The Vicious Cycle No One Models Adequately:

Medicaid cuts → increased uninsured populations → hospitals absorb uncompensated care costs → hospitals raise prices to private insurers (cost-shifting) → insurers increase premiums/deductibles → employer-sponsored insurance drops → more uninsured → worse hospital finances → rural hospital closures → communities lose access AND largest employer → economic multiplier effects exceeding $60 million per facility

1. Speed That Enables Entirely New Capabilities

• Scenarios run in seconds instead of hours because compartmental models track population groups rather than individual people—achieving speedups of tens to hundreds, even thousands of times faster than microsimulation
• Enables previously impossible analysis: the increased speed allows for a large exploratory design, comprehensive probabilistic uncertainty quantification, multi-objective policy optimization, rapid legislative support
• Aggregates strategically: groups individuals by policy-relevant characteristics while preserving essential heterogeneity
• AI-powered calibration: simulation-based machine learning rapidly tunes models to real-world data with full uncertainty characterization

Why it matters: States can evaluate dozens of Medicaid contingency plans in an afternoon, find optimal interventions, and quantify risks

2. Integration That Captures the Full Story

• Explicit closed-loop coupling: Coverage ↔ Provider Capacity ↔ Access ↔ Health ↔ Costs ↔ Premiums solved simultaneously
• Cascading effects modeled: Medicaid cuts → hospital closures → cost-shifting → premium increases → employer coverage drops → more uninsured
• Disease dynamics evolve based on access to preventive care and treatment
• Provider supply responds endogenously to financial constraints, reimbursement rates, patient volume
• Cost-shifting emerges naturally from provider financial models, not imposed through accounting adjustments

Why it matters: Policy consequences ripple across the entire system—HERMES reveals cascades before they happen

3. Modularity That Fits Your Needs and Budget

• Three standalone modules: Coverage Dynamics, Health Dynamics, Provider Dynamics
• Fund one module ($400K-$750K, 12 months) or integrate all three ($1.3M-$1.8M, 42-54 months)
• Each module delivers immediate value; integration multiplies impact through feedback loops
• State-specific customization: $75K-$250K for calibration to local data and policy scenarios

Why it matters: Start where priorities are highest; expand as needs and funding allow; phased development de-risks investment

HERMES development is actively underway. Core modeling components are being constructed and tested against data from multiple national sources, including the Medical Expenditure Panel Survey (MEPS), Current Population Survey (CPS), National Health Interview Survey (NHIS), NHANES, SIPP, and National Health Expenditure accounts. This multi-source data foundation is essential for grounding the model's behavioral and financial parameters in real-world evidence.

Preliminary computational results are validating the core modeling approach, and manuscript preparation documenting methodological contributions is underway. State-level scenario analysis is in progress.

What the investment supports: Active development of a working prototype is not the same as a policy-deployable system. The investment phases described below cover the expert validation, sensitivity analysis, behavioral calibration, and documentation required to ensure that HERMES outputs are defensible under peer review and in the policy arena. Each module requires rigorous human expert review of model structure, parameter plausibility, and scenario interpretation—the intellectual labor that converts a computational framework into a trusted analytical tool.

Compartmental Modeling Foundation

Adapts successful epidemiological SEIR framework to health policy: individuals aggregate into strategic groups (coverage types, disease states, geographic regions, demographic categories). Achieves significant computational speedup versus microsimulation through differential equation systems while maintaining policy-relevant accuracy.

Cutting-Edge Calibration & Validation

• BayesFlow: Simulation-based inference using neural density estimation—handles complex high-dimensional parameter spaces efficiently
• Robust Decision Making: Identifies policies performing well across uncertain futures
• Multi-objective optimization: Pareto frontiers balancing coverage, costs, access, health outcomes, equity
• Validation: Component testing, counterfactual validation, cross-model comparison, expert elicitation
• Open and transparent: Full source code, documentation, replication materials

What these investments cover: These figures reflect the expert validation, behavioral calibration, sensitivity analysis, and quality-assurance processes required to make HERMES outputs defensible under peer review and in the policy arena. Active development is underway; each phase funds the rigorous human expert work needed to convert working model components into a trusted, deployable analytical tool — not merely computational construction.

The Founding Partner fee compensates for development work already invested in HERMES and grants early access to the framework before full public deployment. This is an access and partnership fee — separate from, and complementary to, the phase development investments below.

State Customization Add-Ons:

• Basic ($75K-$100K): State data calibration, standard scenarios
• Comprehensive ($150K-$250K): Custom policy scenarios, ongoing technical support, training