CD Barbell Strategy, When Skipping the Middle Wins
TL;DR
A CD barbell allocates 40-60% of principal to short-term CDs (3-12 months) and the remainder to long-term CDs (5+ years), skipping the 2- to 4-year middle entirely. The strategy outperforms a traditional ladder when the yield curve is inverted or flat. On $200,000 split 50-50, the barbell yields a blended ~4.65% vs a ladder's ~4.48%. The MYGA-anchored barbell yields ~5.10% blended.
Rates expire like milk.
The best MYGA, CD and savings rates change every few weeks. Get a short email when a rate on this page changes.
We’ll email you when a rate on this page changes. Hans Goldstein · NPN 20602398.
Rather talk it through? Or book 15 minutes on Hans’s calendar.
CD vs MYGA, after tax
What $100,000 earns in 5 years: CD vs MYGA, after tax
Rates used: CD 5.00%, the top nationally available 5-year CD (Bankrate, October 14, 2026); change it to the rate you are offered. MYGA 5.95%, the best A-rated rate we track as of October 13, 2026: Delaware Life Pinnacle Plus 5-year (AM Best A-, negative outlook, 7/31/2026), $100,000 and up, effective October 6, 2026 (Delaware Life rate sheet).
| CD, interest taxed every year | $121,081 |
|---|---|
| MYGA, tax-deferred, before tax | $133,507 |
| MYGA after paying tax on all the interest at withdrawal | $126,136 |
| Tax deferral alone adds about | $674 |
| MYGA ahead of the CD, after tax | $5,055 |
Assumes 22% federal and 0% state income tax, no withdrawals, rates held for the whole term. MYGA interest is taxed as ordinary income when you withdraw it, not as capital gains. Gains taken before age 59 1/2 can owe a 10% IRS penalty, and cashing out early can cost a surrender charge. In an IRA both grow tax-deferred, so the edge there is the rate alone. Bank and credit union deposits are FDIC or NCUA insured up to the limits; a MYGA is backed by the issuing insurer's claims-paying ability and your state's guaranty association up to its limits, not the FDIC. General education, not tax, legal or financial advice.
The thesis in one sentence
When the middle of the yield curve pays less than both the short end and the long end, allocating to the middle is paying tuition for nothing. The barbell takes the two highest-yielding points and ignores the rest.
What the 2026 curve looks like
Indicative best-available CD APYs as of mid-2026:
| Term | Best APY | Curve shape note |
|---|---|---|
| 3-month | 4.75% | Anchored near Fed Funds |
| 6-month | 4.70% | Slight discount |
| 1-year | 4.70% | Flat to 6-month |
| 2-year | 4.45% | Belly discount |
| 3-year | 4.30% | Lowest point |
| 4-year | 4.40% | Recovering |
| 5-year | 4.55% | Term premium |
| 5-yr MYGA | 5.55% | Insurance-side premium |
The 3-year is the worst-yielding point on the bank-CD curve. A ladder that allocates 20 percent of principal there is, by construction, dragging blended yield down. The barbell avoids this entirely.
Worked example: $200,000 at 50-50
Split: $100,000 into a 6-month CD at 4.70 percent, $100,000 into a 5-year CD at 4.55 percent.
Blended 5-year math. The short rung rolls 10 times over 5 years; assume reinvestment at the same 4.70 percent (rate-flat scenario). The long rung locks at 4.55 percent for the full period.
- Short side compounded at 4.70%: $100,000 × (1.047)5 = $125,800. Interest: $25,800.
- Long side compounded at 4.55%: $100,000 × (1.0455)5 = $124,800. Interest: $24,800.
- Total interest: $50,600. Blended effective yield: ~4.65 percent.
Comparison: traditional 5-rung ladder on $200,000. $40K each at 4.70 / 4.45 / 4.30 / 4.40 / 4.55. Assuming flat reinvestment at the same rates, total interest over 5 years is approximately $48,300. Blended yield: ~4.48 percent.
Comparison: 3-year bullet on $200,000. Locked at 4.30 percent for 3 years, then forced to reinvest at the unknown year-3 rate. Year 1 to 3 interest: $26,900. The reinvestment risk in years 4 and 5 is unhedged.
Comparison: MYGA-anchored barbell. $100,000 in 6-month T-bills rolling at 4.85 percent, $100,000 in a 5-year MYGA at 5.55 percent.
- Short side: $100,000 × (1.0485)5 = $126,700. Interest: $26,700.
- Long side: $100,000 × (1.0555)5 = $130,900. Interest: $30,900.
- Total interest: $57,600. Blended effective yield: ~5.10 percent.
The MYGA-anchored barbell adds $7,000 of interest over 5 years versus a CD-only barbell on the same $200,000 starting principal, plus tax-deferred compounding on the long side.
Why the barbell works on an inverted curve
Inverted curves penalize the middle. A flat or slightly inverted curve in mid-2026 (3-year at 4.30 percent versus 6-month at 4.70 percent and 5-year at 4.55 percent) means the middle terms are charging you yield for taking on more duration risk than the short end without compensating you for the additional lock-up.
If you believe the Fed will hold or cut, the long end locks the current rate before it can drop. If you believe the Fed will hike, the short end rolls into the new higher rate within months. The middle does neither. It is the worst of both worlds when the curve is flat or inverted.
Weighting the two ends
The split between short and long is the only real decision in barbell construction. Three common weightings:
| Weighting | Short / Long | Best for |
|---|---|---|
| Liquidity-tilt | 60% / 40% | Emergency-fund buyers; uncertain near-term cash needs |
| Balanced | 50% / 50% | Generic conservative income; default for most retirees |
| Yield-tilt | 30% / 70% | Maximum blended yield; bets on stable or falling rates |
The yield-tilt is mechanically the highest-yielding because the long end out-yields the short end by 100+ bps when you replace the long CD with a MYGA. The liquidity-tilt is for buyers who suspect they may need access within 12 months.
When this beats simpler approaches
- Yield curve is flat or inverted. A normal-sloped curve favors the ladder; a flat or inverted curve favors the barbell.
- You have a clear long-horizon bucket plus a near-term liquidity need. The barbell maps cleanly onto a 50 percent "I will need this in 12 months" and 50 percent "this is for year-5+ income" allocation.
- You can use a MYGA on the long end. The MYGA structurally improves the math by 80 to 100 bps on the long side, which is where most of the dollar-weighted yield comes from.
- Allocation is large enough to clear MYGA minimums. Most A-rated MYGAs require $25K to $50K minimum; many top-rate carriers want $100K+.
When simpler is better
- You have under $50K to allocate. The barbell needs scale to clear minimums on both ends; small accounts should use a single bullet or short ladder.
- The curve is normal-sloped and steep. Each additional year of duration is being paid for; a ladder or long bullet captures that better.
- You cannot tolerate any 5-year lock. The long end of a barbell is, by design, locked for years. If a 12-month maximum lock is your ceiling, run a 3-month / 6-month / 12-month rolling ladder instead.
- You want one account, one maturity, one statement. The barbell is at least two accounts; operational complexity is real.
Where a MYGA replaces the long end
The single highest-leverage change to a CD barbell is replacing the long-end CD with a MYGA. The MYGA pays 80 to 110 bps more yield, defers tax on the credited interest, and (unlike a long CD) typically allows 10 percent annual penalty-free withdrawal after year 1.
The trade is FDIC insurance versus state guaranty association coverage, plus the surrender schedule. For long-horizon money where you have no realistic plan to break the lock, the MYGA is the better instrument. See our CD ladder vs MYGA ladder analysis for the side-by-side math.
Operational checklist
- Confirm the curve is flat or inverted before committing to a barbell. If the 3-year out-yields the 6-month by 25+ bps, use a ladder instead.
- Split principal 50-50 as the default. Adjust to 30-70 long-tilt only if you are comfortable with the longer lock.
- Use a top-yielding online bank or T-bills for the short end; minimum-deposit thresholds matter at the small end.
- Use a MYGA from an A-rated carrier for the long end if the long-CD spread to the MYGA is more than 60 bps (it almost always is in 2026).
- Re-evaluate the short-end roll every maturity. If 6-month yields drop below the 5-year, consider extending the rolling short rung.
Related strategy guides
- CD ladder vs bullet strategy and reinvestment risk
- CD ladder versus MYGA ladder side-by-side
- CD rates when the Fed is cutting
- 12-item due-diligence checklist before you open
Frequently asked follow-up questions
Want my independent take on whether a CD or MYGA fits your situation?
I'm a licensed independent producer (NPN 20602398) appointed with multiple A-rated carriers. I'll compare what your bank is offering against the top MYGA rates I see this week, and tell you straight which one fits your timeline, tax bracket, and liquidity needs.
No cost, no obligation. Written second opinion within one business day.
Rather talk it through? Or book 15 minutes on Hans’s calendar, or call 213-414-2808.
Hans Goldstein · 213-414-2808 · NPN 20602398, independent licensed producer
We’ll email it to you. Hans replies within one business day. Privacy Policy.
Disclosure
This article reflects publicly available CD, savings, and annuity rate information approximate to the date above. Rates change frequently, often weekly. Always confirm current rates directly with the institution before opening, renewing, or transferring. This is general educational content, not a personalized recommendation, solicitation, or offer of any specific product. Hans Goldstein is an independent licensed insurance producer (NPN 20602398) appointed with multiple A-rated carriers in the fixed-annuity market; Goldstein & Co. LLC is not a bank, broker-dealer, or registered investment adviser. CDs are deposit products of FDIC-insured banks or NCUA-insured credit unions; annuities are insurance contracts backed by the issuing carrier and state guaranty associations. FDIC and NCUA insurance limits are typically $250,000 per depositor per institution per ownership category. Tax discussion reflects federal law as of 2026 and is subject to change; consult a tax professional for your situation.