> For the complete documentation index, see [llms.txt](https://docs.basis.pro/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.basis.pro/economics-and-rewards/why-drr-rises-and-falls.md).

# Why DRR Rises and Falls: A Structural Explanation

{% hint style="info" %}
Operator and jurisdiction: BASIS is operated by BASIS DIGITAL INFRASTRUCTURE LTD, a Seychelles IBC (LEI: [254900IX2F2KCWNSSS64](https://lei.bloomberg.com/leis/view/254900IX2F2KCWNSSS64)).

Research Partner: Base58 Labs contributes execution research, systems modeling, and risk design.
{% endhint %}

The [Glossary](/reference/glossary.md) defines DRR as follows:

> "DRR stands for Daily Reward Rate. It is an estimated daily yield rate on deployed capital, expressed as a percentage. DRR reflects recent strategy performance and is not guaranteed."

DRR is cyclical because it is market-derived. It is not a fixed coupon. In the [Lock-up Economics](/economics-and-rewards/lock-up-economics.md) model, yield is expressed as:

`Yield ~= (1 - b) x C x r`

where `b` is the required liquidity buffer fraction, `C` is total capital, and `r` is the realized strategy yield rate on deployed capital. The variable `r` is externally determined by market conditions, not by BASIS's operational efficiency alone.

When market structure offers stronger carry, wider spreads, higher utilization, or more positive funding, DRR can rise. When those conditions compress, DRR can fall. This is a repeating market cycle, not a one-way decline.

***

## Up cycle: why DRR can expand

DRR tends to rise during strong demand regimes. These are periods when speculative leverage increases, cross-venue price dispersion widens, DeFi utilization rises, and funding rates turn more positive.

The [Funding Rate Dynamics](/strategies/funding-rate-dynamics.md) page defines the funding mechanism directly:

> "Funding Rate > 0 -> Longs pay Shorts. Funding Rate < 0 -> Shorts pay Longs."

For delta-neutral funding capture, positive funding is structurally favorable because the short perpetual leg receives funding while the spot or staked asset leg preserves directional neutrality.

BASIS documents the following historical 8-hour funding ranges as illustrative historical observations, not forward expectations:

| Asset | Typical positive range | Extreme positive range | Typical negative range | Extreme negative range |
| ----- | ---------------------: | ---------------------: | ---------------------: | ---------------------: |
| BTC   |       0.005% to 0.030% |                >0.100% |     -0.005% to -0.020% |                -0.075% |
| ETH   |       0.005% to 0.035% |                >0.150% |     -0.005% to -0.025% |                -0.090% |
| SOL   |       0.010% to 0.050% |                >0.200% |     -0.010% to -0.030% |                -0.100% |

### Why PAXG is not in this table

This table is quoted directly from BASIS's Funding Rate Dynamics page, which covers BTC, ETH, and SOL only. PAXG uses a different yield structure and is intentionally excluded here.

| Factor                | BTC / ETH / SOL perpetuals           | PAXG perpetual                                                                       |
| --------------------- | ------------------------------------ | ------------------------------------------------------------------------------------ |
| Funding history depth | Multi-year track record              | Approximately since March 2025, a shorter sample                                     |
| 24h perpetual volume  | Multi-billion dollars per asset      | Roughly $90 to $100 million, about two orders of magnitude smaller                   |
| Observed funding band | Wide, matches the ranges shown above | Narrower, has stayed inside a tighter band than BTC's typical range                  |
| Yield documentation   | Single funding-rate mechanism        | Multiple channels, see Golden BASIS spot-perp module and PAXG Real-World Asset Yield |

{% hint style="info" %}
Because of these differences, BASIS documents PAXG yield separately rather than folding it into the BTC, ETH, SOL funding table above.
{% endhint %}

The same page notes that a midpoint funding rate of 0.015% per 8 hours annualizes to approximately 16.4% simple gross before costs. It also notes that BTC funding above 0.100% per 8 hours annualizes to more than 109.5% simple gross before costs. These figures illustrate the scale that funding-based carry can reach when speculative demand is strong. They are historical reference points, not projections.

In up cycles, more than one BASIS yield source can expand at the same time:

* Spatial Arbitrage (BQAE)
* Funding Rate Capture (Delta-Neutral Basis Trade)
* Blue-Chip DeFi Lending + Liquid Staking
* PAXG Real-World Asset Yield

These sources are diversified and not perfectly correlated, but strong risk appetite in the broader market can lift more than one of them at the same time.

***

## Down cycle: why DRR compresses

DRR compresses when expected edge compresses across the platform's yield sources.

Recent market conditions illustrate this side of the cycle. BTC perpetual funding on major venues has recently traded near flat, occasionally printing negative, well below the typical positive BTC range of 0.005% to 0.030% per 8 hours documented above. Sentiment indicators such as the Crypto Fear & Greed Index have been concentrated in Extreme Fear territory, consistent with reduced speculative leverage. When leveraged long demand declines, positive funding falls toward zero or turns negative.

Realized volatility does not need to collapse for this to happen. Price can continue moving while leverage-driven demand for perpetual longs stays weak, which compresses funding-based carry specifically, separate from general price volatility. This is a demand-side and leverage-side compression, not simply "the market went quiet."

This is the same cycle described in the funding mechanics: when funding becomes less positive, the expected carry available to a delta-neutral, short-perpetual structure declines.

***

## The floor mechanism: why the reward rate does not go negative for users

This is the central mechanical point of this page and it should be read independently of the disclosures that follow it.

BASIS's documented control stack is specifically designed so that adverse funding conditions are absorbed by the system through monitoring, exposure reduction, and pausing, rather than being passed through as a negative rate against a user's reward balance. The design objective, and the documented outcome of that design, is that DRR compresses toward a protective floor near 0% under adverse conditions. It does not settle into negative territory and it does not debit previously credited stToken units.

This follows directly from four layers of documented control.

### 1. EV gating

The [Arbitrage Economics: Edge vs Cost](/research-library-deep-dives/arb-economics.md) page defines:

`EV = E[Edge] - sum(C_trade) - R_premium`

The documentation states:

> "BASIS initiates a trade only when EV > 0 under conservative parameters."

A candidate slice that is not EV positive is rejected rather than executed. This rule applies across all four yield sources. It stops new deployment before an unfavorable trade is entered, rather than entering it and hoping conditions improve.

### 2. The funding response table

The [Funding Rate Dynamics](/strategies/funding-rate-dynamics.md) page documents BASIS's automatic response to negative funding:

| Funding Rate (8h)  | BASIS Response                                                  |
| ------------------ | --------------------------------------------------------------- |
| > 0%               | Normal operation                                                |
| 0% to -0.005%      | Elevated monitoring, no immediate reallocation                  |
| -0.005% to -0.020% | Exposure reduction on the affected asset, potentially up to 50% |
| < -0.020%          | Strategy pause on the affected asset until conditions normalize |

This table exists precisely to prevent adverse funding from persisting at scale. Mildly negative funding triggers monitoring. More adverse funding triggers exposure reduction on the affected asset. Deeply adverse funding triggers a pause on that asset until conditions normalize. The documentation states the intent directly:

> "The objective is not continuous deployment at any cost. BASIS prioritizes deterministic execution, math-constrained risk management, and capital preservation when expected carry becomes structurally unfavorable."

The response table works progressively, not instantly. Its purpose is to compress the size and duration of any negative-funding exposure toward zero as quickly as the documented thresholds allow, rather than to let a module remain fully deployed against a persistently negative rate. Combined with EV gating on new slices, this is the mechanism that keeps the platform's realized reward rate from settling into sustained negative territory.

### 3. BSCB and DMM

The [BSCB Sentinel Circuit Breaker](/risk-safety-and-asset-protection/bscb.md) is:

> "An automated, rule-based safety mechanism that halts or reduces execution activity when predefined risk thresholds are breached."

Triggers include slippage anomalies, venue API failures, margin health deterioration, and pricing reference instability.

[DMM, Defensive Maintenance Mode](/risk-safety-and-asset-protection/dmm.md), is the most severe operational state. On activation, it applies a full halt to automated trading, staking actions, swaps, and withdrawal processing until human review and a formal Root Cause Analysis are complete. It is documented as a safety mechanism, not a discretionary performance tool.

Together, EV gating, the funding response table, BSCB, and DMM enforce one principle: do not continue deploying capital at any cost when expected carry or operating conditions become structurally unfavorable.

### 4. BIVB and stToken quantity preservation

The [BIVB & stTokens](/whitepaper/bivb-and-sttokens.md) design converts native assets into staking tokens at a strict 1:1 quantity ratio. Rewards accrue in real time in the same stToken. The documented worked example shows a deposit of 1 BTC growing to 1.10 BTC in quantity terms over time.

This reward-accrual mechanism is quantity-additive. It is not designed to subtract previously credited stToken units because a module becomes temporarily ineligible. When a module pauses, the documented outcome is that new rewards from that module stop accruing for that window. It is not a mechanism that reaches back and reduces stToken already credited to a user's Staking Wallet.

The [FAQ](/faq/faq.md) confirms this pause behavior directly:

> "Can rewards pause? Yes. Rewards may pause when protective controls are triggered, when diagnostic monitoring is active for root cause analysis, or when strategy modules become temporarily ineligible due to slippage, venue incidents, or stress conditions. These pauses are a safety feature."

***

## This is a different claim from yield-level or principal-value guarantees

The floor statement above is a mechanism-level statement about the reward-transmission path. It states that the documented control stack is designed to compress adverse funding toward a near-0% floor rather than a negative rate, and that already-credited stToken balances are not reduced because a module becomes temporarily ineligible.

That is a separate claim from saying how large a positive DRR will be, or that it will stay above any specific percentage.

BASIS does not guarantee the size or level of positive yield. Actual DRR depends on realized strategy performance, prevailing market conditions, execution costs, available liquidity, and the number of eligible opportunities at any given time.

BASIS also does not guarantee the USD-equivalent market value of a user's principal. BTC, ETH, SOL, and PAXG can decline in fiat terms even while their staked quantity is preserved or growing. The BIVB and stToken design preserves native-asset quantity mechanics. It does not fix or protect the fiat-denominated value of that quantity, which is a separate, market-driven variable.

These two statements are not in tension. A reward rate that is mechanically floored near zero is a distinct fact from a promise about how large future rewards will be, or about the fiat value of a user's holdings.

***

## Diversification across yield sources

BASIS uses four yield sources that are diversified and not perfectly correlated:

| Yield source                            | Primary driver                                            |
| --------------------------------------- | --------------------------------------------------------- |
| Spatial Arbitrage (BQAE)                | Cross-venue dispersion and executable price differences   |
| Funding Rate Capture                    | Perpetual funding and basis conditions                    |
| Blue-Chip DeFi Lending + Liquid Staking | Utilization, staking rates, protocol-level opportunity    |
| PAXG Real-World Asset Yield             | Gold-linked asset deployment and related market structure |

This diversification supports floor-like behavior in the blended DRR. If one module pauses because it is not EV positive, other modules may remain eligible. If conditions compress broadly across all four sources at once, the blended DRR moves toward the protective near-0% state described above rather than toward a negative number, because each individual module is subject to the same EV gating and response controls.

***

## Illustrative range, not guaranteed, not a forecast

The table below is illustrative. It describes documented mechanism and historical range inputs. It is not a forecast, a promised rate, or a cap.

| Reference state                     | Documented basis                                                           | Mechanical meaning                                                                                    |
| ----------------------------------- | -------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------- |
| Lower bound: protective pause state | Affected source fails EV gating or funding falls below -0.020% per 8 hours | Affected source contributes near 0% for that window. No debit is applied to already-credited stToken. |
| Mid-cycle positive reference        | Funding at approximately 0.015% per 8 hours                                | Approximately 16.4% simple gross annualized before costs                                              |
| High-cycle positive reference       | BTC funding above the documented extreme of 0.100% per 8 hours             | More than 109.5% simple gross annualized before costs                                                 |

The lower bound in this table is the pause mechanism itself, not a guaranteed positive percentage above zero. The upper reference is drawn from BASIS's own documented historical extreme funding data. It is not a cap and not a forward projection.

***

## Industry-wide funding compression

Yield compression has also appeared recently in other public delta-neutral funding-capture products in the broader market. This is directional evidence that funding-driven yield compression is a market-cycle effect across the category, not specific to BASIS.

Historically, funding-driven yield has re-expanded when speculative leverage returns, positive funding normalizes, and executable spreads widen again. BASIS's control stack is designed to participate when EV is positive and to reduce or pause when it is not, in either direction of the cycle.

***

## Closing disclosure

Historical performance figures on this page are illustrative and reflect past conditions only. BASIS targets yield through market-neutral execution and deterministic risk controls. Actual results depend on realized strategy performance. BASIS does not guarantee any specific yield level or the USD-equivalent value of a user's principal.

Third-party exchange risk, counterparty risk, smart contract risk, regulatory risk, and other catastrophic tail risks are addressed in the Risk Disclosure (Master) page and DMM documentation, and are not restated here.

None of the disclosures above change the mechanism-level statement made on this page: under BASIS's documented controls for ordinary adverse-funding conditions, EV gating rejects unfavorable new trades, the funding response table compresses and pauses adverse exposure, BSCB and DMM provide additional automated and manual safeguards, and the BIVB/stToken design does not subtract previously credited reward units. Together, these are the reasons the platform's reward-transmission path is designed to compress toward a near-0% floor rather than to transmit a negative rate to users.

***

## See Also

* [Yield Sources: Where Returns Come From](/economics-and-rewards/yield-sources.md)
* [Lock-up Economics (Capital Efficiency)](/economics-and-rewards/lock-up-economics.md)
* [Funding Rate Dynamics](/strategies/funding-rate-dynamics.md)
* [BSCB: Sentinel Circuit Breaker](/risk-safety-and-asset-protection/bscb.md)
* [DMM: Defensive Maintenance Mode](/risk-safety-and-asset-protection/dmm.md)
* [Risk Disclosure (Master)](/risk-safety-and-asset-protection/risk-disclosure.md)