WT Week 9 Flashcards
Motivation
Gender Roles and the Misallocation of Labour Across Countries (Ashraf et al., 2022)
To investigate the effect of gender roles on misallocation.
Special settings
Gender Roles and the Misallocation of Labour Across Countries (Ashraf et al., 2022)
The largest share of housework is attributed to women across all countries and time. Important to understand the effect on productivity. If men have an advantage for work outside the home, then the current allocation is efficient along gender lines. If not, then gender roles lead to misallocation as some women who stay home would be more productive in the workforce and some of the men in the workforce would be more productive at home.
Theory
Gender Roles and the Misallocation of Labour Across Countries (Ashraf et al., 2022)
If skills were better matched to roles across gender, there would be improvements in the productivity of each company/economy
Empirical design
Gender Roles and the Misallocation of Labour Across Countries (Ashraf et al., 2022)
“Investigates effect of gender roles on misallocation using microdata on gender pay gap among the employees of a large MNC combined with labour force participation across cohorts in the 101 countries of operation.
Quantify cost of misallocation through estimating parameters of the firm’s wage policy and matching mean and variance, to then show implied productivity by gender as well as average pay and utility of staying at home
Use ORBIS data to assess whether estimates are informative for other firms in the economy”
Data
Gender Roles and the Misallocation of Labour Across Countries (Ashraf et al., 2022)
White-collar, regular employees from a large MNE (100,000), with a homogenous workforce as standarised education requirements upon entry (college degree). Annual compensation data 2015-2019.
Key findings
Gender Roles and the Misallocation of Labour Across Countries (Ashraf et al., 2022)
“Women are paid more than men with the same experience/tenure/job when LFPR is low
Gap shrinks as LFPR increases and converge to standard levels for industry (-10%) when LFPR is highest
Women are more positively seelcted than men due to the marginal women moving from inside to outside home being more productive than the opposite man
Large gender productivity gap; estimated productivity gap larger in countries with weaker gender equity labour laws and with more conservative gender norms”
Interpretation / policy implications
Gender Roles and the Misallocation of Labour Across Countries (Ashraf et al., 2022)
“Implications for firms’ diversity policies:
- Average female applicant is more productive than the average male applicant
- Paying women hires more is justifiable on the grounds of productivity, regardless of whether diversity has an effect of its own
- Aiming for gener equity (pay, promotions, dismisalls) can turn out to be inequitable because selection generates different distributions of productivity; i.e. gender equity policies might hurt women as they limit the firm’s ability to reward performance”
Motivation
Who Becomes an Inventor in America? The Importance of Exposure to Innovation (Bell et al., 2019)
To understand what policies can be most effective in increasing innovation.
Special settings
Who Becomes an Inventor in America? The Importance of Exposure to Innovation (Bell et al., 2019)
“Innovation viewed as engine for economic growth, making it the centre for policies (including tax cuts to investment in STEM).
Lack of evidence on effectiveness as know little about the factors inducing people to become inventors.”
Empirical design
Who Becomes an Inventor in America? The Importance of Exposure to Innovation (Bell et al., 2019)
“Study lives of 1M+ US innovators using de-indentified database linking patent records to tax and school district records.
Tracking individuals from birth onwards to identify the key factors determining who becomes an inventor, as measured by filing a patent.”
Theory
Who Becomes an Inventor in America? The Importance of Exposure to Innovation (Bell et al., 2019)
Likelihood of innovation is highly determined by childhood experiences and background, making for inefficient allocation of skills, such that white males are overrepresented and would be better off in other roles.
Data
Who Becomes an Inventor in America? The Importance of Exposure to Innovation (Bell et al., 2019)
“Patent records: 1.7M patents granted between 1996-2014 in US (database by Google); 1.6M patent applications between 2001-2012 (by Strumsky)
Tax records: federal income tax records 1996-2012 (including gender, age, location, own and parental income); linked using inventor’s name, city and state, with 83% being successfuly linked
NYC School District records: test scores in childhood spanning 1988-1989 and 2008-2009, covering 2.5M children; linked with tax data at 89% matching rate”
Key findings
Who Becomes an Inventor in America? The Importance of Exposure to Innovation (Bell et al., 2019)
“Large disparities in innovation rates by socioeconomy class, race, and gender
Class: Children with parents in top 1% of income distribution are 10x more likely to become inventors than those with below-median income parents
Race: White children 3x more likely to become inventors than black children
Gender: Only 18% of inventors are female; gender gap shrinking gradually but will take 118 years to reach gender parity at current rate
Very little explained by differences in ability for early-ages via test scores as socioeconomic class and race are the confounding factors which explain any existing of disaprity, i.e. to become a US inventor relies on excelling in maths+sciences and having a rich family
By 8th grade, half of gap in innovation by income can be explained by test score differences as low-income children fall behind high-income peers overtime
Living in proximity to other inventors has a large effect, both on the likelihood of becoming an inventor and the type of invention itself (i.e. growing up in Silicon Valley are more likely to file patents in computing). Also matters in gender-specific manner, as innovation is only influenced by having own-gender inventors in the area. Low-income/minorities/women are less likely to have exposure through families and neighbourhoods
If girls were exposed to female inventors as boys to male inventors, gender gap in innovation would fal by half
Large disparity in income among inventors: top 1% earn $1.04M whereas top 20% earn $377K on average.
Gender/class/race remain under-represented across top inventors, suggesting ‘lost Einsteins’
Financial incentives unlikely to increase innovation as: 1) only effect small subset of individuals who have exposure to innovation; 2) unlikely to influence decisions of top inventors who matter most for economic growth (i.e. happy to work in their field whether earn $950K instead of $1M)
Predictions haven’t been tested empirarcally”
Motivation
Dossi (2023)
What are the consequences of the racial gap in science and innovation? In the United States, Blacks are significantly underrepresented in every scientific and innovation field, being one-fourth as likely to become scientists and one-eighth as likely to hold a patent compared to Whites. This underrepresentation may lead to private and societal costs, such as unequal sharing of patent benefits and economic and welfare losses for society. Recent initiatives like the “Unleashing American Innovators Act” aim to diversify the inventor pool, but the implications of the racial gap remain largely unexplored. This paper aims to document how the racial composition of scientists affects the direction and rate of medical research and innovation.
Special Settings
Dossi (2023)
The study leverages various data sources, including US patents, medical research articles, clinical trials, and research grants, combined with the racial distribution of last names from the US Census. This unique combination allows for the identification of race-specific research patterns. The methodology is feasible due to the availability of comprehensive data on patents and the racial categorization of last names, enabling a novel analysis of racial disparities in innovation.
Theory
Dossi (2023)
The paper formalizes a general equilibrium Roy model incorporating racial frictions and endogenous occupational choice. It hypothesizes that the racial composition of scientists influences the direction of research due to homophily (preference for similarity), evolutionary advantages in ancestral environments, and socio-economic factors intertwined with race. The model predicts that removing barriers to entry for minority scientists would increase the overall number of inventors and improve the quality of innovation.
Data
Dossi (2023)
The study uses data from the USPTO, PubMed, ClinicalTrials.gov, and NIH research grants, merged with the 2010 US Census data on last names. This dataset allows for the analysis of patent applications, granted patents, research articles, and clinical trials, with a focus on racial disparities. Key variables include patent acceptance rates, impact of granted patents, and the racial composition of clinical trial participants and research subjects.
Key Findings
Dossi (2023)
- Black scientists are more likely to research diseases prevalent among Blacks, while White scientists focus on diseases common among Whites.
- Black scientists are three times more likely to design clinical trials with Black participants and twice as likely to publish articles focused on Black individuals.
- Patent applications from Black inventors are granted less often, but their granted patents have a higher impact.
- An exogenous increase in HIV-related mortality among Blacks led to a significant rise in Black scientists researching HIV.
- Removing barriers for minority inventors would increase the overall number of inventors by 1 percentage point, a 10% increase from the baseline.
Interpretation / Policy Implications
Dossi (2023)
The findings suggest that racial disparities in the scientific workforce lead to a misallocation of talent and a narrower scope of research focus, which may hinder overall innovation and societal welfare. Policies aimed at reducing barriers for minority scientists, such as improving access to resources and addressing discrimination in the patenting process, could enhance the diversity and quality of innovation. Additionally, encouraging minority participation in science and innovation could lead to more inclusive research addressing the health needs of underrepresented populations, ultimately benefiting society as a whole.
Emperical settings
Dossi (2023)
Additionally, the analysis examines the differential acceptance rates of patent applications and the impact of granted patents, controlling for various factors such as technology class, assignee fixed effects, and state of residence. The study also employs a general equilibrium Roy model to estimate the potential increase in the number of inventors if barriers were removed, providing a comprehensive empirical framework to understand the racial disparities in science and innovation.
