Research and development (R&D): expense vs investment—why this distinction matters for investors

Research and development is a paradox in financial reporting. Accounting rules treat R&D as an operating expense that immediately reduces earnings, yet economically it functions as an investment in future revenue and competitive advantage. This distinction creates an analytical challenge: A company slashing R&D to boost short-term earnings might be destroying long-term value. Conversely, a company investing aggressively in R&D might be sacrificing today's earnings for tomorrow's dominance. Understanding how to interpret R&D spending is essential for accurate company valuation and risk assessment.

Quick definition: Research and development (R&D) encompasses all costs to discover, develop, and improve products, services, or processes. It appears as an operating expense on the income statement but is often better understood as an investment in future competitive position and revenue growth.

Key takeaways

• R&D is expensed immediately under GAAP, but its benefits accrue over years or decades
• R&D intensity (R&D / Revenue) varies dramatically by industry: 15–20% for pharma, 10–25% for software, 2–5% for retail
• Rising R&D as a percentage of revenue during strong growth can signal healthy investment in innovation
• Declining R&D as a percentage of revenue might indicate cost-cutting that sacrifices future growth for near-term earnings
• Some analysts capitalize a portion of R&D (treating it as an asset) to normalize comparisons across companies
• The success of R&D is measured not by spending level but by the productivity of that spending—do R&D projects generate profitable new products?

The accounting rule: why R&D is expensed, not capitalized

Under Generally Accepted Accounting Principles (GAAP), R&D costs are expensed immediately—they reduce earnings in the period incurred—because the future benefits are uncertain. When a company spends $100 million on research, accountants cannot predict whether that research will generate new products, when those products will reach market, or how profitable they will be.

By contrast, when a company purchases a building for $100 million, the asset is capitalized on the balance sheet and depreciated over 20–30 years. The building's utility and lifespan are relatively certain, so the cost is matched to the periods in which the asset generates revenue.

R&D doesn't receive capitalization treatment because:

1. Uncertainty: The research might fail. For every drug that reaches market, a pharmaceutical company invests in dozens that don't.

2. Timing: The payoff might be years or decades away. A researcher's work today might not generate revenue for 10 years.

3. Identifiability: It's hard to link specific R&D spending to specific future products. A software company's R&D contributes to multiple products simultaneously.

This conservative accounting approach protects investors by not overstating assets on the balance sheet. However, it creates an important analytical problem: Expensing R&D artificially depresses reported earnings, making companies that invest heavily in innovation appear less profitable than they truly are.

The economic reality: R&D as investment

While GAAP requires immediate expensing, economically R&D is an investment. The company is trading current earnings for future revenue and competitive advantage.

Consider the pharmaceutical industry. Developing a new drug requires $1–3 billion and 10–15 years of research, clinical trials, and regulatory approval. Under GAAP, all of this is expensed as it's incurred. The company's earnings look terrible during development. However, if the drug becomes a blockbuster generating $5 billion in annual revenue for 20 years of patent protection, the R&D investment was worth hundreds of billions in value.

This reality has led many analysts to adjust reported earnings by capitalizing a portion of R&D spending—treating it as an asset built over time rather than an immediate expense. The adjustment recognizes that R&D has future value not captured in GAAP earnings.

Reported Earnings (GAAP):

Revenue − COGS − SG&A − R&D = Reported Earnings

Adjusted Earnings (capitalizing R&D):

Revenue − COGS − SG&A − R&D Amortization = Adjusted Earnings
(R&D Amortization is typically 3–5 years worth of R&D)

By amortizing R&D over 3–5 years instead of immediately expensing it, adjusted earnings show a more accurate picture of sustainable profitability. However, this adjustment is not standardized; different analysts apply different assumptions.

R&D intensity by industry

R&D spending as a percentage of revenue varies dramatically by industry, reflecting the importance of innovation and the risk of obsolescence:

Highest R&D intensity (15–30% of revenue):

• Pharmaceuticals: 15–25%, due to the long development cycle, regulatory requirements, and risk of failure. A company must invest in dozens of potential drugs to get one to market.
• Biotechnology: 20–40%, especially for early-stage companies with minimal revenue but heavy research spending.
• Software and IT: 10–25%, depending on the company. Platform companies like Microsoft or AWS invest heavily in new features and cloud infrastructure.
• Semiconductors: 15–20%, reflecting the cost of chip design, fabrication plant improvements, and process innovation.

Medium R&D intensity (5–15% of revenue):

• Automotive: 5–10%, for vehicle design, electric vehicle development, and autonomous driving research.
• Industrial equipment: 5–12%, for manufacturing process improvements and new product lines.
• Consumer electronics: 8–15%, for product design and new technologies.
• Communications and aerospace: 10–15%, driven by regulatory requirements and long development cycles.

Lower R&D intensity (1–5% of revenue):

• Retail and e-commerce: 1–3%, focused on online experience and logistics optimization.
• Financial services: 1–4%, limited product innovation beyond regulatory changes.
• Utilities and energy: 1–2%, stable technologies with minimal innovation.
• Food and beverage: 1–3%, new product flavors and packaging but no technological breakthroughs.
• Telecommunications: 2–4%, infrastructure maintenance and incremental service improvements.

These ranges are not absolute. A software company with only 5% R&D might be underinvesting and facing obsolescence. A retailer with 3% R&D might be appropriately investing given its business model.

Assessing R&D quality: spending versus productivity

The amount a company spends on R&D matters far less than the productivity of that spending. Two companies might each invest 20% of revenue in R&D, but one generates blockbuster new products while the other invests in projects that never reach market.

R&D productivity metrics (often disclosed in earnings calls or 10-K sections):

1. New product revenue percentage: What percentage of annual revenue comes from products launched in the last 3–5 years? Higher is better, indicating R&D is delivering new revenue streams.

2. Time-to-market: How long does it take from research initiation to product launch? Shorter is usually better, though complex products (pharmaceuticals, aircraft) require longer timelines.

3. Patent productivity: How many patents does the company file per R&D dollar? More patents doesn't automatically mean more value, but it can indicate R&D output.

4. Product pipeline: Does the company disclose how many drugs are in clinical trials, or how many products in development? A robust pipeline suggests future revenue potential.

5. R&D success rate: For industries where it's measurable (pharma, software), what percentage of R&D projects reach commercialization? Higher success rates indicate efficient resource deployment.

6. Return on R&D: What is the incremental revenue or gross profit generated per R&D dollar invested? This is harder to calculate but reveals true productivity.

Most companies don't explicitly disclose these metrics, but investors and analysts can piece together signals from earnings calls, product announcements, and patent filings.

The risk of R&D cuts

When a company reduces R&D spending to boost short-term earnings, it may be mortgaging the future. In competitive industries, underinvestment in R&D can lead to:

1. Product obsolescence: Competitors' newer products capture market share. The company's installed base gradually shrinks.

2. Talent drain: Research scientists and engineers leave for better-funded competitors. Rebuilding these capabilities takes years.

3. Competitive disadvantage: New entrants with better-funded R&D disrupt the market. The company becomes a slow-moving incumbent.

4. Margin compression: Without innovation, the company competes on price, eroding gross margins.

This risk is highest in technology and pharmaceutical industries, where innovation is the source of competitive advantage. A company cutting R&D in these industries to hit short-term earnings targets might score points with investors expecting near-term performance but is setting itself up for long-term decline.

A flowchart of R&D analysis

Real-world example: Intel's R&D investment and struggles

Intel's R&D spending and the company's competitive decline illustrate the importance of R&D productivity.

2015: Intel reported revenue of $55.4 billion and R&D spending of $13.1 billion, an R&D intensity of 23.6%. The company was investing heavily in next-generation chip manufacturing processes.

2019: Intel reported revenue of $71.9 billion and R&D spending of $14.1 billion, an R&D intensity of 19.6%. Despite strong revenue, R&D intensity had declined slightly, signaling a shift toward profitability focus.

2023: Intel reported revenue of $54.2 billion and R&D spending of $17.8 billion, an R&D intensity of 32.8%. The company had increased R&D spending dramatically to regain manufacturing process leadership after falling behind TSMC and Samsung.

The story is instructive: Intel's R&D spending had become less productive in the 2015–2019 period. Despite heavy investment, the company failed to maintain manufacturing process leadership. Competitors (TSMC, Samsung) with comparable or higher R&D spending achieved better results. By 2023, Intel recognized the problem and increased absolute R&D spending while revenue fell—a costly but necessary course correction.

This example demonstrates that high R&D spending alone doesn't guarantee success. R&D execution, organizational agility, and strategic focus matter equally.

Comparing R&D across competitors

To assess whether a company is investing appropriately in R&D:

1. Calculate R&D intensity (R&D / Revenue) for the target company and three to five competitors using the latest annual report.

2. Compare the ratios:

   • If target is 3+ points below industry average: Underinvesting, risk of obsolescence or competitive loss
   • If within 3 points of average: Investing at market level
   • If 3+ points above average: Either investing for growth or inefficiently deploying R&D (assess revenue growth)

3. Track the trend: Plot R&D intensity for 5 years. Rising intensity might signal increased innovation investment or eroding productivity. Declining intensity might signal maturation or cost discipline.

4. Assess product pipeline: Does the company disclose how many products are in development? A robust pipeline suggests R&D is generating future revenue.

5. Monitor competitive wins and losses: Are the company's new products gaining market share or losing it? Are competitors' new products disrupting the market? This reveals R&D effectiveness.

Capitalizing R&D: the analyst adjustment

Some equity analysts capitalize a portion of R&D to provide a normalized earnings figure. The standard approach:

1. Estimate the useful life of R&D (typically 3–5 years)
2. Capitalize current-year R&D as an asset on the balance sheet
3. Amortize prior-year R&D over the estimated useful life
4. Add back current-year R&D expense and subtract R&D amortization to calculate adjusted earnings

Example:

Company with $100M revenue, $20M R&D

Reported Earnings (GAAP):
Revenue $100M - COGS $40M - SG&A $25M - R&D $20M = $15M

Adjusted Earnings (3-year R&D amortization):
Assuming 3 prior years of $20M R&D each:
Revenue $100M - COGS $40M - SG&A $25M - R&D Amortization $20M = $15M

(R&D Amortization = ($20M + $20M + $20M) / 3 = $20M)

In this case, capitalizing R&D doesn't change earnings because R&D has been stable. However, if R&D is growing, capitalization smooths the impact:

Company with rising R&D:
Year 1: $20M R&D
Year 2: $25M R&D
Year 3: $30M R&D

Reported Earnings impact:
Year 3: −$30M (full expense)

Adjusted Earnings impact (3-year amortization):
Year 3: −($20M + $25M + $30M) / 3 = −$25M

Capitalization is not standardized and should not be applied blindly. Use it to understand the potential impact of R&D on comparability, but don't mistake adjusted earnings for GAAP earnings.

Common mistakes in R&D analysis

1. Assuming more R&D spending is always better: A company wasting R&D dollars on unproductive projects is worse off than a company investing less but more wisely.

2. Ignoring strategic timing of R&D spending: A company facing near-term earnings challenges might legitimately defer discretionary R&D. Understanding the strategy matters.

3. Penalizing early-stage companies for high R&D ratios: A five-year-old biotech company with $5 million revenue and $20 million R&D has a 400% R&D ratio, but this is normal and expected. Early-stage companies are all investment.

4. Forgetting that R&D success is measured in outcomes, not inputs: Spending more on R&D is meaningless if the outputs don't generate new products or competitive advantage.

5. Overlooking outsourced R&D: Some companies outsource R&D to universities, contract research organizations, or partners. These costs might not appear in the R&D line item but represent real innovation investment.

6. Not adjusting for one-time R&D charges: Occasionally, a company records acquisition-related R&D costs (fair value of acquired in-process R&D) as an unusual expense. Separate these from ongoing R&D for trend analysis.

FAQ

Should I capitalize R&D when analyzing a company? It depends on your purpose. For comparing near-term earnings, use GAAP figures. For assessing long-term sustainable earnings, capitalizing R&D provides useful context. Many financial analysts and research firms capitalize R&D for publication.

What is the right R&D spending level? Industry-dependent. Software and pharma typically require 15%+ of revenue. Retail and utilities might get by with 2–3%. Compare to direct competitors.

How do I account for R&D done by customers or partners? This is often not captured in the company's R&D spending but represents real innovation. It's visible in joint ventures, partnerships, or customer development programs. Read the 10-K narrative for clues.

Can R&D spending be reduced without harming long-term competitiveness? Yes, if the reduction targets inefficient projects or duplication. No, if it's across-the-board cutting that eliminates essential innovation. Context and strategy matter.

What if a company acquires another company partly to gain its R&D team and technology? The acquisition price typically includes a premium for R&D value. After acquisition, the combined company's R&D spending might decline because duplication is eliminated. This is often positive (greater efficiency) not negative.

How is R&D different from capital expenditure on manufacturing equipment? Both are investments, but R&D is expensed under GAAP while capital equipment is capitalized. This creates a comparison problem: a capital-heavy company might show higher reported earnings than a research-heavy company with identical economic performance.

Should I use R&D as a percentage of revenue or absolute R&D dollars for comparison? Always use percentage. Absolute dollars vary with company size and are not comparable. A company spending $10 billion on R&D might be less productive than one spending $2 billion if the latter generates more new products.

Related concepts

• Operating expenses: SG&A, R&D, and more
• Selling, general and administrative expenses (SG&A)
• Depreciation and amortisation on the income statement
• Understanding the income statement: structure and purpose

Summary

Research and development is the paradox at the heart of financial reporting: Accounting rules treat it as an immediate expense that depresses earnings, yet economically it functions as an investment in future competitive advantage. This distinction is critical for investors. A company cutting R&D to boost near-term earnings might be sacrificing long-term value. Conversely, a company investing aggressively in R&D might be appropriately positioning itself for future growth, even if reported earnings suffer today. R&D intensity varies dramatically by industry—15–25% for pharmaceuticals and advanced software, but only 1–3% for retail and utilities. The key to R&D analysis is not the spending level but the productivity: Does the company's R&D generate blockbuster new products or does it fund unproductive research? By tracking R&D intensity over time, comparing it to competitors, and assessing the quality of new products and pipelines, investors can determine whether a company is investing wisely in innovation or burning cash on unproductive research.

Next

Depreciation and amortisation on the income statement