Burlington, Vermont: Historical Economic Development and Industry Case Studies

To understand the application of complex federal and state tax incentives within a specific municipality, one must first analyze the region’s economic metamorphosis. Burlington, Vermont, provides a profound case study of industrial evolution, transitioning from a 19th-century resource extraction hub into a 21st-century nucleus of technology, life sciences, and advanced manufacturing.

The historical foundation of Burlington’s economy was built upon its strategic geographic location. Around 1800, the early village of Burlington Bay was populated by a handful of innkeepers, retail merchants, shipbuilders, and specialty workshops, evidenced by enduring architecture such as the Old Stone Store built in 1798. Following the opening of the Champlain Canal in 1823, Burlington was suddenly plugged into a vast North American trade network. Steamboats and canal barges poured into the harbor, and the city systematically expanded its shoreline using timber cribs filled with stone to make room for expansive lumberyards, warehouses, rail spurs, and wharves. By the late 1800s, Burlington had become a bustling lumbering and manufacturing center—at one point recognized as the third-largest lumber market in the world. The city was incorporated in 1865, and its Victorian-era prosperity left behind magnificent architecture designed by Ammi B. Young, H.H. Richardson, and McKim, Mead & White.

As the global lumber boom eventually subsided, Burlington’s waterfront and adjacent neighborhoods, such as the Maple-Kilburn area and Lakeside Avenue, pivoted toward diversified manufacturing. From before 1869 to the present, a wide range of companies conducted manufacturing in industries as diverse as cotton textiles, spool and bobbin production, Venetian blind crafting, marble and granite finishing, and small machinery. Concurrently, the University of Vermont (UVM) was laying the intellectual groundwork for the city’s future. The institution pioneered practical and engineering education, believed to be the first non-military institution to offer engineering courses, and admitted women to full membership in the scholarly society Phi Beta Kappa as early as 1875.

Today, this rich industrial history, combined with the dense intellectual capital generated by UVM’s recent designation as an R1 research university by the Carnegie Classification of Institutions of Higher Education, has cultivated an ecosystem highly conducive to generating Qualified Research Expenses (QREs). The following five case studies illustrate how distinct industries rooted in Burlington’s economic history engage in technical activities that satisfy the stringent requirements of United States Internal Revenue Code (IRC) § 41 and Vermont 32 V.S.A. § 5930ii.

Case Study 1: Aerospace and Defense Manufacturing

The defense manufacturing lineage in Burlington traces its origins directly back to the Lakeside Avenue mills. Before 1894, this area was fertile farmland on the banks of Lake Champlain, but it was soon transformed when the Draper Corporation of Massachusetts constructed a new mill to test a novel loom design and take advantage of favorable freight rates. This facility evolved into the Queen City Cotton Company, creating Vermont’s first factory housing. The critical turning point occurred in the early 1940s when the mill buildings were seized by the United States Government from the E.B. & A.C. Whiting Tulatex Corporation and reallocated to Bell Aircraft for wartime weapons production, specifically the manufacturing of airplane turrets. Following World War II, General Electric purchased the facility to manufacture weapons systems, operating it until the late 1990s when it was acquired by Lockheed Martin, and subsequently by General Dynamics.

General Dynamics, a premier American defense contractor that receives over three percent of total federal spending on contractors, has maintained operations in this sector, evolving from rudimentary wartime machining to the development of advanced, fully integrated land, sea, and air systems. Their portfolio includes artificial intelligence, vision systems, electronic warfare components, and advanced combat vehicles.

Modern aerospace and defense engineering inherently involves immense technical risk. When engineering teams develop next-generation stabilizing mechanisms for aircraft turrets or vehicle-mounted weapon systems, they face profound technical uncertainty regarding the structural integrity of novel composite materials under extreme kinetic stress and thermal variance. To resolve this, engineers rely on the hard sciences—specifically mechanical engineering, materials science, and physics. The process of experimentation involves developing complex computer-aided design (CAD) models, utilizing finite element analysis (FEA) software to simulate stress, and physically testing prototypes in environmental chambers. Hypotheses regarding alloy compositions or algorithmic dampening controls are systematically tested, analyzed, and refined iteratively until rigorous military specifications for reliability and performance are achieved.

Under the United States federal tax code (IRC § 41), the wages of the mechanical engineers designing these components, the software engineers writing the targeting code, and the technicians building the prototypes constitute highly eligible QREs. The costs of the raw materials, such as specialized composites and circuitry, that are consumed or destroyed during destructive physical testing qualify as supply QREs. Because General Dynamics operates physical facilities within Burlington, the wages paid specifically to the Vermont-based engineers and the supplies consumed within the state’s borders are fully eligible for the Vermont state R&D credit under 32 V.S.A. § 5930ii. This requires careful geographic sourcing by corporate tax departments to ensure that costs associated with operations in other states, such as the company’s headquarters in Reston, Virginia, are strictly excluded from the Vermont apportionment schedule.

Case Study 2: Specialty Food and Beverage Formulation

Burlington is globally recognized as the birthplace of the premium specialty food movement, a sector epitomized by Ben & Jerry’s. The company was founded in 1978 by Ben Cohen and Jerry Greenfield in a renovated gas station on the corner of College and St. Paul Streets in downtown Burlington. Leveraging Vermont’s rich agricultural heritage—particularly its dairy and maple syrup production—the founders rapidly expanded their footprint. In 1980, they rented space in an old spool and bobbin mill on South Champlain Street, a direct adaptive reuse of the city’s 19th-century manufacturing infrastructure, to begin packing their ice cream in pints. The company successfully navigated intense early competition, including a famous 1984 lawsuit against Pillsbury (the parent company of Häagen-Dazs) characterized by the “What’s the Doughboy Afraid Of?” campaign, eventually growing into a multinational brand operating as an independent subsidiary of Unilever.

While the brand is famous for its marketing and social mission, maintaining market share in the modern era requires sophisticated food science and ongoing, capital-intensive research and development. This is particularly true as the company adapts to stringent clean-label standards and the massive consumer shift toward plant-based, non-dairy alternatives. Developing a non-dairy, oat-based ice cream that perfectly mimics the texture of dairy is not a mere culinary exercise; it is an exercise in complex food chemistry. Traditional dairy fat acts as a natural emulsifier, providing the creamy mouthfeel and structural stability consumers expect. When transitioning to an oat base, severe technical uncertainty arises regarding how to achieve colloidal stability, prevent ice crystal formation during the freezing process, and match the structural viscosity of dairy without using artificial stabilizers that are banned by the brand’s clean-label ethos.

The company’s research and development scientists must rely on biological science and organic chemistry. The process of experimentation involves formulating multiple variations of lipid matrices, such as combining oat milk with specialized plant-derived oils or sunflower lecithin. Scientists conduct rheological testing to measure viscosity, microscopic analysis to observe emulsion stability, and extreme shelf-life stress testing under variable temperature controls to ensure the product does not separate or degrade during global distribution.

The United States R&D tax credit is not limited to high-tech machinery or software; food science is a heavily supported industry. The wages of food scientists, chemists, and quality assurance engineers actively engaged in overcoming these oat-base formulation challenges are qualified under federal law. The cost of raw ingredients, such as oats, specialized cocoa, and novel emulsifiers used in experimental batches that are discarded rather than sold, qualify as supply QREs. Because Ben & Jerry’s maintains primary manufacturing and R&D facilities in Vermont, including their headquarters in South Burlington and factory in Waterbury, the formulation science conducted at these local facilities generates significant Vermont-sourced QREs. If the company pays a third-party laboratory physically located in Vermont to conduct advanced microbiological testing on shelf-life, a statutorily defined percentage of those contract research payments would also qualify for the state credit.

Case Study 3: Automotive Digital Marketing Software Development

Despite its reputation as a rural, mountainous state, Vermont possesses a thriving, high-density software sector. This industry developed in Burlington primarily due to a unique congruence of factors: an influx of computer science and engineering talent from UVM, and an intensely outdoorsy, locavore culture that actively attracts highly skilled tech workers seeking a specific quality of life. In 2004, the growing number of software firms banded together to form the Vermont Software Developers’ Alliance (vtSDA) to support an ecosystem that eventually grew to hundreds of companies.

A dominant historical anchor in this sector was Dealer.com, a global leader in online marketing solutions, inventory management, and CRM software specifically designed for the automotive industry. Operating out of a retrofitted Pine Street manufacturing plant—formerly the home of Specialty Filaments workers—the company created a modern corporate environment featuring indoor tennis courts and organic cafes, completely transforming the industrial space into a digital innovation hub. The company experienced explosive growth, recognizing revenue increases of over 1,270 percent during a five-year period, driven entirely by continuous software engineering.

Building a scalable, cloud-based software architecture capable of rendering real-time, dynamic automotive inventory across thousands of dealership websites simultaneously involves massive technical uncertainty. Software engineers face continuous challenges related to database concurrency, latency reduction, and algorithmic optimization for digital retailing integration across fragmented legacy dealership management systems. Relying on the principles of computer science, engineering teams engage in a rigorous process of experimentation, typically structured through Agile software development lifecycles. They hypothesize that a new caching algorithm, or a fundamental architectural shift from a monolithic structure to containerized microservices, will reduce server load times and improve data synchronization. They test these hypotheses by writing original code, running load simulations, conducting regression testing, and pushing beta versions to sandbox environments, continuously iterating on the codebase based on highly specific performance metrics.

Software development is one of the most heavily scrutinized areas under Section 41 of the federal tax code, primarily due to the complex Internal Use Software (IUS) regulations. Because Dealer.com’s software is developed to be licensed and sold to third-party automotive dealerships, it is classified as Non-IUS, meaning it is not subject to the punishing High Threshold of Innovation test that restricts internal back-office software claims. The wages of full-stack developers, quality assurance engineers, and systems architects writing code to resolve capability and design uncertainties are highly eligible QREs. The software industry generates the vast majority of its QREs through W-2 wages rather than physical supplies. Therefore, the ability to claim the Vermont state credit under 32 V.S.A. § 5930ii hinges entirely on the developers being physically located in Vermont. If a Burlington-based software firm utilizes developers working remotely from other states, the wages paid to those out-of-state employees must be rigorously stripped out of the BA-404 state apportionment calculation, as they fail Vermont’s strict geographic sourcing requirement.

Case Study 4: Renewable Energy and Biomass Engineering

Burlington has garnered international recognition as one of the first municipalities in the United States to source 100 percent of its citywide electricity grid from renewable energy. This commitment to sustainability is deeply tied to the region’s abundant natural forestry resources. The city’s clean energy journey began in 1978 when local leaders made the strategic decision to decommission an aging, highly polluting coal plant and replace it with the McNeil Generating Station, located in the Intervale area near Burlington’s Old North End. Operated by the Burlington Electric Department (BED) and jointly owned by Green Mountain Power and VPPSA, the 50-megawatt plant runs primarily on locally sourced biomass—specifically, wood chips harvested within a 60-mile radius of the station.

In recent years, driven by the city’s aggressive Net Zero Energy Roadmap, BED and its partners have moved beyond basic electricity generation to engineer an advanced District Energy system. This complex infrastructural project is designed to capture thermal waste heat from the McNeil plant’s steam turbines and distribute it via subterranean piping to commercial sectors, such as the University of Vermont Medical Center, to drastically reduce commercial natural gas usage and eliminate approximately 13,000 tons of carbon dioxide emissions annually.

Retrofitting a 40-year-old biomass plant to serve as the anchor for a commercial district energy network poses unique and severe technical engineering challenges. Engineers face immense uncertainties regarding thermal fluid dynamics—specifically, how to efficiently capture low-grade steam turbine waste heat, transfer it through massive heat exchangers into a closed-loop municipal piping network, and minimize thermal loss across long physical distances through varying geological conditions. Furthermore, they face complex environmental engineering uncertainties regarding the optimization of electrostatic precipitators and air quality control devices to strictly limit nitrogen oxide and particulate emissions in densely populated adjacent neighborhoods. The process of experimentation involves complex mechanical and thermodynamic modeling. Engineers test different steam extraction parameters, evaluate alternative composite insulation materials for the subterranean piping, and simulate thermal load balancing under extreme Vermont winter conditions to ensure continuous reliability.

Research conducted by municipal utility departments can be structurally complex from a tax perspective, as tax-exempt entities cannot directly utilize nonrefundable income tax credits. However, because the McNeil Generating Station is a joint venture that includes private, tax-paying entities like Green Mountain Power (which owns 31 percent of the facility), and because the project utilizes private engineering contractors who may retain financial risk and economic rights to their designs, the federal R&D credit remains highly applicable. The development of customized emissions scrubbing mechanisms or novel thermal transfer designs relies on engineering principles and seeks to improve plant capability, satisfying the statutory requirements of IRC § 41. Furthermore, district energy research is inherently site-specific. The engineering assessments, geological surveys for pipe routing, and thermodynamic testing are conducted physically in Burlington. If a private utility partner incurs these costs using local engineers, the wages and localized supply consumption directly translate to highly valuable Vermont QREs eligible for the state credit.

Case Study 5: Biotechnology and Life Sciences Instrumentation

The most robust intersection of academic research and commercial manufacturing in Vermont is found in the life sciences sector. A prime historical example is BioTek Instruments. The company was originally founded by University of Vermont Professor Norman Alpert and was subsequently grown by his sons, Briar and Adam Alpert, into a global leader in life science technology. Operating out of a 40,000-square-foot facility in Winooski, immediately adjacent to Burlington, BioTek develops, engineers, and manufactures precision microplate readers, washers, dispensers, and integrated software. These instruments are vital tools utilized globally in pharmaceutical research, epidemiology, and basic science research focusing on disease mechanisms such as Alzheimer’s, Parkinson’s, and cancer.

The region’s biotechnology footprint has expanded significantly in recent years through deliberate technology transfer initiatives designed to move ideas from UVM research labs into the commercial sector. This is exemplified by UVM’s opening of the BioLabs Innovation Center, a $2.2 million conversion of facility space that provides state-of-the-art laboratory infrastructure and incubator space for early-stage life science startups. This alignment with institutions in major biotech hubs like Boston and San Diego capitalizes on UVM’s deep research capabilities.

Developing precision life science instrumentation requires bridging multiple, highly complex scientific disciplines. When a company like BioTek designs a new automated microplate washer, engineers face profound technical uncertainty regarding micro-fluidics. They must ensure that the mechanical dispensing mechanism can handle highly viscous biological reagents rapidly and without cross-contamination, while simultaneously ensuring the optical sensors in the microplate reader can accurately measure fluorescence or luminescence at the microscopic level across thousands of test wells. This fundamentally relies on biological science, optical physics, and mechanical engineering. The process of experimentation is exceptionally rigorous. Engineers design prototype fluidic manifolds, subject them to extensive cyclic fatigue testing, analyze dispensing accuracy down to the microliter using advanced statistical software, and systematically refine the pump mechanisms, optical alignments, and software algorithms based on the failure points identified during empirical testing.

The development of new biomedical laboratory equipment represents the absolute gold standard for IRC § 41 eligibility. The research inherently relies on hard sciences, seeks to create a vastly improved product, faces technical uncertainty at the outset, and follows a strict scientific method of evaluation. Under federal law, the wages of biomedical engineers, software developers, and optical physicists are fully qualified. Furthermore, the expensive prototypes and raw materials consumed during the testing phases that cannot be sold to customers are considered eligible supply QREs. Regarding state eligibility, BioTek maintains a fierce commitment to local manufacturing, ensuring that virtually all of their engineering and manufacturing occurs within Vermont. This localization makes their federal QRE base almost entirely applicable to the Vermont state credit. For emerging startups utilizing the new UVM BioLabs space, the rental cost of specialized computers or laboratory equipment leased exclusively for research activities within the Colchester facility also qualifies as a Vermont-specific QRE. If these startups contract UVM to perform basic research on their behalf, those payments may qualify for the 65 percent (or potentially 75 percent consortium) contract research inclusion rate under both federal and state law.

Exhaustive Analysis of United States Federal R&D Tax Credit Requirements

The United States tax code incentivizes technological innovation through a highly structured system of tax credits. The legislative intent is to spur domestic economic growth by directly offsetting the inherent financial risks associated with the process of discovering new information and developing new products, processes, or software. Internal Revenue Code (IRC) Section 41 establishes the Credit for Increasing Research Activities. It is imperative to distinguish this from IRC Section 174, which governs the treatment of research and experimental expenditures as standard deductions. In contrast, Section 41 provides a highly valuable dollar-for-dollar reduction in a taxpayer’s actual tax liability based on incremental increases in qualified research spending.

Because tax credits are a matter of legislative grace, they are narrowly construed by the courts and the Internal Revenue Service (IRS), placing the absolute burden of proof upon the taxpayer to substantiate every dollar claimed. To claim the federal credit, a taxpayer’s activities must strictly adhere to a statutorily defined four-part test, which serves as the universal definition of “qualified research.”

The Four-Part Test for Qualified Research

Under IRC § 41(d), an activity must simultaneously satisfy all four of the following criteria to be considered qualified research. The failure to meet even one criterion results in the total disqualification of the associated expenses.

Section 174 Permitted Purpose

The expenditure must be eligible for treatment as a research and experimental expenditure under IRC § 174. The research must relate to a new or improved function, performance, reliability, or quality of a “business component”. A business component is statutorily defined as a product, process, computer software, technique, formula, or invention to be held for sale, lease, or license, or to be used by the taxpayer in a trade or business. The IRS Audit Techniques Guide explicitly instructs examiners to be alert to claimed expenses for research related to non-functional aspects of a business component; research related to style, taste, cosmetic, or seasonal design factors is explicitly excluded by law.

Technological in Nature

The process of experimentation must fundamentally rely on principles of the hard sciences. Specifically, the statute identifies engineering, physical science, biological science, or computer science as the only acceptable foundations. Knowledge derived from the social sciences, arts, humanities, economics, or market research is strictly prohibited from qualifying, ensuring that the credit rewards true scientific advancement rather than clever marketing or business administration optimization.

Elimination of Technical Uncertainty

At the onset of the research project, there must be a genuine, documented uncertainty regarding the capability or method of developing the business component, or the appropriate design of the business component. If the solution to a problem is readily apparent to a competent professional in the specific field utilizing existing knowledge, technical uncertainty does not exist. The taxpayer must prove that the information available to them did not establish the capability or method for achieving the desired result, necessitating an experimental phase to discover that new information.

Process of Experimentation

This is the most heavily litigated prong of the four-part test. The statute dictates that substantially all of the activities must constitute elements of a process of experimentation designed to evaluate one or more alternatives to achieve a result. This involves a methodical, scientific approach: developing a hypothesis as to how a new alternative might be used to develop a business component, systematically testing that hypothesis (e.g., through physical modeling, computer simulation, or structured trial and error), analyzing the resulting data, and ultimately refining or discarding the hypothesis based on the empirical results.

Qualified Research Expenses (QREs)

If a specific project successfully meets the rigorous four-part test, the costs directly associated with that activity can be aggregated as Qualified Research Expenses (QREs) under IRC § 41(b). The tax code is highly prescriptive regarding what constitutes a QRE; if an expense is not explicitly set forth in Section 41(b), a taxpayer may not claim it, regardless of its necessity to the research. QREs are strictly limited to four main categories.

The Complexities of Internal Use Software (IUS)

Software development faces an elevated level of scrutiny under the federal tax code, representing one of the most complex areas of R&D tax compliance. The regulations distinguish heavily based on the intended end-user of the software.

Software developed by the taxpayer primarily for their own internal general and administrative functions is classified as Internal Use Software (IUS). This is specifically targeted at back-office functions such as financial management (accounting, inventory management, budgeting), human resource management (payroll, benefits administration), and support services (data processing, facility management). Historically, the legislative intent was to exclude IUS from the R&D credit entirely, viewing it as standard business operational expenditure rather than true innovation. However, final regulations established narrow exceptions, allowing IUS to qualify if, and only if, it satisfies an additional three-part “High Threshold of Innovation” (HTI) test.

To satisfy the grueling HTI test, the internal software must:

• Be Highly Innovative: The software must be intended to result in a reduction in cost, an improvement in speed, or other measurable improvement that is both substantial and economically significant to the taxpayer. Incremental improvements do not qualify.
• Involve Significant Economic Risk: The taxpayer must commit substantial resources to the development, and there must be substantial uncertainty, explicitly because of technical risk, that the resources can be recovered within a reasonable period.
• Not Be Commercially Available: The software cannot be purchased, leased, or licensed and used for the intended purpose without massive modifications that would themselves satisfy the innovation and risk requirements.

Conversely, software developed to be sold, leased, licensed, or marketed to third parties (Non-IUS) does not face the HTI test and is evaluated under the standard four-part test. Furthermore, Dual Function Software (DFS)—software that supports internal administration but also allows third parties to interact directly with the taxpayer’s business systems—is subject to complex safe harbor rules. Under the safe harbor, if a taxpayer reasonably anticipates at the onset of development that at least 10 percent of the software’s use will be by third parties, 25 percent of the research costs are automatically eligible for the credit without facing the HTI test.

Exhaustive Analysis of Vermont State R&D Tax Credit Requirements

The State of Vermont heavily leverages the federal statutory definitions established under IRC § 41, creating a system of legislative conformity. However, it applies its own localized mathematical adjustments and strict geographic restrictions to ensure that the tax incentive directly stimulates scientific research and economic growth within its own sovereign borders.

Statutory Provisions and Core Calculation

Under the Vermont Statutes Annotated, specifically 32 V.S.A. § 5930ii, a taxpayer of the state is eligible for a credit against Vermont income taxes equal to 27 percent of the amount of the federal tax credit allowed in the taxable year for eligible research and development expenditures under 26 U.S.C. § 41(a). It is important for practitioners to note that prior to January 1, 2014, the state credit rate was 30 percent, but legislative amendments reduced and stabilized the rate at the current 27 percent.

This state credit operates as a nonrefundable incentive, meaning it can reduce a taxpayer’s liability to zero, but the state will not issue a refund check for any excess credit generated beyond the tax owed. The credit can be applied against either Vermont personal income tax or business and corporate income tax liabilities. To mitigate the nonrefundable nature of the credit, the statute explicitly provides that any unused credit available may be carried forward for a period of up to 10 years, allowing companies engaged in long-term, pre-revenue R&D phases to preserve the value of the incentive.

Geographic Sourcing and Proration Mechanics

The most critical and challenging requirement of the Vermont R&D tax credit is geographic sourcing. The statute explicitly states that the credit applies only to eligible expenditures that are made within this State. Vermont does not create its own separate base amount rules or alternative simplified calculation methodologies; instead, it relies entirely on the complex federal calculations but restricts all variables exclusively to Vermont-sourced data.

To calculate the base amount required for the state credit, a taxpayer must undergo a rigorous, multi-year data extraction process:

• The taxpayer must identify only Vermont-sourced QREs and only Vermont-apportioned gross receipts for the prior four tax years to establish the fixed-base percentage under federal rules.
• The fixed-base percentage is calculated as the sum of prior Vermont QREs divided by the sum of prior Vermont gross receipts. (For startup companies with no prior data, Vermont conforms to the federal startup rules, utilizing a 3 percent fixed-base for the first five years, scaling up over a decade).
• The final base amount is then determined by multiplying the established fixed-base percentage by the average Vermont gross receipts for the prior four years.

For pass-through entities or corporations operating facilities in multiple states, this statutory framework mandates a highly precise proration of all federal QREs. Supplies must be physically consumed in Vermont, contract research must be physically performed by third parties located in Vermont, and employee W-2 wages must be allocated strictly to personnel who are physically located in Vermont while conducting, supporting, or supervising the research.

Tax Administration, Compliance, and Unprecedented Transparency

To properly claim the state credit, taxpayers are required to file Form BA-404, Tax Credits Earned, Applied, Expired, and Carried Forward, alongside their standard Vermont corporate (Form CO-411) or business income tax returns. The Department of Taxes provides strict guidance regarding corporate structures. In cases involving unitary combined group returns, state tax credits cannot be freely combined or shifted across the affiliated group to offset the liabilities of highly profitable out-of-state subsidiaries; the application of the R&D credit is strictly limited to the specific member entity to which the credit is originally attributed.

Furthermore, Vermont mandates an unprecedented level of corporate transparency for this specific tax incentive. Under 32 V.S.A. § 5930ii(c), the Vermont Department of Taxes is legally compelled to publish annually, on or before January 15, a public list containing the names of all taxpayers who have claimed the R&D credit during the most recently completed calendar year. This forces companies to weigh the financial benefit of the credit against the public disclosure of their tax strategies and R&D activities.

Tax Administration Guidance and Landmark Case Law

The interpretation of IRC § 41, and by direct extension the eligibility for the Vermont state credit, has been shaped significantly by federal tax court rulings. Understanding this jurisprudence is essential for structuring research activities and surviving subsequent audits by the IRS or the Vermont Department of Taxes.

Defining the Process of Experimentation: Union Carbide

In Union Carbide Corp. v. Commissioner (T.C. Memo. 2009-50), the Tax Court provided foundational guidance on the statutory meaning of a “process of experimentation.” The IRS frequently challenges taxpayers on this prong, arguing that standard engineering trial-and-error does not meet the legal threshold. The court in Union Carbide ruled decisively that the process requires the application of the scientific method—defined legally as an analytical technique by which a hypothesis is formulated and then systematically tested through observation and experimentation.

The court elaborated that a taxpayer must develop a specific hypothesis as to how a new alternative might be used to develop a business component, test that hypothesis methodically in a scientific manner, comprehensively analyze the resulting data, and then actively refine or discard the hypothesis based on those results. This ruling explicitly invalidates simple, undocumented “trial and error” that lacks systematic rigor, requiring businesses to maintain formal testing protocols and iterative design documentation.

The Substantially All Rule and Fraction Mechanics: Little Sandy Coal

The 2023 U.S. Court of Appeals for the Seventh Circuit opinion in Little Sandy Coal Co., Inc. v. Commissioner represents one of the most critical recent developments in R&D case law, particularly regarding the intersection of IRC § 41 and IRC § 174. The case centered intensely on the “substantially all” rule found in IRC § 41(d)(1)(C), which mandates that at least 80 percent of a taxpayer’s research activities for a given business component must constitute elements of a true process of experimentation.

The IRS had challenged the taxpayer’s claim regarding the design and construction of large vessels. While the appellate court ultimately affirmed the denial of the credit due to the taxpayer’s failure to adequately document exactly which employee activities constituted research, it provided a massive, taxpayer-favorable clarification on the mathematics of the 80 percent fraction calculation.

The higher court ruled that costs associated with the “direct support” and “direct supervision” of research activities—provided they qualify as deductible research expenses under IRC § 174—must be included in both the numerator and the denominator of the 80 percent calculation. Prior to this appellate intervention, the Tax Court had erroneously interpreted the statute in a manner that placed support and supervision time in the denominator (expanding the total base of activities) but excluded it from the numerator, artificially suppressing the resulting percentage and making the 80 percent threshold virtually impossible for industrial manufacturers to clear. Despite winning the highly technical legal argument on the formula construction, the taxpayer in Little Sandy lost the case entirely because they “failed to provide a principled way to determine the portion of employee activities that constituted elements of a process of experimentation”.

Strict Documentation Requirements: George v. Commissioner

The burden of proof in all tax matters falls entirely on the taxpayer. In the recent decision George v. Commissioner (T.C. Memo. 2026-10), the United States Tax Court violently reinforced the foundational principle that the four-part test must be substantiated through contemporaneous records.

In this case, the court struck down the taxpayer’s claim in its entirety because the company relied on reconstructed narratives assembled by tax consultants years later during the audit process, rather than systematic, timely documentation generated by the engineers during the actual research phase. The ruling highlights massive audit risk as the Internal Revenue Service increases scrutiny of base period calculations and research credit claims lacking timely documentation. For practitioners and corporate tax directors, this confirms that while genuine experimental activity in non-traditional industries can absolutely qualify for the credit, a lack of contemporaneous time-tracking data, test results, and organized cost data will result in a total, irreversible disallowance of QREs by examining agents.

Claiming the R&D credit at both the federal and state levels requires a highly defensive posture regarding documentation and cost allocation. Software developers at Burlington firms must utilize agile project management software to link specific code commits directly to technical uncertainties. Biomedical engineers must maintain rigorous, dated laboratory notebooks detailing the formulation of hypotheses and test results. Failure to proactively map these daily activities to the statutory requirements of IRC § 41 ensures failure upon audit.

The information in this study is current as of the date of publication, and is provided for information purposes only. Although we do our absolute best in our attempts to avoid errors, we cannot guarantee that errors are not present in this study. Please contact a Swanson Reed member of staff, or seek independent legal advice to further understand how this information applies to your circumstances.