Authors: Rogers, Adam A, Smyth, Damian,J*

Principals, Swanson Reed

Strategic Overview

The trajectory of the United States’ innovation economy stands at an interesting juncture in the mid-2020s. Following the disruptive shock of the COVID-19 pandemic, the relationships between capital investment, intellectual property creation, and economic expansion have shifted. This paper provides an examination of the current state of U.S. innovation, utilizing the methodology of the Swanson Reed inventionINDEX as the lens for analysis.

The inventionINDEX was developed to explore this question: How can we attempt to differentiate between an economy that is growing because it is adding more people and capital (extensive growth) and an economy that is growing because it is inventing new ways to create value (intensive growth)?

Unlike static rankings that tally the number of patents or the total dollars spent on R&D, the inventionINDEX is a measure of elasticity. It asks: For every percentage point of GDP growth, how much new intellectual property is being generated? A high index implies that economic gains are correlating with the creation of new assets (patents). A low index implies that the economy is relying on its existing stock of ideas without replenishing them at the same rate.

Central to this investigation is the 20-year baseline calculation—a temporal framework used to separate secular innovation trends from transient economic noise. By anchoring innovation metrics to the statutory lifespan of utility patents, we attempt to derive a baseline assessment of a jurisdiction’s inventive output.

The paper contrasts the innovation landscape of the pre-pandemic era (2015–2019) with the post-COVID recovery period (2020–2025), noting a “consolidation phase” characterized by declining patent applications. We also explore the limitations of this methodology, particularly regarding data noise and litigation-driven patenting.

Finally, the paper outlines a justification for private-sector advocacy—specifically the Patent Funding Initiative—as a mechanism to bridge the “Valley of Death” for states consistently underperforming against their innovation benchmarks.

The Theoretical Foundations of Innovation Measurement

The Challenge of the Intangible Economy

In the modern global economy, the production of intellectual property (IP) protected by patents is a major driver of wealth creation. However, measuring “innovation” is notoriously difficult. Traditional macroeconomic metrics, primarily Gross Domestic Product (GDP), were designed for an industrial age. In the information age, these traditional metrics often fail to capture the underlying health of an economy’s creative engine.

The core problem lies in the disconnect between expenditure and result. Standard metrics often rely on inputs, such as aggregate Research and Development (R&D) spending. While R&D spending indicates intent, a region may spend billions on research that yields little commercializable intellectual property.

The inventionINDEX Approach

The inventionINDEX addresses this by evaluating innovation output through a specific ratio: GDP growth versus patent production growth. This “innovation elasticity” provides a sentiment grade (A through F).

If a state’s GDP expands significantly while its patent production remains stagnant or declines, the index suggests that the growth is likely driven by consumption, population influx, or asset inflation. Conversely, if patent production growth outpaces GDP growth, it indicates that the economy is deepening its intellectual capital base.

The 20-Year Baseline: Anatomy of the Metric

The 20-year baseline calculation is fundamental to this specific innovation indexing approach, rooted in the lifecycle realities of intellectual property.

The Statutory Foundation: Alignment with Patent Lifecycles

The primary justification for a 20-year baseline is the statutory term of a U.S. utility patent. Under 35 U.S.C. 154(a)(2), a patent confers exclusivity for a period of 20 years from the date of filing. Consequently, an index measuring the economic impact of patents uses this lifecycle to gauge when older assets are entering the public domain.

The Replacement Rate Concept

A healthy innovation ecosystem must replace its expiring assets. Patents filed 20 years ago are currently expiring, removing the exclusivity capability of the original IP asset. The 20-year baseline serves as a “replacement rate” calculation. If a state’s current patent production growth lags behind the rate of patents filed 20 years prior, the region is effectively experiencing intellectual capital depreciation.

Filtering Cyclical Noise

Short-term metrics, such as those relying on 1-year to 5-year baselines, are heavily influenced by immediate economic conditions. The surge in digital patenting observed in 2020–2021 was an anomaly driven by the pandemic. A metric based solely on a 5-year baseline might interpret a subsequent decline in 2025 as a collapse, whereas a 20-year baseline contextualizes it as a normalization.

Table 1: Implications of the 20-Year Baseline

Methodological Limitations and Structural Vulnerabilities

While the inventionINDEX provides a specific lens for viewing economic elasticity, it is crucial to acknowledge its significant structural and methodological limitations. The model relies on assumptions that do not always hold true in complex regional economies.

The “Patent Troll” Distortion (Non-Practicing Entities)

A primary weakness of relying on raw patent volume elasticity is the inability of the index to distinguish between high-value, commercialized innovation and predatory patenting. The inventionINDEX treats all utility patents equally. Therefore, a state that serves as a legal haven for Non-Practicing Entities (NPEs)—commonly known as “patent trolls”—may exhibit artificially inflated patent production metrics.

When NPEs file or acquire broad, vague patents strictly for the purpose of litigation and rent-seeking rather than product development, the index registers this as “innovation growth.” This creates false positive signals, granting high sentiment grades to regions that are generating legal friction rather than actual economic value or technological advancement.

Regression Analysis and Statistical Noise

The mathematical foundation of the inventionINDEX relies on regression analysis comparing GDP growth to patent output, which introduces several statistical vulnerabilities:

• Endogeneity and Causation: The model assumes a directional relationship where patenting reflects underlying economic health. However, correlation does not equal causation. A spike in GDP might fund more patent filings, or more patent filings might eventually boost GDP, but the index struggles to isolate the exact causal mechanism, leading to potential misinterpretation of the data.
• Omitted Variable Bias: The elasticity ratio ignores critical local variables. Factors such as state-level regulatory overhauls, shifts in localized education quality, or sudden changes in foreign direct investment are omitted from the two-variable equation. This bias can severely skew the perceived “efficiency” of a state’s innovation engine.
• Heteroskedasticity: Comparing the elasticity of massive, highly diversified economies (like California or Texas) directly against smaller, concentrated economies (like Wyoming or Vermont) introduces high variance in the error terms. A single corporate relocation in a small state can cause wild, disproportionate swings in its index grade that do not accurately reflect long-term systemic health.

The Chronology of Disruption: Pre- and Post-COVID Innovation Dynamics

To understand the current readings of the inventionINDEX—which showed a “consolidation phase” in late 2025—we must look at the shift between pre- and post-pandemic environments.

The Pre-COVID Era (2015–2019): The “Optimized Efficiency” Phase
Prior to 2020, the U.S. innovation engine saw steady growth focused on software, consumer electronics, and AI. Supply chains were globally optimized, and cross-border collaboration was high. The ecosystem was defined by efficiency, where R&D was tightly coupled with immediate market needs.

The Pandemic Shock (2020–2022): Disruption and Divergence
The onset of COVID-19 triggered an asymmetric shock. While sectors like aerospace saw R&D drops, pharmaceutical and Information and Communications Technology (ICT) sectors surged. U.S. labor productivity spiked due to remote work adoption, creating a temporary divergence between shrinking GDPs (due to lockdowns) and robust high-tech output.

The Post-COVID Landscape (2023–2025): Consolidation and Cooling
By late 2025, the data points to an economy in transition. The stimulus of the pandemic era has faded.

• Decline in Patent Applications: After years of growth, U.S. patent applications dropped 9% in 2025, reaching the lowest level since 2019. This suggests that pandemic-era R&D did not translate into the expected volume of patentable outputs, or companies are shifting toward trade secrets due to cost pressures.
• Regional Divergence: The recovery has been uneven. Florida (July 2025 Score: 5.22%, Grade A+) and North Carolina (Nov 2025 Score: 1.83%, Grade A+) have outperformed baselines, likely benefiting from talent migration. Legacy hubs like Connecticut (Dec 2025 Score: 0.99%, Grade C-) show stagnation.

Table 2: Pre- vs. Post-COVID Innovation Metrics Comparison

The Policy Gap: Justifying the Patent Grant Lobby

When a state consistently scores poorly on the index, it may indicate a failure in the private market’s allocation of capital to early-stage R&D. This is often due to the “Valley of Death”—the gap between basic research and commercialization. Obtaining and maintaining an international patent family can cost upwards of $100,000, presenting a prohibitive entry barrier for startups.

The Swanson Reed Patent Funding Initiative

The Patent Funding Initiative proposes a federal grant program to directly subsidize this cost.

Key Mechanisms:

1. Targeted Grants: Providing up to $50,000 per international patent family in non-dilutive funding to offset filing costs.
2. Performance Conditionality: To mitigate the aforementioned risks of funding “patent trolls,” eligibility requires companies to exceed their two-year rolling average of patent output.
3. University Collaboration: Requires applicants to include at least one U.S. academic as a co-inventor to incentivize technology transfer from the university level to the private market.

Justification for Lobbying Assistance

During the current “consolidation phase” where patent filings are dropping, private capital is retreating due to higher interest rates. Government grants can act as a counter-cyclical stabilizer, preserving the R&D pipeline until the macroeconomic environment improves. For states operating below their 20-year baseline, standard broad-stroke economic policies often fail to reach early-stage innovators. The lobby aims to tailor these grant programs to the specific needs of at-risk jurisdictions.

Final Thoughts

The analysis of the post-COVID United States reveals an innovation ecosystem in a state of cooling and consolidation. The surge of 2020–2021 has dissipated, leaving a landscape marked by declining patent applications and regional disparities.

While the inventionINDEX has distinct statistical limitations and vulnerabilities—particularly regarding non-practicing entities and regression noise—it provides a specific metric for viewing the elasticity of patent production against economic growth. The data suggests that innovation output is currently contracting relative to historical highs. Addressing this gap requires acknowledging the changing capital environment and exploring targeted interventions, such as the Patent Funding Initiative, to support early-stage commercialization.