funda17 Flashcards
(16 cards)
Public Sector Innovations and Inequality
*Public sector innovations often reduce inequality, particularly in medicine and healthcare.
*Example: Innovations like eradicating malaria or curing cancer benefit society broadly, especially in countries with universal healthcare.
*Targeting big societal challenges often helps lower-income populations, as they are the most affected by these issues.
Private Sector Innovations and Inequality
*The private sector’s impact on inequality is complex:
Skill-biased change: Increases returns to skilled workers, widening wage gaps.
Labor displacement: Reduces labor reliance, increasing returns to capital.
Market power: Concentration of power in industries (e.g., big tech) raises inequality.
*Inequality-reducing innovations include agricultural improvements, mobile telephony, and social media.
Innovation and Top Income Inequality
Aghion et al. (2019) studied the relationship between innovation and income inequality in U.S. states (1976–2013).
Focus: Top 1% income inequality, measured by the share of income owned by the top 1% (e.g., ~32% in New York, 2013).
Data: Panel dataset of 51 states over 37 years.
Measuring Innovation’s Impact
*Innovation was measured by patents per capita, weighted by citations within 5 years (indicating patent quality).
*Findings:
*Patent numbers per capita increased by 1.6 times during the study period.
*70% of patent growth was driven by new inventors, reflecting innovation’s expansion.
*Patents were tied to the geographic locations of inventors.
Implications of Innovation and Inequality
More patents per capita were linked to higher income shares for the top 1%, showing innovation often benefits high-income earners disproportionately.
Balancing innovation’s benefits with its inequality effects requires policies ensuring broader access to gains.
Public sector efforts can counteract inequality by focusing on universal benefits, while private sector innovations need careful regulation.
Innovation and Top Income Inequality
Study: Aghion, Akcigit, Bergeaud, Blundell, Hemous (2019).
Findings:
Innovation increases the income share of the top 1%, with a larger impact when innovation comes from new entrants (startups) rather than incumbents.
Reason: Social mobility—new innovators rise up income rankings.
Broader Inequality: No effect on the Gini coefficient, a measure of overall income inequality.
Wealth and R&D Efforts
Wealthier individuals are more willing to pay for innovative products and services.
Corporations respond to this demand by directing R&D efforts toward areas with higher expected returns (e.g., luxury healthcare or advanced education).
This trend can exacerbate inequality by prioritizing innovations that benefit higher-income groups.
Induced Innovation and Infant Mortality
Study: Cutler, Meara, Richards-Shubik on neonatal care (1963–1998).
Findings:
Significant investments in neonatal intensive care (e.g., post-1963) led to a 7% reduction in infant mortality per 100 NIH grants.
However, progress disproportionately benefited white infants, especially for conditions like Respiratory Distress Syndrome (RDS).
Outcome: Inequality in mortality reduction, with white infants seeing greater improvements.
Numerical Example of Induced Inequality
Disease 1 (affecting white infants more): Receives more funding due to higher absolute mortality.
Disease 2 (affecting Black infants more): Receives less funding.
Result: Mortality reductions are greater for Disease 1 (50%) compared to Disease 2 (25%), widening health outcome disparities between groups.
Social Barriers to Innovation
Study: Bell, Chetty, Jaravel, Petkova, Van Reenen (2019).
Findings:
A child’s likelihood of becoming an inventor is highly correlated with socioeconomic status, race, and gender.
Exposure to innovation (mentoring, networks) during childhood has a causal effect on becoming an inventor.
Underrepresented groups (e.g., low-income families, minorities, women) often lack access to such exposure, limiting their participation in innovation.
“Lost Einsteins” and Policy Solutions
Lost Einsteins: Untapped potential in underrepresented groups due to barriers like lack of access to resources and mentorship.
Policy Recommendations:
Enhance exposure to innovation for disadvantaged children (e.g., STEM education, mentorship programs).
These efforts can significantly increase both the number of inventors and the diversity of high-impact innovations.
Core Problems in the Economics of Innovation
Market Failures and the Need for an Economics of Innovation: Unlike typical goods, the production of new ideas suffers from market failures that prevent efficient production and diffusion. The core question is, why do we need a special field of the economics of innovation, when we don’t need an economics of apples or cars? This is because of three key issues:
◦
Non-Convexity: The production of new ideas often involves non-convex production sets. That is, ideas can be very costly to develop initially, but then can be used and recombined without bound.
◦
Externalities: The value of new ideas often involves externalities. The social benefits of new ideas are often greater than the private benefits captured by the innovator, leading to underproduction. The use of new ideas by one researcher does not preclude their use by others.
◦
Non-Deterministic Production: The production of new ideas is not deterministic. Research and development is a risky undertaking with uncertain outcomes.
*
Innovation and Economic Growth: The paper highlights that innovation is not just another economic object, but plays a cumulative and fundamental dynamic role in economic growth. Suboptimal innovation policy leads to suboptimal growth, as seen in exogenous, endogenous, and Neoschumpeterian growth models.
*
Challenges in Empirical Analysis of Innovation Policy: Measuring and valuing new ideas is difficult. Isolating empirical variation to create counterfactuals of how innovation might have been different with a different set of incentives is also very challenging
Science as a Non-Market Incentive
Motivations Beyond Profit: Scientific research is often driven by non-market incentives such as curiosity, creativity, academic tenure, and the pursuit of scientific credit and recognition. Scientists seek to establish priority of discovery.
◦
Priority: The reward structure is often winner-take-all. There is a tension between allowing a project to mature for longer (improving its quality) versus publishing or patenting to establish priority. High-potential projects are often executed with the lowest quality due to competition among researchers.
◦
Scientific Norms and Institutions: Publication in academic journals establishes priority and citations provide a measure of a scientific contribution’s importance.
*
Autonomy and “Taste” for Science: Scientists often have a “taste” for science, meaning they are willing to accept lower wages in exchange for choosing their own research projects and participating in scientific communication. Scientists are willing to pay to do science.
*
Tradeoffs in Research Funding: Academia may be a cheaper input to innovation due to scientists’ willingness to accept lower wages. However, scientists may work on projects they find personally interesting or prestigious, rather than projects with commercial value. Giving control to firms can direct scientists to work on more valuable projects, but is more expensive since scientists must be paid a wage premium
Market-Oriented Policies for Innovation
Tax and Subsidy Policies:
◦R&D Tax Credits: These policies aim to reduce the costs of research, and can be effective in boosting R&D. A 10% fall in the tax price of R&D generates at least a 10% increase in R&D in the long run. However, firms may relabel expenses as R&D to take advantage of these credits.
*Direct Subsidies: Governments often subsidize research investments at private firms, particularly small firms. Small firms are more likely to be liquidity-constrained, and direct grants are complementary to other R&D investments. Larger firms tend to use grants to pay for inframarginal investment, making tax credits a substitute.
*
Patents:
◦Incentives: Patents grant inventors the right to exclude others from economically exploiting an invention for a limited time in exchange for disclosing how the invention functions. They aim to balance the need to reward inventors with the need to promote innovation.
◦Patent Process: The patent application process involves submitting a written description of the invention, including a discussion of prior art, and a specific list of claims. Examiners determine if the application is patentable. The process often involves multiple rounds of rejection and revision.
◦Patent Limitations: Many inventions are not patented, and the propensity to patent varies across industries. Patents also vary significantly in their quality or value, meaning simple counts of patents are not always meaningful measures of innovation.
◦Strategic Use: Firms may tie incentive pay to patenting, which can influence how patents are interpreted. Firms also use patents to signal their technological lead, potentially reducing competition.
*
Labor Market Policies:
◦Immigration: Policies that attract high-skilled immigrants to the U.S. increase innovation. Immigrants account for a disproportionately high percentage of inventors and innovation, particularly when measured by the quality of patents.
◦Non-Compete Agreements: These agreements affect labor mobility and the formation of spin-out companies. California law prohibits non-compete agreements, whereas Massachusetts has historically enforced them. Non-competes can discourage firm-specific investments in human capital and relationships between customers and suppliers
Innovation, Diffusion, and Growth
Diffusion: The social value of innovation depends on its diffusion, which refers to the change over time in who produces and uses the invention. Diffusion includes both technology diffusion (adoption by new users) and knowledge diffusion (spread of ideas).
*
Barriers to Diffusion:
◦Tacit Knowledge: Many inventions are tacit, rather than codified, and require explanation to be exploited. Even within a firm, transferring knowledge about production can be costly.
◦Information Asymmetry: Asymmetric information and social contacts can affect the spread of new technologies. Firms often make new products similar to existing products to facilitate learning.
◦Adoption Costs: The costs of adoption or information transmission can limit diffusion.
◦Strategic Barriers: Some agents benefit from preventing the use of new technologies, as in the case of “goldbricking” by workers.
*
Measuring Diffusion:
◦Patent Citations: Patent citations are used to study diffusion, but have limitations as they are not a complete metric of spillovers. Examiner-added citations, for example, do not measure information flow between inventors.
◦
Geographic Localization: Knowledge spillovers often have a local component. Geographic proximity increases the likelihood of patent citations.
*Text Analysis: Machine learning and text analysis of patents can provide insights into technology diffusion beyond citations.
*
Standard Setting Organizations (SSOs): SSOs attempt to avoid wasteful innovation and marketing of technologies that will not be adopted by developing standards for interoperability. SSOs also address concerns with “standard-essential” patent holders demanding high licensing fees after standards have been adopted
Innovation and Inequality
Bidirectional Link: The paper examines the bidirectional link between inequality and innovation. It explores whether innovation contributes to rising inequality and how policies like the patent system affect inequality.
*
Product Variety and Inflation: Higher-income households experience faster increases in product variety and lower inflation. This may be due to the increased demand for “premium” products, which leads to disproportionate benefits for wealthier consumers.
*
Lost Potential Innovators: The paper explores whether inequality leads to a loss of potential innovators.
◦Parental Resources: The probability of an individual becoming an inventor is strongly correlated with their parental income. The probability of patenting for an individual whose father is at the top of the income distribution is about ten times larger than that of someone with a father at the bottom of the income distribution. This is true even in countries with less income inequality and higher social mobility.
◦Gender and Racial Gaps: There are also significant gender and racial gaps in innovation. The gap in patenting between African Americans and whites is larger during periods of ethnic and political violence. Top-scoring girls in math competitions are drawn nearly exclusively from a small set of elite schools, which suggests that many girls with the ability to reach high levels of math achievement are not doing so.
◦Global Inequality: Students from lower income countries who achieve the same scores in international math competitions are less likely to contribute to math research than their counterparts from wealthier countries
