debt_leverage_profile
1.65x Total Debt / Equity (Elevated leverage)
High
SEC-XBRL
interest_rate_sensitivity
~$15-20M annual interest expense increase per 100bps rise; ~$1.4B variable-rate debt exposure makes leverage ratio sensitive to rate hikes
Inferred
Agent_Inference
geopolitical_supply_exposure
High intensity; OPEC+ supply decisions and Russia-Ukraine conflict drive price volatility.
Medium
GICS-commodity-overlay-v1
supply_chain_dependency
Cushing, Oklahoma crude hub and Permian Basin pipeline corridors; disruption risk concentrated in Mid-Continent and Gulf Coast logistics nodes
Inferred
Agent_Inference
international_expansion_readiness
Delek US operates almost entirely domestically; negligible sovereign currency devaluation exposure across international revenue markets
Inferred
Agent_Inference
geographic_footprint
Essentially 100% US-domestic operations (Texas, Tennessee, Arkansas, New Mexico); no material foreign-currency revenue exposure
Inferred
Agent_Inference
commodity_exposure_profile
High intensity; commodities: Crude Oil, Natural Gas, Coal, Refined Petroleum Products, Uranium, Steel (equipment); geopolitical: OPEC+ supply decisions and Russia-Ukraine conflict drive price volatility.
Medium
GICS-commodity-overlay-v1
vendor_lock_dependency_score
Pipeline and crude suppliers (Plains All American, Magellan) provide non-substitutable midstream access; single-source crude delivery risk exceeds 30% in certain refinery catchments
Inferred
Agent_Inference
business_model_type_primary
Minimal operational disruption; refining and retail operations are largely on-premise/industrial-control-system-based with limited cloud dependency
Inferred
Agent_Inference
business_model_type_secondary
Back-office ERP and trading analytics may have cloud exposure, but core refining operations would continue uninterrupted within 30 days
Inferred
Agent_Inference
switching_cost_profile
Low API coupling risk; operations are physical-asset-intensive (refineries, pipelines, retail fuel stations) with minimal SaaS or API-dependent workflows
Inferred
Agent_Inference
howey_test_risk_index
Fails Howey Test; revenue model is commodity product sales (refined fuels), not an investment contract — no securities classification risk
Inferred
Agent_Inference
regulatory_burden_tier
High
Medium
GICS-regulatory-overlay-v1
data_sovereignty_risk
Low GDPR exposure (no EU operations); moderate CCPA risk via retail fuel loyalty programs collecting California consumer data
Inferred
Agent_Inference
antitrust_exposure_flag
Moderate; regional refining market concentration in Mid-Continent and retail fuel pricing in southern US markets warrants periodic FTC scrutiny
Inferred
Agent_Inference
regulatory_exposure_profile
High burden; regimes: EPA, FERC, DOE, CFTC, OSHA, SEC; Accelerating emissions mandates and methane rules threaten capex economics.
Medium
GICS-regulatory-overlay-v1
revenue_model_type
~95% transactional (spot refined product sales, rack pricing); <5% recurring via wholesale supply agreements and retail fuel contracts
Inferred
Agent_Inference
monetization_vector
Commodity margin capture — crack spread between crude input cost and refined product (gasoline, diesel, jet fuel) wholesale and retail sales price
Inferred
Agent_Inference
pricing_architecture
Crack-spread-driven commodity pricing; no proprietary pricing power — margins collapse when crude-product differentials compress below $8/bbl
Inferred
Agent_Inference
pricing_power_rating
Low; refined fuel is fungible commodity priced off NYMEX benchmarks; Delek is price-taker in wholesale markets
Inferred
Agent_Inference
target_gross_margin_bracket
Refining gross margin typically 6-12% of revenue; retail segment adds ~2-4% incremental margin; consolidated gross margin ~4-10% cyclically
Inferred
Agent_Inference
churn_vulnerability_index
No free-rider problem; fuel is purchased transactionally with no subscription structure, but wholesale contract customers exhibit moderate stickiness
Inferred
Agent_Inference
headcount_cost_structure
Revenue growth is largely capital- and throughput-linear, not headcount-linear; doubling refinery throughput requires <20% headcount increase
Inferred
Agent_Inference
marginal_cost_of_growth
High capital marginal cost (refinery capacity expansions $200M+), but low incremental labor cost; growth is capital-intensive, not people-intensive
Inferred
Agent_Inference
franchise_compliance_risk
Low franchise compliance risk; retail sites are mostly company-operated or independent dealers under supply agreements, not traditional franchise model
Inferred
Agent_Inference
customer_acquisition_metric
At 10x scale, wholesale fuel distribution economics improve via fixed-cost leverage, but crude sourcing and pipeline access become binding constraints
Inferred
Agent_Inference
network_effect_present
No meaningful network effects; refining and fuel retail are point-of-sale commodity businesses with no cross-user value creation
Inferred
Agent_Inference
asset_efficiency_ratio
0.6% Return on Assets (Excellent)
High
SEC-XBRL
recession_resistance_tier
Tier 3 moderate resilience; fuel demand is relatively inelastic short-term but crack spreads compress during recessions as industrial demand falls
Inferred
Agent_Inference
customer_segment_primary
Wholesale fuel purchasers (distributors, airlines, trucking fleets, commercial accounts); top 10 wholesale customers likely represent 30-40% of refinery output
Inferred
Agent_Inference
customer_segment_secondary
Retail consumers via ~250 DKS convenience/fuel stations in southern US; high fragmentation, no single retail customer represents material concentration
Inferred
Agent_Inference
characteristic_occupations
["11-0000 Management Occupations", "13-0000 Business and Financial Operations Occupations", "15-0000 Computer and Mathematical Occupations", "17-0000 Architecture and Engineering Occupations", "19-0000 Life, Physical, and Social Science Occupations", "23-0000 Legal Occupations", "41-0000 Sales and Related Occupations", "43-0000 Office and Administrative Support Occupations", "47-0000 Construction and Extraction Occupations", "49-0000 Installation, Maintenance, and Repair Occupations", "53-0000 Transportation and Material Moving Occupations"]
High
SOC-2018/GICS-overlay
agent_automatable_labor_share
0.33 (HIL — ~33% of characteristic roles agent-automatable)
Medium
SOC-2018 + agentic-exposure-v1
capital_expenditure_profile
4.9% CapEx / Revenue (Low-CapEx Asset-Light)
High
SEC-XBRL
sec_cik
0001694426
High
SEC-EDGAR