Autocallables, under the wrapper:How the structure works, term by term
In one sentence: a contract that pays income while an index stays above a floor, hands your money back early if the index recovers to its starting point, and passes the loss through to you if the index ends far enough down. Everything else is detail. This page walks through all of it, with worked examples.
The Structure
An autocallable is a set of conditional payments
One contract sits underneath every product in this category. It references an index level, fixed on the day the contract is struck. On each observation date it pays a coupon if the reference sits above the coupon barrier. It redeems early, at par, if the reference sits at or above the autocall level. At final maturity, if the reference has finished below the contract’s downside level, a loss passes through to principal.
That is the core of the payoff. A few thresholds and a calendar. How the loss passes through depends on whether the contract uses a barrier or a buffer, covered below.
An autocallable ETF does not hold one contract. It typically holds swap exposure to an index that references a laddered series of them, so strike dates and observation dates are spread across the calendar rather than concentrated on a single day. That reduces reliance on any single strike or observation date. It does not reduce market exposure.
Illustrative mechanics only. The path shown is hypothetical, is not a projection, backtest, or simulation, and does not represent the performance of any fund or index. In this illustration the reference stays above the coupon barrier at each observation and closes above the autocall level, which redeems the contract early at par.
An autocallable pays a regular coupon and returns principal, so long as the reference index does not finish below a set level.
The Terms
Six terms define the payoff
| Coupon barrier | The level the reference must hold on an observation date for that period’s coupon to accrue. Below it, that coupon is not paid. |
| Autocall level | The level that triggers early redemption at par, cancelling the remaining coupon schedule. Rising markets can shorten contract life; realized life depends on the market path and cannot be predicted. |
| Downside term | Barrier or buffer, typically observed only at final maturity. A risk barrier is a cliff: finish below it and principal participates in the full decline, one for one. A risk buffer absorbs the first tranche of decline; only the shortfall beyond it passes through, often multiplied by a gearing factor. Full comparison in the next section. |
| Tenor and non call period | How long a contract can live, and how long before it becomes callable. Additional time is not a recovery mechanism; principal stays exposed to the reference index for longer. |
| Memory feature | With memory, missed coupons accrue and can be paid later if the reference recovers. Without it, each observation stands alone and a missed coupon is gone. Two quoted rates are not comparable until this question is answered. |
| Observation method | When the levels are checked. Scheduled dates only for coupons and calls. A downside level observed once at maturity behaves very differently from one monitored continuously. Same depth, different exposure. |
Every one of these is a number in the offering documents. None of them is visible in a distribution rate.
The six interact. A deep downside level with a shallow coupon barrier, a buffer with a long tenor, memory with monthly observations: each combination prices differently and loses differently. Reading the six together is the whole discipline of comparing these products.
The Downside Term
Barrier vs buffer: same number, different loss
A barrier is a cliff. One threshold, checked at final maturity. Finish above it and principal comes back whole. Finish below it and the entire decline from the initial level passes through. An index at 51% of its strike returns par. An index at 49% returns 49. Two points of index separate a full recovery of principal from a loss of more than half.
A buffer is a slope. The first tranche of decline is absorbed. Only the shortfall beyond the buffer reduces settlement value, and a gearing factor multiplies it. With a 50% buffer and a 200% gearing factor, an index at 49% of strike produces a 1% shortfall, times two, so settlement at 98. At 25% of strike the shortfall is 25%, times two, settlement at 50. The loss is gentler at the edge and accelerates as the index falls further, capped at the actual decline of the index.
Neither shape is safer. They are different answers to where the pain lands, and the buffer is paid for out of the coupon. A contract sorted into a screen by its headline number alone hides which shape it carries.
Hypothetical mechanics only. Illustrates a 50% barrier against a 50% buffer with a 200% gearing factor at final maturity. Not a projection, backtest, or simulation, and not the terms of any specific product.
Side by side, hypothetical terms
| 50% barrier | 50% buffer, 200% gearing | |
| Index finishes at 55 | 100 | 100 |
| Index finishes at 49 | 49 | 98 |
| Index finishes at 44 | 44 | 88 |
| Index finishes at 25 | 25 | 50 |
| Cost of the difference | Higher coupon | Lower coupon |
Settlement value per 100 at final maturity, hypothetical contract, strike 100. Buffer losses beyond the buffer are geared at 200% and capped at the actual decline of the index. Neither design is safer; they are different loss shapes, priced differently through the coupon.
Worked Examples
One contract, three paths
Nobody learns a payoff from definitions. Walk one hypothetical contract through time instead. Strike 100. Coupon barrier 60. Downside level 50, observed only at final maturity. Autocall level 100, checked monthly after a one year non call period. Tenor five years. Coupons monthly, contingent, no memory. A sample of observation dates is shown for each path. All figures are hypothetical mechanics, not projections, backtests, or simulations.
Path A
Recovers and gets called
| Month 3 | Index 97 | Coupon paid |
| Month 6 | Index 91 | Coupon paid |
| Month 12 | Index 96 | Coupon paid |
| Month 15 | Index 104 | Called at par |
Index at or above 100 on the first callable observation. Contract ends at month 15. Every scheduled coupon was paid, principal returned at par, and the remaining coupon schedule is cancelled. The capital now has to be redeployed, potentially into contracts struck on different terms.
Path B
Dips, pauses, survives
| Month 6 | Index 71 | Coupon paid |
| Month 18 | Index 55 | No coupon |
| Month 30 | Index 63 | Coupon paid |
| Month 60 | Index 82 | Par at maturity |
The index never reaches 100 on a callable date, so the contract runs its full five years. Coupons pause on every observation below 60, and with no memory feature those missed coupons are never recovered. At maturity the index sits above 50, so principal returns at par.
Path C
Finishes below the level
| Month 12 | Index 80 | Coupon paid |
| Month 36 | Index 48 | No coupon |
| Month 48 | Index 45 | No coupon |
| Month 60 | Index 44 | Loss applies |
The index finishes at 44, below the 50 level. With a barrier, the full 56% decline passes through: settlement 44 per 100. The same path under a 50% buffer with 200% gearing settles at 88 per 100. The downside term, not the path, decided the outcome.
The three outcomes, generalized
| 1. | The index holds the coupon barrier throughout. Income paid on every observation, principal returned at maturity or on an earlier call. |
| 2. | The index spends time between the coupon barrier and the downside level. Income pauses on those observations, principal returned at maturity. |
| 3. | The index finishes below the downside level at maturity. Income paused and principal impaired, per the contract’s barrier or buffer mechanics. |
The Reference
What the levels are actually measured against
A barrier or buffer is a level on a specific index, and it is not always the benchmark named in conversation. It is often a rules based index constructed on that benchmark, with a volatility target, exposure bounds that can include leverage, and an annual deduction applied on top.
The distinction matters because the levels are measured against the constructed index. A given decline in the underlying benchmark does not translate into the same decline in the reference index, and the relationship shifts with the volatility path. Where exposure bounds allow leverage, the reference index can decline faster than the benchmark itself, carrying it toward its levels sooner. A stated cushion is a cushion on the reference index, not on the benchmark.
Five parameters to locate in any reference index
| Volatility target | Sets exposure to the equity index by formula. Higher measured volatility produces lower exposure, and the reverse. Targets vary meaningfully from product to product. |
| Exposure bounds | The floor and cap on that formula. Any exposure above 100% is leverage, it moves without discretion, and it increases losses as well as gains. |
| Deduction factor | A fixed annual amount subtracted from the index. It accrues every day, in every market condition, and it also differs from product to product. |
| Financing cost | A rate charged on the exposure, usually a benchmark rate plus a spread. |
| Reset frequency | How often exposure is recalculated. Between resets the prior exposure is carried, whatever the market does in the interval. |
Where to find them
All five parameters are set out in a product’s offering documents, usually in the prospectus description of the reference index. No two products in the category share the same set, and each one changes how the barriers behave.
Read them before comparing any two products, because a screen sorted by coupon or by barrier depth silently assumes the reference indices behave the same. They do not.
Pricing The Coupon
The coupon has a source
A coupon is a price, not a gift. It is the payment received for what the contract sold, and it has three components.
| Sold: the upside | Participation in gains above the autocall level. However far the index rises, the contract holder receives coupons and par, nothing more. |
| Sold: the tail | The cushion below the downside level. The holder accepts the loss past the barrier or the geared loss past the buffer. A deeper cushion, a buffer instead of a barrier, or a memory feature all buy protection back, and each is paid for through a lower coupon. |
| Charged: construction | Reference index deductions, financing costs, and transaction costs all contribute to the economics. They accrue regardless of market direction. |
Volatility is what prices all of it. Higher implied volatility on the reference index makes the sold upside and the sold tail worth more, so the coupon quoted against them is larger. A bigger coupon is not a better deal; it is the market’s payment for more risk sold or a more volatile reference.
The useful question is never how large the coupon is. It is what was sold to produce it.
Two contracts on the same index can quote very different coupons because different things were sold. A screen sorted by rate hides exactly that.
What Can Go Wrong
The failure modes are specific
Autocallable strategies tend to struggle in identifiable conditions, and those conditions are worth naming before an allocation, not after.
| Slow declines | A gradual drift lower can carry a reference index through its levels without producing the volatility spike that would reduce exposure. The mechanism that cushions a fast fall may never engage in a slow one. |
| Capped upside | Strong equity markets are not shared beyond the coupon. An early call returns capital at par, so participation in further gains ends there. |
| Coupons can stop | Contingent coupons are not paid for observations where the reference sits below the coupon barrier. In unfavorable conditions an investor may receive little or no income. |
| Geared downside | A buffer reduces the loss at the edge and accelerates it past the edge. With a 200% gearing factor, every additional point of index decline beyond the buffer costs two points of settlement value, capped at the actual decline of the index. Buffer designs trade a cliff for a steeper slope, and pay for it with a lower coupon. |
| Marks, not cashflows | Laddering diversifies strike dates. It reduces, but does not eliminate, reliance on any single contract. It does not diversify the market, because every contract in the ladder references the same index. And a shareholder owns a fund whose net asset value is marked every day, and may sell well before any contract reaches its maturity. |
| Reinvestment | Early redemptions in rising markets return capital that has to be redeployed, potentially into contracts struck on different terms. |
| Counterparty and derivatives | Swap based exposure carries the creditworthiness of the swap counterparty, and a fund could lose money if that counterparty fails to meet its obligations. Derivatives also add valuation, correlation, and liquidity risk. |
| Distribution character | Gains earned through a swap are generally not net investment income for accounting purposes, so distributions may be classified largely as return of capital. Return of capital reduces net asset value. A distribution rate is not a total return, and the two should be compared, not conflated. |
Distribution rate vs total return, worked
A hypothetical fund, one year, per 100 of starting value. Same 13 distribution in both cases; only the total return differs.
| Case | Total return | Distributed | End value |
| Distributions exceed return | 4 | 13 | 91 |
| Return exceeds distributions | 16 | 13 | 103 |
Hypothetical figures for illustration only, not any fund’s results. Both distributions could carry the same return of capital label for accounting purposes. The label alone does not say which case occurred; the comparison of total return against the amount distributed does.
Why the label appears
Gains earned through a total return swap are generally not classified as net investment income, so there is little accounting income for distributions to be charged against and they are labeled return of capital. That is an accounting and tax characterization. It can coexist with a rising net asset value or a falling one, which is exactly why the label cannot substitute for the total return comparison.
Structures
Not every autocallable product is the same structure
Products under this name differ on two independent axes: what they reference, and how the downside works. What causes a loss is different in each combination, so a screen sorted by distribution rate is not comparing like with like.
Index with volatility target
References a rules based index constructed on a broad equity benchmark, with a volatility target and an annual deduction. The barrier risk here is as much index construction risk as market risk.
Worst of basket
References several indices or ETFs, and the payoff follows the weakest of them. The barrier risk is largely correlation risk. Diversification in the basket works against the holder, not for them.
Single stock
References one company. The barrier risk is company specific gap risk, which no index level volatility mechanism smooths. Coupons on this design are often quoted at wider levels, reflecting that concentration.
Cutting across all three: barrier vs buffer on the downside, memory vs no memory on the coupon, and income vs accumulation designs where coupons compound inside the product rather than being distributed. Headline coupons can look similar across every combination even though what produces a loss is different in each.
Before You Allocate
Questions worth asking
Not the marketing shorthand, the index named in the prospectus. Then check whether that index is the equity benchmark itself or a constructed index built on it. If it is constructed, find the volatility target, the exposure bounds, the deduction factor, the financing rate, and the reset frequency.
Contingent coupons accrue only when the reference holds the coupon barrier on an observation date. A memory feature allows missed coupons to be recovered later if conditions improve. Without one, each observation stands alone and a missed coupon is gone. The two designs are not comparable on headline rate alone.
A barrier is a cliff: finish below it and the full decline from the initial level passes through. A buffer absorbs the first tranche of decline, and only the shortfall beyond it passes through, often multiplied by a gearing factor. Same headline number, entirely different loss shape, and the buffer is paid for through a lower coupon.
It multiplies losses beyond a buffer. With a 50% buffer and a 200% gearing factor, an index finishing at 49% of its strike produces a 1% shortfall, times 200%, so a 2% reduction in settlement value. Each further point of decline costs two points of value, capped at the actual decline of the index.
Depth and observation method both matter. A level observed only at final maturity behaves very differently from one monitored continuously. And the depth is measured against the reference index, so a stated cushion does not translate one for one into a cushion on the underlying equity benchmark.
It reduces reliance on any single strike date, observation date, or maturity. That is a real improvement over holding one contract. It does not diversify the market, because every contract references the same index. Ladder depth is a timing diversifier, not a risk eliminator.
Gains earned through a total return swap are generally not classified as net investment income, so there is little accounting income for distributions to be charged against and they are labeled return of capital. That label is an accounting and tax characterization, and by itself it does not tell you whether capital was eroded. The meaningful comparison is a fund total return against what it actually distributed. Where distributions exceed total return, net asset value declines.
Two conditions work against it, for mechanical reasons. A strong sustained rally, because participation stops at the coupon and early calls return capital at par. And a gradual decline in which volatility never rises enough to reduce exposure before the downside levels are reached.
Vocabulary
Same term, six names
Issuers name the same thresholds differently, which makes two offering documents harder to compare than they should be. The translations first, then a working glossary.
| Coupon barrier | Also appears as interest barrier, contingent coupon barrier, or income barrier. |
| Risk barrier | Also appears as maturity barrier, protection barrier, protection level, principal barrier, or knock in level. |
| Autocall level | Also appears as call barrier, call level, autocall barrier, or redemption trigger. |
| Non call period | Also appears as lockout period or no call period. |
| Contingent coupon with memory | Also appears as Phoenix coupon or snowball feature. |
Autocallable. A contract that pays contingent coupons while a reference index holds a set level, redeems early at par if the index recovers to its autocall level, and passes losses to principal if the index finishes below its downside level at maturity.
Reference index. The index the levels are measured against. Often a constructed index, not the benchmark itself.
Strike level. The reference index level fixed on the day a contract is struck; every threshold is a percentage of it.
Observation date. A scheduled date on which the index is checked against the coupon barrier and, outside the non call period, the autocall level.
Contingent coupon. A coupon paid only when the reference index satisfies the coupon barrier on an observation date.
Memory feature. Missed coupons accrue and can be paid later if the index recovers above the coupon barrier.
Non call period. The initial span during which a contract cannot be autocalled.
Tenor. The maximum life of a contract from strike date to final maturity.
Risk barrier. A cliff style downside level; finishing below it passes the full decline through, one for one.
Risk buffer. A downside mechanic absorbing an initial tranche of decline; only the shortfall beyond it reduces settlement value.
Gearing factor. A multiplier on losses beyond a buffer, capped so the loss does not exceed the actual decline of the index.
Worst of. A structure referencing several indices or assets, where the payoff follows the weakest performer.
Volatility target. A formula setting the constructed index’s exposure to the benchmark; higher measured volatility, lower exposure.
Deduction factor. A fixed annual amount subtracted from a constructed index, accruing daily. Also called a decrement.
Financing cost. A rate charged on the constructed index’s exposure, usually a benchmark rate plus a spread.
Laddering. Holding many contracts with staggered strike dates so observations and maturities spread across the calendar.
Vintage. A single contract within a ladder, identified by its strike date.
Mark to market. Daily valuation at current prices, so fund value fluctuates even between observation dates.
Return of capital. An accounting and tax characterization of a distribution not charged against net investment income. The label alone does not indicate whether capital was eroded.
Distribution rate. An annualized rate implied by a fund’s most recent distribution. Not a total return; the two should be compared, not conflated.
The terms are the product. Read them.
REX offers autocallable ETFs built on these mechanics. Fund terms, data, and prospectuses are on the suite page.
Further Reading
Outside analysis of this category
Independent analysis of autocallable ETFs is worth reading, including work that is critical of the volatility target index design used across much of the category. Eric Barthe examines the reference index construction, the laddered exposure, and the distribution characterization in detail in “Autocallable ETFs: What Are You Actually Invested In?” on LinkedIn.
Links to third party content are provided for information only. REX Financial is not affiliated with the author, does not control third party content, and does not endorse, adopt, or verify its analysis or conclusions.
Important Information
This material is for educational and informational purposes only. It is not investment advice, not a recommendation, and not an offer to buy or sell any security. It does not take into account the objectives, financial situation, or needs of any investor. Descriptions of autocallable structures on this page are general and simplified; the terms of any specific product are governed by that product’s offering documents, which should be read carefully before investing.
All illustrations, worked examples, index paths, levels, coupon outcomes, and settlement values on this page are hypothetical and are provided to show mechanics only. They are not projections, backtests, or simulations, they do not describe the terms of any specific product, and they do not represent the past or future performance of any fund or index. Past performance is not indicative of future results.
Autocallable structures involve significant risks, including the risk of substantial loss of principal, up to and including a total loss. Contingent coupons are not guaranteed and are not paid for observations on which a reference index is below its coupon barrier; in unfavorable market environments an investor may receive little or no income. Where a reference index finishes below a contract’s risk barrier at final maturity, principal participates in the decline of the reference index; where a contract uses a risk buffer, losses beyond the buffer may be multiplied by a gearing factor, capped at the actual decline of the reference index. Constructed reference indices may apply volatility targets, leverage, deductions, and financing costs that cause them to behave differently from, and underperform, the equity benchmarks on which they are built; leverage within a reference index increases losses as well as gains. Funds that obtain exposure through swap agreements are subject to counterparty risk, and derivatives involve valuation, correlation, liquidity, and leverage risks. Distributions from such funds are not guaranteed, may be reduced or eliminated, and may include return of capital, which reduces a fund’s net asset value over time. A distribution rate does not represent a fund’s total return.
Investors should carefully consider the investment objectives, risks, charges, and expenses of any fund before investing. Fund specific risks, terms, and standardized information are set out in each fund’s prospectus and summary prospectus, available at rexshares.com. Please read the prospectus carefully before investing. Investing involves risk, including the possible loss of principal. Complex products of this type are not suitable for all investors.
