Contrarian playbook from a pre-retirement portfolio rebuild — $2.5M, retiring at 57, US large cap concentrated, skipping TLH
Late-40s couple, $2.5M across taxable + 401k + Roth + HSA. Retiring 57/55. Just spent serious time rebuilding the strategy and landed on something that goes against a lot of conventional ChubbyFIRE advice. Posting because I think the gain-harvest math is underdiscussed here and the TLH critique is unpopular but worth debating.
Allocation philosophy: US large cap concentrated through accumulation, diversify in 401k right before retirement
Standard advice says diversify internationally and across asset classes throughout accumulation. I'm doing the opposite. Reasoning:
- US large cap has outgrown international by such a wide margin over 15+ years that even after major drawdowns, post-crash balances exceed "diversified" portfolios that supposedly protected on the way down
- I optimize for terminal wealth at age 90, not Sharpe ratio at age 50
- The volatility I "saved" by holding VXUS would cost me hundreds of thousands in compounding by age 65
- I can accept a 50% drawdown in 2027 because I'm still adding to it and not selling until 57
The 401k rebalance is the key insight that makes this work
Most people forget: rebalancing inside a 401k is completely tax-free. No capital gains, no wash sales, no 1099. At age 55 (2 years before retirement) I can shift $1.7M of allocation from S&P 500 institutional fund to 70/20/10 US/Int'l/Bonds in an afternoon for $0 in tax.
This gives me:
- Growth tilt during the 20+ year compounding window that matters most
- Defensive allocation right when sequence-of-returns risk peaks (first 5 years of retirement)
- Zero tax friction on the pivot
Why I'm skipping tax loss harvesting
I know this is heresy here but hear me out:
- TLH only defers tax — it reduces basis, meaning bigger gain later
- For a buy-and-hold investor with no plans to sell until retirement, the deferral is solving a problem I don't have
- The $3K annual ordinary income offset = ~$700 in actual tax savings for someone in my bracket. Not worth the complexity
- Robo-advisor Direct Indexing left me with 113 individual stock positions when I tried to leave Wealthfront. The cleanup is brutal: ACAT to self-directed broker, then weeks of consolidating small positions, unwanted dividend stocks, fractional shares
- Wash sale risk multiplies across accounts (spouse + IRAs = same taxpayer per Rev. Rul. 2008-5) — coordinating TLH across 6 accounts is a tax-prep nightmare
My gain-harvest plan ages 57-62 (the unsung ChubbyFIRE move)
This is where I think the strategy really pays off:
- After retirement at 57: no W-2, no Social Security yet (delaying to 70), no RMDs until 73
- Gap years 2032-2037 give me 6 years of near-zero ordinary income
- 0% federal LTCG bracket up to ~$96,700 MFJ
- I'll harvest ~ $600K of gain realized at 0% federal, 4.5% NC state only
- Resets cost basis on the entire taxable account, eliminating embedded gain
- Versus selling in retirement at 73+ with RMDs pushing me into 15% LTCG bracket + NIIT = 23.3% effective
Tax savings on ~$200K embedded gain alone: ~$20K. Scale to the entire portfolio over 6 years and you're talking $50K-100K saved versus the "buy and hold forever, sell as needed" approach.
Operational diversification matters more than asset diversification at this size
SIPC caps at $500K per broker. At our size we exceed that at any single firm. So I split:
| Broker | Issuer | Holding |
|---|---|---|
| Schwab | Direct ownership | Individual stocks (post-cleanup) |
| Vanguard | Vanguard | VTI + VGT |
| Merrill Edge | iShares | IUSG |
| Employer 401k | State Street | SSSYX (S&P 500 institutional) |
| Wife's 401k | State Street | SSSYX |
| Fidelity Roth × 2 | Fidelity | FSPGX |
| Fidelity HSA | Fidelity | FSPGX |
If one broker gets ransomware'd, frozen by fraud lock, or has a multi-day outage, I lose access to a fraction not everything. Different issuers means a BlackRock event doesn't take down my whole equity sleeve.
Bonus: Merrill Edge holdings get me to BofA Preferred Rewards Platinum Honors ($100K+ tier). Worth $1K-1.5K/year passively.
Why I killed my robo-advisor account after 5+ years
Wealthfront Direct Indexing accumulated to $550K with $350K basis ($200K embedded gain). The TLH "benefit" came with hidden costs I didn't appreciate at signup:
- Can't disable TLH on Direct Indexing (only on basic ETF portfolios)
- Account becomes effectively trapped — selling triggers $46K tax bill at current rates
- ACAT'ing all 113 positions to a self-directed broker is the only realistic exit
- The 0.25% AUM fee compounds to real money on a 7-figure balance
If I'd just bought VTI from day one I'd be in the same financial position with zero complexity and zero exit cost.
TL;DR
- Concentrate in US large cap during accumulation
- Skip TLH if you don't plan to sell until retirement
- Diversify by broker and issuer, not just by asset class
- Rebalance inside 401k at age 55 for tax-free pivot to defensive allocation
- Harvest gains at 0% LTCG in the 57-62 gap years before SS and RMDs
- Hold the taxable account until death for step-up basis on whatever's left
- Avoid robo-advisor Direct Indexing — the exit costs aren't disclosed at signup
Curious if anyone else here is running concentrated + gain-harvest vs. diversified + TLH. The math seems to favor the former pretty heavily for our profile but I want to stress-test it.