Bloom, Nicholas, John Van Reenen, and Heidi Williams. 2019. “A Toolkit of Policies to Promote Innovation.” Journal of Economic Perspectives, 33 (3): 163-84.DOI: 10.1257/jep.33.3.163

Overview

• Context: The US, since the 1970s, has seen a persistent slowdown in productivity growth, mirrored in other developed countries. This trend is accompanied by stagnant real wage growth and rising wage inequality.
• Purpose: The authors aim to advise policymakers on efficiently investing in innovation policy, exploring various levers and their effectiveness.

Key Points

1. R&D Spending Trends:
   • Growth Over Time: R&D spending as a percentage of GDP increased from 1.3% in 1953 to about 2.7% in 2015.
   • Shift in Funding Sources: Initially, the federal government was a major funder, but over time businesses have taken the lead, with their spending exceeding that of the federal government by more than double in 2015.
2. Worker Participation in R&D:
   • Stable Workforce Participation: In the U.S., the fraction of the workforce engaged in research has stabilized between 0.7% and 0.9%.
   • International Comparison: This is comparable to the EU but lower than Japan, where about 1% of the workforce is involved in research.
3. Government Role in Innovation:
   • Necessity of Intervention: Despite the risks of government failure, intervention is often essential due to market failures.
   • Examples of Success and Failure: Successful projects include jet engines, GPS, and the internet. Failures include the Concorde Anglo-French supersonic jet.
4. Measuring Spillovers:
   • Case Studies: Examine individual innovations, like Griliches’ study on hybrid corn, showing substantial social returns.
   • Production Function Approach: Links productivity growth to R&D investments, but faces challenges in establishing causality.
   • Patent Counts: Analyzes patent citations to infer knowledge spillovers, though citations can sometimes be strategic or added by examiners.
5. Financial Constraints on Innovation:
   • Social vs. Private Returns: Social returns to R&D are estimated at about 60%, compared to private returns of around 15%.
   • Challenges in Equity Funding: Pitches for funding often struggle due to the secrecy required before patenting, creating an information asymmetry.
   • Difficulties in Government Intervention: Governments often lack precise information to select high-quality investments, leading to challenges in designing effective policies.
6. Tax Incentives for R&D:
   • How They Work: Governments provide additional deductions against tax liabilities for R&D expenses.
   • OECD Figures: The U.S. federal R&D tax credit reduces the cost of U.S. R&D spending by about 5%, ranking in the bottom one-third of OECD nations in terms of generosity.
   • Challenges: Firms may reclassify other spending as R&D; relocation of activities towards areas with more generous fiscal incentives.
7. Patent Boxes:
   • Operation: They apply a lower tax rate to revenues linked to patents, encouraging firms to shift their intellectual property royalties into different tax jurisdictions.
8. Government Research Grants:
   • Functioning: Target specific types of R&D, like basic research in universities or applied R&D in industry.
   • Spillovers to Private Firms: Publicly funded R&D can lead to increased patenting and innovation in the private sector.
   • Crowding Out and Crowding In: Public R&D may replace (crowd out) or supplement (crowd in) private R&D, affecting overall innovation.
9. Human Capital Supply:
   • Bianchi and Giorcelli (2018) Study: Demonstrates how increasing STEM graduates leads to more innovation.
   • Role of Immigrants: Immigrants are more likely to be innovators and contribute substantially to the U.S. innovation landscape.
   • Innovators from Lower-Income Backgrounds: Fewer innovators emerge from these backgrounds; improving access to education and exposure to role models can counter this.
10. Intellectual Property and Patent Eligibility:
    • Patent Rights: Protect innovations that are novel, nonobvious, and useful.
    • Court Decisions: Recent rulings have carved out areas where patents may not apply, balancing social benefits against costs.
11. Impact of Competition on Innovation:
    • Schumpeter’s Argument: Monopoly profits incentivize innovation, and competition reduces these profits.
    • Arrow’s Replacement Effect: New entrants have no existing rents to lose and are more motivated to innovate.
    • Bloom and Romer’s Perspectives: Highlight the complex relationship between competition, innovation, and economic growth.
12. Focusing on Small Firms:
    • Rationale and Challenges: Small firms are often more responsive to support policies, but focusing solely on them may discourage growth.
13. Moonshots and Mission-Oriented Approach:
    • Justifying Large-Scale Projects: Projects like these are essential for addressing major challenges like climate change.
    • Political Considerations: They may also be more politically viable, leading to sustainable funding for research.

Innovation Policy Toolkit

Policy	Quality of Evidence	Conclusiveness of Evidence	Net Benefit	Time Frame	Effect on Inequality
Direct R&D Grants	Medium	Medium	**	Medium run	↑
R&D Tax Credits	High	High	***	Short run	↑
Patent Box	Medium	Medium	Negative	NA	↑
Skilled Immigration	High	High	***	Short to medium run	↓
Universities: Incentives	Medium	Low	*	Medium run	↑
Universities: STEM Supply	Medium	Medium	**	Long run	↓
Trade and Competition	High	Medium	***	Medium run	↑
Intellectual Property Reform	Medium	Low	Unknown	Medium run	Unknown
Mission-Oriented Policies	Low	Low	*	Medium run	Unknown

Notes:

• Net Benefit: Rated on a scale where three is the highest, indicating the strength and magnitude of average effects.
• Time Frame: Short run (next 3-4 years), medium run, or long run (approximately ten years or more).
• Effect on Inequality: ↑ indicates likely increase, ↓ indicates likely decrease, NA denotes not applicable, and Unknown where the impact is uncertain.

Toolkit Findings

1. Short-Term Effectiveness: Research and development (R&D) tax credits and direct public funding are highlighted as the most effective policies in the short term. These approaches can quickly stimulate R&D activities.
2. Long-Term Strategy: For longer-term impact, increasing the supply of human capital is more effective. This includes expanding university admissions in science, technology, engineering, and mathematics (STEM) fields. The benefits of this approach unfold over a longer period as the increased number of skilled individuals enters the workforce.
3. Impact of Skilled Immigration: Skilled immigration is shown to have significant positive effects even in the short run. This is due to the immediate contribution of skilled immigrants to the innovation ecosystem.
4. Moderate Benefits of Competition and Trade: Competition and open trade policies have more modest benefits for innovation. However, they are financially cost-effective, making them attractive options. These policies can stimulate innovation by exposing firms to larger markets and increased competition.
5. Inequality Considerations:
   • R&D Subsidies and Trade: Likely to increase inequality, as they increase the demand for highly skilled labor. In the case of trade, some communities might suffer due to job losses and the need for trade adjustment.
   • Increasing Skilled Labor Supply: Expected to reduce inequality by making high-skilled labor less scarce, thereby easing competitive pressures.
6. Diverse Perspectives: The table acknowledges that different policymakers and stakeholders might have varying opinions on the effectiveness and desirability of these policies.
7. Framework for Debate: The authors aim for this framework to spark further discussion and analysis regarding policies that can restore equitable growth in the modern economy.