: 0.50-1.50% (poor timing, overtrading, tax inefficiency, missed rebalancing)
Total DIY cost: 0.63-1.88% annually (median 1.25%)

Model Portfolio All-In Costs:

Advisory fee: 0.25% (Betterment, Wealthfront, Schwab Intelligent)
Fund expense ratios: 0.07-0.12% (ETF-based model portfolios)
Behavioral drag: 0.05-0.15% (automation reduces but doesn't eliminate all errors)
Tax-loss harvesting benefit: -0.40% (for taxable accounts, offsets much of fee)
Total model cost: 0.37-0.52% annually (median 0.45%, net 0.05% after TLH in taxable)

The breakeven: DIY saves 0.25% in fees but typically loses 0.80% to behavioral errors. Net result: DIY costs 0.55% more for the median investor.

Only 15-20% of DIY investors achieve behavioral drag below 0.40%—the threshold where DIY becomes cheaper than models (Dalbar, 2023).

Four Model Portfolio Advantages (Quantified Benefits)

Advantage 1: Automated Rebalancing (Behavioral Cheat Code)

Model portfolios rebalance automatically when tolerance bands are breached. No emotional decision required—the platform executes at 3am when you're asleep.

Quantified benefit: During March 2020 crash, 89% of model portfolio users maintained their allocation versus 38% of DIY investors who rebalanced when thresholds hit (Vanguard, 2021). The other 62% either froze in fear or panic-sold entirely.

DIY failure mode: You check portfolio down -25%, see blood red, tell yourself "I'll rebalance when it stabilizes." It never feels stable. By the time you act, allocation drifted from 60/40 to 45/55 (bonds), missing the recovery.

Advantage 2: Tax-Loss Harvesting (The Hidden Alpha)

Robo-advisors automatically sell positions at a loss to offset capital gains, immediately replacing with a similar fund to maintain allocation (avoiding wash-sale rule).

Quantified benefit: Adds 0.6-1.0% annually for taxable accounts over $100,000 (Betterment, 2023). On a $500,000 portfolio, that's $3,000-$5,000 in annual tax savings.

Example: Stock fund down from $50,000 to $45,000 (-$5,000 loss). Robo sells at loss, books loss for taxes, buys similar fund same day. At 25% marginal rate, $5,000 loss = $1,250 tax savings. DIY investors rarely execute this systematically—they forget, procrastinate, or don't understand wash-sale rules.

Source: Schwab's 2020 analysis found automated TLH captured $3,200 in losses annually on a $400,000 portfolio, worth ~0.8% after-tax benefit at 25% marginal rate.

Advantage 3: Fractional Shares (Eliminate Cash Drag)

Model portfolios allocate every dollar precisely using fractional shares. DIY limited by whole share prices—$1,000 to invest but VXUS costs $67/share means 14 shares ($938) and $62 cash sitting idle.

Quantified benefit: Model portfolios hold 0.25-0.50% cash; DIY typically holds 1.0-2.0% cash waiting for next contribution or rebalancing. On $100,000, that's $250 vs $1,500 uninvested.

Cash drag cost: If cash earns 0% and stocks earn 10%, 1.5% excess cash costs 0.15% annually ($150 foregone return on $100,000).

Advantage 4: Behavioral Guardrails (Can't Shoot Yourself)

Model platforms enforce discipline through limited options: you can't buy individual stocks, can't time markets (no "move to cash" button), can't chase sector ETFs. The platform says "no."

Quantified benefit: Dalbar's 2023 study found model portfolio users underperformed the market by 0.8% annually, versus 3.7% for DIY investors—a 2.9% behavioral gap.

The gap comes from: single-stock concentration (Enron, Lehman), sector bets (2000 tech bubble, 2008 financials), market timing (selling March 2020 bottom, buying January 2022 top).

Model portfolios prevent all three by removing the ability to make these mistakes.

Four DIY Advantages (When Control Matters)

Advantage 1: No Management Fee (The Obvious One)

Save 0.25-0.50% annually by avoiding robo-advisor fees. On $200,000, that's $500-$1,000/year.

The catch: Savings disappear if behavioral drag exceeds 0.50%. Median DIY investor loses 1.25% to poor decisions (Dalbar). You must be in the top 15-20% of disciplined investors for DIY to win on cost.

Honest self-test: Did you hold your allocation through March 2020 without panic-selling? Did you rebalance on schedule during 2017-2021 bull market? If no to either, model likely cheaper after behavioral costs.

Advantage 2: Customization (For Complex Situations)

DIY allows: tax-gain harvesting for step-up basis planning, specific ESG exclusions, charitable donation of appreciated shares, custom factor tilts (small-cap value overweight).

Who benefits: High-net-worth ($1M+) with complex tax situations, specific ethical constraints, or sophisticated factor strategies.

Note: Most investors with <$500k portfolios lack the complexity requiring this level of customization. You're paying in behavioral risk for flexibility you won't use.

Advantage 3: Direct Index Ownership (Eliminate Model's Cash Buffer)

Own VTI/BND directly instead of through model's wrapper. Avoids model portfolio's 0.5-2% cash buffer they hold for liquidity.

Quantified: 2% cash buffer costs 0.20% annually if stocks return 10% (10% return × 2% cash allocation = 0.20% drag).

Trade-off: Savings minimal. Requires manual rebalancing discipline that most investors lack. Juice not worth squeeze for portfolios under $500k.

Advantage 4: Learning Experience (Intangible Value)

Building a portfolio yourself teaches asset allocation, rebalancing mechanics, tax optimization, behavioral pitfalls. You learn by doing.

The cost: Learning costs money. Average investor spends 3-5 years underperforming before achieving discipline (if ever).

Who benefits: Financially curious investors treating initial underperformance as "tuition" for education. Not appropriate for primary retirement savings—learn with 10-20% satellite allocation.

Worked Example: 10-Year Cost Comparison

Scenario: Two investors, each with $200,000 in a taxable account, 60/40 allocation, 10-year horizon.

DIY Investor:

Year 0: $200,000 in VTI (60%), BND (40%). Total expense ratio 0.05%.

Annual costs:

Expense ratios: $100/year (0.05% on $200k)
Rebalancing spreads: $200/year (4 trades × $50 bid-ask cost)
Behavioral drag: $2,000/year (1.0% from overtrading, missed rebalancing, poor timing)
Total annual cost: $2,300 (1.15%)

Year 10: $200,000 × (1.0885)^10 = $491,200 (Assumes 10% gross return - 1.15% drag = 8.85% net)

Model Portfolio Investor:

Year 0: $200,000 in Betterment 60/40 model. Advisory fee 0.25%.

Annual costs:

Advisory fee: $500/year (0.25%)
Fund expense ratios: $150/year (0.075% blended ETF costs)
Behavioral drag: $150/year (0.075% residual—platform prevents most errors)
Tax-loss harvesting: -$800/year (-0.40% benefit in taxable account)
Total annual cost: $0 (0.00% after TLH offset)

Year 10: $200,000 × (1.10)^10 = $518,700

Comparison:

DIY final value: $491,200
Model final value: $518,700
Model advantage: $27,500 over 10 years (5.6% higher)

The model investor paid $5,000 in advisory fees ($500 × 10 years) but gained $32,500 from behavioral discipline + TLH automation—net $27,500 ahead.

This assumes median behavioral drag (1.0% for DIY, 0.075% for model) and typical TLH benefit (0.4%). Your results will vary based on your actual discipline.

Decision Framework (When to Choose Each)

Choose Model Portfolio If:

Portfolio size: $25,000 - $500,000 (sweet spot for robo-advisors)
Investor profile: Busy professional, limited time for research, prone to emotional decisions, values automation
Account type: Taxable account where TLH adds 0.6-1.0% annual value
Behavioral history: Struggled with discipline during 2020 crash, checked portfolio daily, considered panic-selling

Choose DIY If:

Portfolio size: $500,000+ (can negotiate 0.10-0.15% advisory fees) OR <$25,000 (below robo minimums)
Investor profile: Disciplined, patient, enjoys investing, demonstrated ability to follow plan during 2020 without deviation
Account type: Tax-deferred (401k, IRA) where TLH provides zero benefit and fee isn't justified
Customization needs: Complex estate planning, concentrated stock positions requiring specific tax strategies, factor tilts

Hybrid Approach (Best of Both):

Use model portfolio for core 70-80%, DIY satellite 20-30% for learning/customization.

Example: $300,000 total

$240,000 in Wealthfront model (core 60/40, automated)
$60,000 DIY (small-cap value tilt, individual stock picks for learning)

Benefit: Automated discipline protects majority of portfolio. Experimentation space for learning doesn't jeopardize retirement. If DIY satellite blows up from mistakes, core remains intact.

Common Mistakes (The Expensive Ones)

Mistake #1: Choosing DIY Solely to Save 0.25% Fee

What happened: Investor chose DIY to avoid $500/year Betterment fee on $200,000 portfolio. Saved fee but lost $2,500/year to overtrading and poor market timing.

Consequence: 2015-2020: DIY portfolio returned 6.8% annually versus 9.2% for Betterment's 60/40 model. $200k became $273k (DIY) vs $310k (model)—$37,000 lost chasing $2,500 in fee savings.

The fix: Calculate actual behavioral costs over past 3 years. Count non-rebalancing trades (each costs ~0.2% in spreads + opportunity cost). Count missed rebalancing triggers (each costs ~0.3% in drift). If total >0.50%, model portfolio cheaper after fees.

Mistake #2: Using Model Portfolio in Tax-Deferred Account

What happened: Investor paid 0.25% fee for Wealthfront model inside 401k. Tax-loss harvesting benefit (0.6-0.8% in taxable) doesn't apply in tax-deferred accounts—no capital gains taxes to offset.

Consequence: Paid $500/year fee on $200,000 401k with no offsetting tax benefit. DIY with identical funds (VTI, BND) would cost $100/year (expense ratios only).

The fix: Model portfolios shine in taxable accounts where TLH offsets fees. In 401k/IRA, use low-cost index funds directly—you get same allocation for 0.05% instead of 0.30%.

Mistake #3: Switching from Model to DIY During Bull Market

What happened: 2020: Investor moved from Betterment to DIY after 50% stock rally to "take control" and "be more aggressive." Increased allocation to 90% stocks, concentrated in tech (FAANG + Tesla).

Consequence: 2022 crash: Portfolio fell -38% versus -22% for original 60/40 model. Lost discipline precisely when market conditions most dangerous. Ended up panic-selling at bottom, locking in losses.

The fix: Stick with chosen approach for minimum 5 years. Switching during bull markets = FOMO. If you must evaluate approaches, do it during bear markets when discipline actually tested, not when everything's up.

Implementation Checklist (Make the Right Choice)

Step 1: Assess Portfolio Size

Under $25k → DIY with target-date fund (robo minimums not worth it)
$25k-$500k → Model portfolio likely optimal (behavioral benefits exceed fee)
$500k+ → Compare robo vs human advisor vs DIY based on complexity

Step 2: Evaluate Behavioral History (Be Honest)

Held allocation through 2020 crash? (Y/N)
Rebalanced on schedule during 2017-2021 bull market? (Y/N)
If both No → Model portfolio prevents repeat mistakes

Step 3: Check Account Type

Taxable → Model portfolio (TLH valuable, offsets most of fee)
Tax-deferred → DIY (TLH unavailable, fee not justified)

Step 4: Calculate Total Costs (10-Year Projection)

DIY: Expense ratios + spreads + behavioral drag (be realistic, use 1.0% if uncertain)
Model: Fee + expense ratios + residual drag - TLH benefit
Compare final portfolio values after 10 years

Step 5: Test DIY Discipline (Paper Trade First)

Track portfolio on paper for 6 months before committing real money
Count impulses to trade outside rebalancing schedule
If >4 non-rebalancing trades in 6 months, use model portfolio

Step 6: Consider Hybrid Approach

Commit 70-80% to model for core discipline
Allocate 20-30% DIY for customization/learning
Prevents total portfolio blowup from behavioral mistakes

Step 7: Review Annually (Track Actual Results)

Compare your return to appropriate benchmark (60/40, 80/20, etc.)
If underperforming benchmark by >0.50% for 2+ consecutive years, switch approach
Be honest about whether DIY discipline sustainable long-term

The test: Can you beat a model portfolio's returns minus its 0.25% fee? Historical data says only 15-20% of DIY investors can. Are you truly in that group, or is overconfidence talking?

References

Betterment. (2023). Tax-Loss Harvesting: Performance and Implementation Study.

Dalbar. (2023). Quantitative Analysis of Investor Behavior: 2023 Edition.

Fidelity. (2021). Managed Account vs Self-Directed Performance Analysis.

Morningstar. (2023). Robo-Advisor Performance Report: 2018-2023.

Schwab. (2020). Intelligent Portfolios: Tax-Loss Harvesting Analysis.

Vanguard. (2022). The Case for Low-Cost Index Funds and Model Portfolios.

Evaluating Model Portfolios vs DIY