Asset Risk Framework

Our multi-layer risk management approach integrates external risk ratings with proprietary quantitative models. The objective is to assess the creditworthiness of DeFi lending markets by evaluating the probability of capital loss across technical, economic, and execution dimensions.

Birch Hill Ratings (BHR) quantify the probability of capital impairment in DeFi lending markets. Each rating synthesizes protocol-level security, liquidation execution feasibility, and simulated performance under tail risk scenarios to provide investors with traditional credit-style risk assessments.

Technical Risk

All assets and asset markets are evaluated using Exponential Finance's A-F risk rating system, which provides comprehensive technical coverage across:

• Blockchain Risk: Network security, validator concentration, upgrade mechanisms

• Protocol Risk: Code quality, audit history, economic design

• Asset Risk: Intrinsic value, market capitalization, centralization vectors

• Pool Risk: Collateral quality, liquidation mechanisms, oracle dependencies

Economic Risk

Collateral Value at Risk

The Collateral Value at Risk (CVaR) model determines optimal supply caps for each lending market by quantifying the maximum capital at risk during extreme liquidation events. This helps us understand the liquidity profile of an asset or market on a particular network, helping us refine ratings to network-specific tail risks.

The Supply Cap is a function of available DEX liquidity, adjusted for volatility (VaR) and liquidation economics.

Where:

• L_i (Liquidity) = The available depth of the collateral asset on DEXs at a specific slippage tier.

• VaR_i = (Value at Risk): The 99th percentile worst-case liquidation volume derived from historical volatility simulations

• LP_i (Liquidation Penalty): The incentive paid to liquidators. This represents the "cost" of exiting the position.

• α = (Safety Factor): A proprietary scalar value reducing our capacity usage to ensure safety buffers during liquidation.

Simulation & Modeling

Once we understand technical and economic risks, we measure performance over thousands of dynamic agent-based simulations to build scenario analysis and "What If?" reporting. We leverage classical Geometric Brownian Motion for predicting volatile asset paths, and Ornstein-Uhlenbeck methodology for predicting stable asset pairs. The results of our simulation & modeling inform both the Birch Hill Risk Engine as well as Birch Hill Risk Ratings.

Ratings Process

Step 1: Technical Foundation Ingest Exponential Finance risk ratings (A-F scale) covering blockchain, protocol, asset, and pool-level risk factors.

Step 2: Economic Adjustment Calculate market-specific CVaR using:

• Historical simulation (1-year lookback, daily returns)

• 99th percentile worst-case liquidation volume

• DEX liquidity mapping at various slippage tiers

• Correlation factors for related collateral markets

Step 3: Stress Testing Run Monte Carlo simulations using:

• Geometric Brownian Motion (volatile assets)

• Ornstein-Uhlenbeck (stable assets)

• Multi-scenario analysis across price shocks, liquidity crises, and correlation breakdowns

Step 4: Final Rating Assignment Synthesize technical scores, liquidity profiles, and simulation results into traditional credit ratings.

Birch Hill Rating Map

BHR	Exponential Fi Base	Liquidity Profile	Simulation Performance	Interpretation
AAA	A	Deep (α ≥ 0.75)	Stable across all scenarios	Exceptional quality, minimal tail risk
AA	A	Strong (α = 0.65-0.74)	Resilient in stress tests	High quality, low default probability
A	A / B+	Adequate (α = 0.60-0.64)	Moderate stress tolerance	Investment grade, acceptable volatility
BBB	B	Adequate (α = 0.60-0.64)	Passes baseline scenarios	Lower investment grade
BB	B / C+	Limited (α = 0.50-0.59)	Vulnerable in tail events	Speculative, elevated risk
B	C	Constrained (α < 0.50)	Significant stress exposure	Highly speculative
CCC	D/E/F	Any	Failed stress scenarios	Substantial credit risk

Rating Modifiers:

• Liquidity depth upgrades strong technical ratings (A → AA/AAA)

• Weak liquidity constrains ratings regardless of technical score

• Simulation failures cap maximum rating at CCC

• Correlation risk downgrades correlated portfolio concentrations by one notch

Continuous Compliance Monitoring

Automated screening against regulatory sanctions lists, stolen asset databases, and protocol security assessments ensures institutional-quality risk management.

Screening Checkpoints:

1. Market Onboarding: Pre-investment screening of all contract addresses

2. Capital Allocation: Transaction-level verification before execution

3. Daily Monitoring: Active borrower address screening

4. Monthly Rebalancing: Comprehensive re-screening before reallocation

Data Sources:

• OFAC SDN List, UN Security Council Sanctions, EU Financial Sanctions

• Chainalysis, Hypernative

• Known exploit addresses and protocol incident databases