Structured Products Teardown: How Banks Charge 3% and How to Replicate for Free

"Participate in 100% of S&P 500 upside with complete principal protection." Sounds perfect—until you understand that JP Morgan's 3-year principal-protected note embeds a 2.8% annual fee that your advisor doesn't mention. The note is a zero-coupon bond (your protection) plus a cheap call option (your participation)—and the bank keeps the difference. Citadel and Goldman's own prop desks replicate these structures for $0 in embedded fees. So can you.

Executive Summary

The Structured Products Universe ($7 Trillion Global Market):

• Principal-Protected Notes (PPNs): Return your principal at maturity + some upside participation. Cost: 2-4% embedded annual fee.
• Autocallable Notes: Pay high coupons (8-12%) if the market doesn't fall below a barrier. If it does: you absorb losses. Cost: complexity premium 1-2%.
• Buffered ETFs (Defined Outcome ETFs): Limit downside (buffer) and upside (cap). Examples: PAUG, PJAN, BJUL. Cost: 0.79% explicit + implicit carry cost.
• Equity-Linked Notes (ELNs): Short-term notes with equity-linked coupons. Used by JEPI. Often fair value but opaque.

The Key Question for Every Structured Product:

• What are the components? (Usually: a bond + options)
• What does each component cost in the open market?
• What is the "all-in" fee (explicit + implicit)?
• Can I replicate this myself more cheaply?
• Does the structure serve a genuine need I can't address otherwise?

Part 1: Principal-Protected Notes—The Best-Selling Structured Product

How the Bank Builds It

Example: JP Morgan 3-Year S&P 500 Principal-Protected Note

• You invest $100,000
• At maturity (3 years): Get back minimum $100,000 (protected) + 80% of S&P 500 gains
• JP Morgan charges: $0 explicit fee (seems free!)

How JP Morgan actually structures it:

Step 1: Buy a zero-coupon bond

• JP Morgan buys a 3-year zero-coupon Treasury bond for $86,260 (at 5% yield, $86,260 grows to $100,000 in 3 years)
• This is your "principal protection" — guaranteed to return $100K at maturity
• Cost to JPM: $86,260

Step 2: Buy a call option on the S&P 500

• With the remaining $100,000 - $86,260 = $13,740, JP Morgan buys a 3-year at-the-money call option on the S&P 500
• A 3-year ATM call on SPX costs approximately $18,000-$22,000 per $100K notional (using Black-Scholes at 18% implied vol)
• JP Morgan has $13,740 but the full call costs ~$20,000
• They can only afford to buy 13,740/20,000 = 68% of a full call
• This is why you only get "80% participation" (approximately — varies by interest rates and vol)

The math: Where the fee hides

Total embedded cost: ~2.1% per year. Plus bid/ask on bond (~0.3%), plus structured note illiquidity premium (~0.3%), plus advisor commission (~0.5-1%).

Total all-in cost: 3.0-3.7% annually. For comparison, VTI costs 0.03%.

What You Actually Receive vs. What You Could Have Built

The note's performance in three scenarios:

The principal protection benefit is only realized if the market falls. In all other scenarios, the note underperforms significantly.

Part 2: DIY Replication — Build Your Own PPN

The Two-Component Replication

Anyone with a brokerage account can build a principal-protected note. Here's how:

Step 1: Calculate bond component

Goal: Have $100,000 in 3 years.
Current 3-year Treasury yield: 4.8%

Bond component needed = $100,000 / (1 + 0.048)^3 = $87,138

Purchase: Buy $87,138 of TBills/TIPS/T-notes maturing in 3 years.
This is GUARANTEED to return $100,000 in 3 years (Treasury = zero default risk).
                    

Step 2: Buy call options with the remaining cash

Remaining capital: $100,000 - $87,138 = $12,862

Option strategy: Buy LEAPS (Long-term Equity Anticipation Securities)
  - SPY or SPX calls, 3 years out, at-the-money
  - $12,862 buys you approximately 65-70% of a $20K ATM call
  - This gives you 65-70% upside participation

Alternatively:
  - Buy 1 SPY LEAPS call at delta 0.50 (ATM), Jan 2029 expiry
  - Cost: ~$45-55 per share (contract = 100 shares × $50 = $5,000)
  - $12,862 / $5,000 = 2-3 contracts
  - Equivalent to owning 200-300 shares of SPY at a fraction of the cost
                    

Total DIY cost:

• Treasury purchase: $0 commission at Fidelity/Schwab/Treasury Direct
• SPY LEAPS options: $0 commission (Fidelity, Schwab, IBKR) + bid/ask spread (~0.10%)
• Total: ~0.10% vs. 3.0-3.7% for JPM note. You save ~2.9% per year.
• On $100K over 3 years: $8,700 saved

When the Bank's Note IS Worth Buying

Be fair: there are situations where structured products provide genuine value:

1. You genuinely cannot tolerate any loss — e.g., this money must be there for a specific purpose in 3 years (college tuition, known expense) and you CANNOT risk a bear market. The insurance cost may be worth it psychologically.
2. Interest rates are very high — At 8% rates, the zero-coupon bond costs only $79K, leaving $21K for options → you get 100% participation AND full protection. This was the case in the 1980s-90s when PPNs were genuinely compelling.
3. You don't have options approval — DIY replication requires options Level 2+ access. If you can't get this (or won't), a buffered ETF might be a simpler alternative.
4. Tax-deferred account — In an IRA, the structured note's ordinary income treatment is neutralized. The option replication's tax advantages disappear. More comparable net-of-tax.

Part 3: Autocallable Notes—The Yield-Seeker's Trap

How Autocallables Work

Autocallables are the most complex and widely-sold structured product. They offer high coupons (8-14% annually) in exchange for taking on specific downside risk.

Example: Goldman Sachs 2-Year Autocallable Note on S&P 500

• Pays 11% annual coupon IF S&P 500 stays above 75% of its starting level
• Autocalls (matures early) if S&P 500 is above starting level on any quarterly observation date
• If S&P 500 falls below 75% at maturity: You absorb all losses below 75% (i.e., if market is down 40%, you lose 40%)

The "barrier" risk everyone ignores:

• The 75% barrier sounds safe — the S&P 500 rarely falls 25%+
• But: In 2008, S&P fell 38%. In 2020 COVID, it fell 34%. In 2022, -25%. All would breach the 75% barrier.
• Frequency of 25%+ drawdowns: ~1 in 7-10 years historically
• If you hold 10 autocallable notes, expect 1-2 to breach the barrier

Autocallable Deconstruction

Autocallables are built from:

• Short put option at 75% strike (you receive premium for taking downside below 75%)
• Bond component (interest rates finance the coupon)
• Structured barrier: Observation dates, autocall feature

The put you're effectively selling:

• A 2-year S&P 500 put at 75% strike costs approximately 4-6% in implied volatility premium annually
• Plus the bond yield covers another 4-5%
• Together: 8-11% coupon. That's the bank's formula.
• You're essentially selling an S&P 500 put. The bank wraps it in complexity, calls it a "note," and charges 1-2% more than the raw option value.

DIY alternative: Sell a 2-year SPY put at the 75% strike directly:

• Same economic exposure
• No bank wrapper fee (1-2% savings)
• Full transparency of your actual risk
• Cash immediately received (put premium) vs. periodic coupon payments
• Requires $50K+ in margin for appropriate position sizing

Part 4: Buffered ETFs — The Most Transparent Structured Product

How Defined Outcome ETFs Work

Buffered ETFs (Innovator Capital's PAUG, PJAN, BJUL; First Trust's BUFR) are the most retail-friendly structured products because they're fully transparent ETFs with daily liquidity.

Example: PAUG — Innovator S&P 500 Power Buffer (August)

• BUFFER: First 15% of losses are absorbed by the product (you lose nothing if S&P falls 0-15%)
• CAP: Maximum upside is approximately 15-18% for the annual outcome period
• TERM: One year (August to August)
• COST: 0.79% expense ratio + implicit carry cost

How Innovator builds it (transparent in fund prospectus):

• Buys a 1-year ATM call on SPDR S&P 500 (upside participation)
• Sells a 1-year call at +15% (cap — defines your maximum gain)
• Buys a 1-year put at 100% (protects you from first 15% loss)
• Sells a 1-year put at 85% (below 15% buffer, you absorb losses again)
• This is a "collar spread" — completely standard options strategy

The cost analysis:

When PAUG is worth 1.75-2.15%/year:

• You need equity exposure but have a specific near-term liability (pension lump-sum decision in 12 months)
• You're in the "red zone" of retirement (5 years before FIRE) and a 15%+ loss would delay retirement
• You want to stay invested during a period of elevated market uncertainty without emotional panic selling
• In these cases, the 15% buffer might be worth 1.5-2% in "psychic value"

When PAUG Is NOT Worth It

• You're in a long accumulation phase (20+ years). Over 20 years, the compound cost of 2%/year vs. VTI at 0.03% is enormous (~37% of final portfolio value)
• You have substantial cash reserves (2-3 years of expenses). Cash is your buffer—you don't need to pay for it in your equity sleeve
• You can handle 15%+ drawdowns emotionally. Then the buffer provides no psychological value, and you're just paying for it for nothing

Part 5: Complete Replication Toolkit

Python Option Pricer — Value Any Structured Product

import numpy as np
from scipy.stats import norm
import pandas as pd

def black_scholes(S, K, T, r, sigma, option_type='call'):
    """
    Black-Scholes option pricer.
    S: Current stock price
    K: Strike price
    T: Time to expiration (years)
    r: Risk-free rate (annual)
    sigma: Implied volatility (annual)
    Returns: option price as % of underlying
    """
    d1 = (np.log(S/K) + (r + 0.5*sigma**2)*T) / (sigma * np.sqrt(T))
    d2 = d1 - sigma * np.sqrt(T)

    if option_type == 'call':
        price = S * norm.cdf(d1) - K * np.exp(-r*T) * norm.cdf(d2)
    else:  # put
        price = K * np.exp(-r*T) * norm.cdf(-d2) - S * norm.cdf(-d1)

    return price / S  # Return as % of underlying

def analyze_structured_product(
    principal=100000,
    term_years=3,
    risk_free_rate=0.048,
    implied_vol=0.18,
    participation_rate=0.80,
    buffer_pct=0.15,
    cap_pct=None,
    product_type='ppn'  # 'ppn', 'buffered', 'autocall'
):
    """
    Deconstruct any structured product into its components.
    Calculate fair value vs. actual price.
    """
    S = 100  # Normalized price

    print(f"\n{'='*55}")
    print(f"STRUCTURED PRODUCT TEARDOWN")
    print(f"{'='*55}")
    print(f"Type:           {product_type.upper()}")
    print(f"Principal:      ${principal:,.0f}")
    print(f"Term:           {term_years} years")
    print(f"Risk-Free Rate: {risk_free_rate:.1%}")
    print(f"Implied Vol:    {implied_vol:.1%}")

    if product_type == 'ppn':
        # Component 1: Zero-coupon bond
        bond_cost = principal / (1 + risk_free_rate)**term_years
        bond_pct = bond_cost / principal

        # Component 2: ATM call option
        call_price_pct = black_scholes(S, S, term_years, risk_free_rate, implied_vol, 'call')
        call_pct = call_price_pct

        # Available for options
        available_for_options = 1 - bond_pct
        participation_achievable = available_for_options / call_pct

        print(f"\nCOMPONENT ANALYSIS:")
        print(f"  Zero-coupon bond cost: {bond_pct:.1%} of principal (${bond_cost:,.0f})")
        print(f"  Available for options: {available_for_options:.1%} (${available_for_options*principal:,.0f})")
        print(f"  ATM call cost:         {call_pct:.1%} (${call_pct*principal:,.0f} for full participation)")
        print(f"\n  Achievable participation: {participation_achievable:.0%}")
        print(f"  Advertised participation: {participation_rate:.0%}")
        print(f"  Participation shortfall:  {participation_rate - participation_achievable:.0%}")

        implicit_fee = (participation_rate - participation_achievable) * call_pct / term_years
        print(f"\n  IMPLICIT FEE: {implicit_fee:.2%}/year")
        print(f"  Fee on ${principal:,.0f}: ${implicit_fee*principal:,.0f}/year = ${implicit_fee*principal*term_years:,.0f} total")

        print(f"\nDIY REPLICATION:")
        print(f"  Buy T-note:       ${bond_cost:,.0f} (returns ${principal:,.0f} at maturity)")
        print(f"  Buy LEAPS call:   ${available_for_options*principal:,.0f}")
        print(f"  Participation:    {participation_achievable:.0%} (same as fair value)")
        print(f"  DIY Cost:         ~0.10% (bid/ask only)")
        print(f"  Bank cost:        ~{implicit_fee + 0.005:.2%}/year")
        print(f"  ANNUAL SAVINGS:   ~{implicit_fee + 0.005 - 0.001:.2%}")

    elif product_type == 'buffered':
        # Buffered ETF: Buy call spread + put spread
        # Long ATM call + Short capped call + Long put at 100% + Short put at (100-buffer%)
        cap = cap_pct if cap_pct else 0.15

        long_call = black_scholes(S, S, 1, risk_free_rate, implied_vol, 'call')
        short_call = black_scholes(S, S*(1+cap), 1, risk_free_rate, implied_vol, 'call')
        long_put = black_scholes(S, S, 1, risk_free_rate, implied_vol, 'put')
        short_put = black_scholes(S, S*(1-buffer_pct), 1, risk_free_rate, implied_vol, 'put')

        net_option_cost = long_call - short_call + long_put - short_put
        theoretical_cap = cap  # Approximately—the cap is set so net cost ≈ 0

        print(f"\nCOMPONENT ANALYSIS (Buffered ETF):")
        print(f"  Long ATM call:         {long_call:.2%} of notional")
        print(f"  Short {1+cap:.0%}-strike call:  -{short_call:.2%}")
        print(f"  Long ATM put:          {long_put:.2%}")
        print(f"  Short {1-buffer_pct:.0%}-strike put:  -{short_put:.2%}")
        print(f"  Net option cost:       {net_option_cost:.2%}")
        print(f"\n  Theoretical cap:       ~{cap:.0%} upside")
        print(f"  Buffer:                {buffer_pct:.0%} protection")
        print(f"  Expense ratio:         0.79%")
        print(f"  Total cost to achieve: {net_option_cost + 0.0079:.2%}/year")

# Run examples
print("\n1. JP MORGAN 3-YEAR PRINCIPAL-PROTECTED NOTE")
analyze_structured_product(
    principal=100000, term_years=3, risk_free_rate=0.048,
    implied_vol=0.18, participation_rate=0.80, product_type='ppn'
)

print("\n\n2. INNOVATOR PAUG 15% BUFFER ETF")
analyze_structured_product(
    principal=100000, term_years=1, risk_free_rate=0.048,
    implied_vol=0.16, buffer_pct=0.15, cap_pct=0.16, product_type='buffered'
)
                    

Part 6: JEPI and ELNs — When Structure Adds Value

JEPI's Equity-Linked Notes: Fair Value or Rip-off?

JEPI uses Equity-Linked Notes (ELNs) rather than directly writing covered calls. This is actually legitimate—in an ETF structure, directly selling options creates regulatory complications. ELNs from a counterparty bank (Goldman, JP Morgan, Morgan Stanley) embed the short call exposure and pay monthly income.

Are JEPI's ELNs fairly priced?

• JEPI's ELN terms are disclosed in fund filings (available at SEC EDGAR)
• Academic analysis suggests ELNs trade approximately at fair value (+/- 0.15%)
• JEPI's real cost is the 0.35% expense ratio + low-vol equity sleeve underperformance vs. S&P in bull markets
• Verdict: JEPI is reasonably priced for what it offers. The "structured product" stigma doesn't apply here.

Part 7: Annuities—The Cousin of Structured Products

Fixed Index Annuities (FIAs): The Retirement Version

Fixed index annuities are sold as "participation in market gains without downside risk"—essentially a PPN with lifetime income features. They're the dominant product sold to pre-retirees by commissioned insurance agents.

How FIAs embed their costs:

• Similar bond + call structure as PPNs
• Additional costs: insurance company profit margin (0.5-1%), surrender charges (7-10% if you exit early), mortality risk adjustment
• All-in cost vs. alternatives: 2.5-3.5% annually

When FIAs make sense:

• You need guaranteed lifetime income (annuitization feature)
• You have maxed all tax-advantaged accounts and need another tax-deferral vehicle
• You're in a high tax bracket and the FIA's tax deferral outweighs the higher costs

When FIAs are a bad deal:

• Your advisor earns 6-8% upfront commission (creates enormous conflict of interest)
• You have surrender charges that lock you in for 7-10 years
• You're under 60 (too early—you'll need liquidity)
• You haven't maxed out your 401(k), IRA, and HSA first

Part 8: The Decision Framework

When to Consider, When to Walk Away

The Three Questions Before Buying Any Structured Product

1. "What are the components?" — Demand the term sheet. It must show: bond component, option component, strikes, notional. If the bank refuses to show components, walk away.
2. "What is the all-in cost?" — Explicit fees + implicit (participation shortfall + participation cap + illiquidity premium + advisor commission). Total this number. Compare to alternatives.
3. "Why can't I replicate this more cheaply?" — For most products, you can. The legitimate cases are: you lack options access, you need tax deferral unavailable elsewhere, or you genuinely need guaranteed lifetime income (annuitization).

Conclusion: Knowledge Is Your Protection

Structured products aren't inherently evil—they're financial engineering. At their best, they provide genuine value by combining instruments in ways individual investors can't easily access. At their worst, they're complexity used to obscure fees.

The $7 trillion global structured products market exists because most buyers don't read the term sheets. You now know how to read them. A principal-protected note is a bond plus a cheap call. An autocallable is a short put wrapped in coupons. A buffered ETF is a collar spread in an ETF wrapper.

Armed with this knowledge, you can either:

• Replicate the product yourself for a fraction of the cost, OR
• Pay the bank's fee knowingly—because the product genuinely serves a specific need in your portfolio that you can't address more cheaply

Both outcomes are better than being sold a structured product you don't fully understand.

Further Reading

Research:

• Bergstresser, D. & Beshears, J. (2010): "Who Selected Adjustable-Rate Mortgages? Evidence from the 1989-2007 Survey"—NBER
• FINRA Investor Education: "Understanding Structured Notes with Principal Protection"
• CFA Institute: "Structured Products and Derivatives" (Level 1 curriculum chapter)

Related Articles:

• Systematic Options Income (DIY alternatives to structured products)
• Carry Strategies (volatility carry vs. structured notes)
• Derivatives & Hedging
• Private Credit (legitimate yield alternative)

Interactive Tools:

• Retirement Cash Flow Analyzer
• Investment Analyzer