IZA Newsroom

IZA – Institute of Labor Economics

Mark Fallak

Claudia Goldin receives 2023 Nobel Prize in Economic Sciences

by

Harvard University’s Claudia Goldin has been honored with the Nobel Prize in Economic Sciences for her groundbreaking research on the role of women in labor markets. IZA extends its heartfelt congratulations to Goldin, who has been a member of the global IZA research network since 2011.

Simon Jäger, Director of IZA, describes the award for Claudia Goldin as “outstanding and well-deserved.” According to Jäger, the prize recognizes the life and work of a researcher who has produced groundbreaking insights into the working world of the last 200 years, especially the role women play in the labor market and how this role has evolved.

Goldin has explored, for example, “how the participation of women in the workforce developed during industrialization and thereafter.” Recently, she has examined the roles of discrimination, social norms, and flexibility in relation to the persistent gender pay gap. Goldin has been able to clearly document this using historical data, “which she tracked down in a kind of detective work,” says Jäger.

The awarding of the Nobel Prize to Claudia Goldin not only acknowledges her outstanding contributions but also serves as a reminder of the importance of her research in unraveling the complexities of gender dynamics in the labor market.

In 2016, Claudia Goldin had already received the IZA Prize in Labor Economics. Her upcoming IZA Prize book, “An Evolving Force – A History of Women in the Economy,” will provide further insights into her impactful work.

How exposure to artificial intelligence affects worker well-being

by

[Editor’s note: The following text is based on a ChatGPT-edited version of a summary provided by the authors of the study.]

Generative AI, exemplified by ChatGPT, has ignited widespread interest, casting a spotlight on its potential to redefine our daily lives and reshape cognitive and professional processes. This transformative force may extend to economic growth, healthcare, safety, and transportation, while simultaneously breaking down barriers to information access, education, and training. Amidst various studies exploring AI’s influence on labor markets and productivity, a critical gap remains in understanding its effects on workers’ well-being and mental health over time.

In response, Osea Giuntella, Johannes König, Luca Stella conducted a groundbreaking study, leveraging longitudinal data from the German Socio-Economic Panel (SOEP). Their focus? To investigate the profound implications of AI technology adoption in the workplace on workers’ well-being, economic concerns, and mental health. With a strategic analysis of occupational exposure and a meticulous study design, the authors reveal a nuanced narrative of the AI revolution’s impact on the German workforce.

The findings indicate a divergence in life satisfaction since the significant uptick in AI adoption in 2015, with exposed workers reporting lower levels. Job satisfaction among AI-exposed workers has witnessed a notable decline, coupled with heightened concerns about job security and personal economic situations. However, intriguingly, the study unveils no discernible impact on workers’ mental health, anxiety, or depression.

This study serves as an initial exploration into the evolving landscape of workers’ perceptions during the AI revolution’s transitional phase. Understanding these diverse impacts is paramount for shaping labor market policies that balance innovation with the protection of employee well-being. As we navigate this era of technological change, policies safeguarding vulnerable workers, fostering effective retraining programs, and providing support during transitions emerge as crucial measures to mitigate the potential adverse consequences of automation technologies on worker welfare.

Tax flight by the super-rich: Does it really happen?

by

The ongoing concern regarding the increasing concentration of wealth frequently prompts calls for higher taxes on wealth. A common worry associated with such tax proposals is the potential for wealthy individuals to evade them by relocating internationally. A new IZA discussion paper by Arun Advani, David Burgherr, and Andy Summers analyzes the reform of capital taxation in the United Kingdom and provides quantitative insights into the actual magnitude of such responses.

Exploring a unique aspect of the UK tax system

The authors leverage a distinctive feature of the UK tax system dating back to the colonial era, which allows a special tax regime for individuals whose “permanent home” or “domicile” is not in the UK. Under this regime, such individuals can opt not to pay tax on returns to capital that come from wealth located outside the UK, such as dividends from foreign shareholdings and foreign bank interest. This exemption applies even if they reside and work in the UK for the entire year, a practice that many individuals seem to follow.

A reform triggering migration effects

In 2015, a reform was announced that terminated access to this special tax regime once individuals had spent 15 years in the UK, with the reform taking effect in 2017. This reform offers an excellent opportunity to examine the migration effects of capital taxes for several reasons: (1) it represented a substantial change, resulting in an 18% reduction in post-tax income; (2) it created variation in the tax rate for the same individuals over time; and (3) it did not immediately impact a group of otherwise very similar individuals who had spent slightly less time in the UK.

Understanding the impact on emigration

The study’s findings reveal relatively modest impacts on emigration. The baseline migration rate among long-term residents was approximately 4%, and the reform increased this rate by about 4.6 percentage points. While this constitutes a substantial relative change in migration rates, it translates to an implied elasticity of emigration in response to a 1% decrease in post-tax income of only 0.26. Notably, older individuals and those who paid relatively little tax before the reform, deriving most of their income from overseas capital returns, exhibited greater responsiveness. However, these groups represent a minority among the population benefiting from the reform.

Implications for capital taxation and labor market policy

These results bear significance not only for our comprehension of responses to capital taxation but also for the design of labor market policy. Many countries are increasingly introducing tax incentives for migrants. This study’s findings suggest that such tax competition is likely to incur high deadweight costs since a relatively weak migration response results in revenue costs outweighing the additional tax revenue generated by the policy.

The long shadow of China’s one-child policy

by

In 1980, China introduced the coercive one-child policy (OCP) which, as its name suggested, limited couples to having only one child. This was introduced by the Chinese government over concerns about overpopulation and economic growth. This policy lasted for more than three decades. The OCP was maintained through a system of fines, financial incentives, and intense propaganda.

Recent generations of Chinese women have grown up subject to this policy. Did exposure to such a restrictive policy and its influence in shaping a small-family culture permanently alter the subsequent fertility of Chinese women even after they were in an environment that did not restrict their fertility?

To answer this question, Siyuan Lin, Laura M. Argys and Susan L. Averett use data from the Census Bureau’s American Community Survey to examine the childbearing decisions of a sample of Chinese women who migrated to the US after growing up for varying lengths of time exposed to the OCP. In the US, these migrants are free to have as many children as they want.

The new IZA discussion paper finds that Chinese women who immigrated to the US and were exposed to the OCP for a longer duration have significantly fewer children compared to similar women who were less exposed to the OCP.

Fertility preferences shaped by small-family culture

This is the case even when accounting for the woman’s natal family size in an attempt to disentangle the effect of growing up in a smaller family from the impact of OCP exposure to a small-family culture in shaping fertility preferences. The results are stronger for currently-married women and hold when using a sample of only Chinese women and when introducing a control group of women from other Asian countries/regions.

The negative impact of exposure to the OCP on fertility is evident even after controlling for patterns of fertility assimilation of migrants and education levels that might have increased due to the OCP. The authors’ estimates suggest that exposure to the OCP from age 6-30 results in about 7 percent fewer children than similar women who were not exposed to the OCP.

Dramatic consequences of strong son preference

The implementation of the OCP, while reducing fertility, has also had unintended consequences. Data from the World Bank show that the old-age dependency ratio, a critical indicator of strain faced by the younger generation to support the older generation, has risen dramatically in recent years. The OCP, enacted in a country with a strong son preference, has also been shown to lead to a skewed sex ratio resulting in a large proportion of unmarried men. In addition to altering the future family prospects of these men, some evidence suggests this imbalance has also led to an increase in crime.

In response to concerns about the consequences of the low level of Chinese fertility and the recently declining population, the government enacted a two-child policy in 2016 that was quickly followed by a three-child policy in 2021.

New three-child policy unlikely to succeed

The finding that the duration of exposure to the OCP reduced subsequent fertility seems to be borne out by preliminary evidence from adopting the two-child policy in 2016. Despite eliminating penalties and encouraging families to have a second child through altered messaging, China’s TFR has continued to fall from 1.67 in 2015 to 1.28 in 2020. Clearly, fertility has not rebounded as policymakers had hoped, hence the enactment of the three-child policy in 2021.

The new research suggests that this policy has little chance of success since the vast majority of women currently of childbearing age in China were exposed to OCP messaging and penalties throughout most of their lives. It is unclear how long it will take to overcome decades of small-family cultural norms or if it is even possible, the authors write.

What we teach children about race and gender

by

Children’s beliefs about their abilities or place in society are known to be shaped at a young age by race- and gender-based messages such as those implicitly conveyed in educational materials. As a result, it is vital that we understand how race and gender are portrayed in the content we present to our children.

A recent IZA discussion paper (forthcoming in the Quarterly Journal of Economics) uses AI tools that convert images into data on representation of skin colors, race, gender, and age in a century of influential children’s books. This work contributes to our knowledge of what messages are conveyed in the content we use to teach and entertain our children. It also expands the set of tools available for social scientists to measure representation in a variety of contexts.

The authors of this study focus on award winning children’s books which are divided into two collections: “Mainstream” books considered to be of high literary or artistic value, and “Diversity” books selected because of how they center experiences of specific underrepresented identity groups in addition to their high literary value. They find that library checkouts for books in the Mainstream collection increase dramatically after receiving an award, implying that children are more likely to be exposed to the content of these particular stories.

The natural first step when measuring representation in images is to apply face detection to identify all depicted faces. However, off-the-shelf face detection models have a significant drawback; while off-the-shelf methods are an excellent way to detect faces in photographs, they perform poorly when applied to illustrations. Since many of the images found in content targeted towards children contain illustrations, this presented a problem. To overcome this, the authors construct a novel training dataset to build a model which can better detect faces in both photographs and illustrations. This model allowed them to detect 2.5 times more faces in their sample of children’s books than they would have using off-the-shelf face detection methods.

Applying their face detection model along with other computer vision tools such as skin segmentation, the authors find that faces depicted in the Mainstream collection are lighter on average than faces depicted in the Diversity collection. This is true even when conditioning on a face’s predicted race (e.g. faces predicted to be Black are lighter on average in the Mainstream collection than faces predicted to be Black in the Diversity collection). Perceptual Tint is used as a measure of how light or dark a skin color is on a scale of 0-100.

Another key finding is that relative to their growing share of the U.S. population, Black and Latinx people are underrepresented in these same books, while White males are overrepresented. Female representation in the text of these stories has grown from about 25% of all gendered terms and names to almost parity in recent years. However, even though females are increasingly present in the text of these stories, they appear less often in text than in images, suggesting greater symbolic inclusion in pictures than substantive inclusion in stories.

In an effort to understand the determinants of demand for representation in children’s books, the authors use data on children’s book purchases to understand who is purchasing books with different levels of representation. They find that people are more likely to purchase children’s books which contain representations of their own identities or the identities of their children. For example, purchasers who identify as Black or as Latinx are more likely to buy books that contain pictured characters with darker skin color, on average, than purchasers who identify as White. They also find that on average, purchasers who have a son purchase books with a lower percentage of female names, as compared to purchasers that have no children.

By merging the location of children’s book purchasers to data on local beliefs from the Cooperative Election Study, the authors show that a greater number of purchases of books from the Diversity collection is associated with a smaller proportion of individuals who believe that undocumented immigrants should be deported and a larger proportion of individuals who believe that White people in the U.S. have certain advantages because of the color of their skin.

Overall, this study uses AI tools to reveal enduring inequality in representation of skin colors, race, gender, and age within influential children’s books. It also establishes evidence that demand for representation in children’s books is related to consumers’ identities as well as their personal and political beliefs. These results suggest how the demand for representation may be a channel through which beliefs about race and gender could propagate across generations through the messages contained in the books parents purchase for their children. The findings in this study generate hypotheses that can motivate and inform subsequent research on the causes and consequences of representation in children’s books.

Democrats do not benefit electorally from popular minimum wage increases

by

Social scientists have shown that voters tend to reward politicians and governments that directly put cash in their hands. However, this kind of direct redistribution through the tax code, while explicitly involving government, is not the only means by which governments may seek to change the income distribution. Regulation of private market transactions may have a similar, if indirect, effect, implicitly redistributing via so-called “pre-distribution” policies. Wage floors, in particular, are implemented with the clear goal of redistributing pre-tax firm income to low-wage workers.

In the United States, polls consistently indicate minimum wage increases are broadly popular – and clearly associated with the Democratic party. A new IZA discussion paper by Emiliano Huet-Vaughn provides the first test of whether large minimum wage increases actually yield electoral gains for Democrats. For both federal and state races, the author finds no evidence that this is generally true using an event-study design and sub-national variation in minimum wages from the early 1990s to recent years. A null result is further confirmed when using a beneficiary-level political sentiment measure and difference-in-difference design.

Various explanations for the finding are explored and dispelled while newly collected survey evidence supports a salience, or inattention, mechanism. Specifically, voters are found to attend much less to a minimum wage increase than to an equivalently-valued direct cash transfer from the government. This suggests putting money in people’s hands may not be enough to receive political credit and that the directness of a transfer may itself matter.

Worker pay rose in U.S. chains that abandoned no-poaching clauses

by

Many US chains formerly prohibited one location from hiring another’s workers using contractual “no-poaching” clauses. When they abandoned this practice under legal pressure, worker pay rose by 4 to 6.6 percent. A new IZA discussion paper by Brian Callaci, Matthew Gibson, Sergio Pinto, Marshall Steinbaum and Matt Walsh investigates the wage impact using nationwide data from a wide array of US industries, from tax preparation to plumbing.

The paper takes advantage of an antitrust enforcement campaign undertaken by the Attorney General (AG) of Washington State. From mid 2018 through early 2020, the AG signed 239 agreements with individual chains, committing them to abandon no-poaching clauses.

Examples include McDonald’s (fast food), Jackson Hewitt (tax preparation), Expedia CruiseShipCenters (travel), European Wax Center (personal care), Hertz (car rental), and Weichert Real Estate Affiliates. The agreements bound chains not only in Washington State, but nationwide.

The result was an increase in labor market competition, with different locations inside the same chain now free to pursue the same workers. The authors estimate the effects of this increased competition on worker pay by comparing chains that agreed to abandon no-poaching clauses to other employers, before and after the agreements with the AG.

Using job vacancy data from Burning Glass and worker pay reports from Glassdoor, they find that abandonment of franchise (chain) no-poaching clauses increased worker pay by 4 to 6.6 percent on average. Pay increases were larger for salaried than for hourly workers. In addition, the study documents spillover pay increases for chain employers that did not sign agreements with the AG, but that compete for labor with chains that did.

Fig. 1C: Event study estimates, inverse sample (Burning Glass data)
Fig. 1D: Event study estimates, inverse sample (Glassdoor data)

These results suggest that low-pay labor markets are not almost perfectly competitive, as is often assumed. More broadly, the paper points to the importance of antitrust policy and enforcement for worker welfare.

Despite the efforts of the Washington State AG, no US court has yet ruled that franchise no-poaching clauses are illegal, and they are still used by some chains. To date private litigation attempting to recover damages for past use of franchise no-poaching clauses has not succeeded.

Woke jobs after Dobbs?

by

In an increasingly politically polarized social environment, companies are more frequently engaging in politically and socially controversial issues, including gun control, LGTBQ issues, climate change, and racial equality. Such engagement may serve as a meaningful signal of company culture for current or prospective workers who support such values, but could alienate those with differing viewpoints. This raises an important question: What are the consequences of employers making socio-political statements?

Following the June 24, 2022, Dobbs v. Jackson Supreme Court ruling, which overturned the federal right to abortion established in Roe v. Wade, hundreds of employers publicly announced policies covering out-of-state travel for abortions and related care. In a new IZA discussion paper, Pawel Adrjan, Svenja Gudell, Emily Nix, Allison Shrivastava, Jason Sockin and Evan Starr examine the labor market implications of these announcements.

After developing a comprehensive list of the companies that made these public declarations, the authors asked how employees and job seekers reacted by studying the evolution of reviews on Glassdoor, focusing on 6.5 million reviews submitted from 2019 through the first half of 2023, as well as 3 billion job seeker clicks on Indeed during this period.

To estimate the impact these announcements had on employees, the researchers compared changes in job satisfaction within companies that made these announcements versus companies that did not. To ensure the non-announcing companies are as comparable as possible, they took each announcing company and identified the top 20 other companies that workers most commonly click on during a search session on Indeed. As an example, for the company Starbucks, the set of similar employers contained a number of fast-food service chains, but also retail stores.

More clicks on job postings

According to the analysis of the Indeed data, firms that announced supportive reproductive care policies saw an 8% increase in clicks on their job postings compared to similar jobs at similar employers that made no announcement. Because trends were similar for both groups of employers prior to the Dobbs decision, but diverged immediately afterwards, findings are attributable to these public announcements of support themselves. Higher job seeker interest was concentrated in Democratic-leaning states and in typically female-dominated jobs in states where abortion became illegal after Dobbs.

Given that employee sentiment is a strong predictor of turnover, job satisfaction is another key outcome for companies to care about. If sentiment improves following these public statements, then retention within the company might be expected to improve as well. If, however, workers respond adversely to their companies taking a strong socio-political stance, then this could be a precursor to existing employees leaving the company.

Lower employee satisfaction

The analysis of the Glassdoor data focuses on the 1-5 star ratings workers provide about senior management and culture. Directly after the announcements, there is a sharp and statistically significant decline of about 0.2 stars in employee satisfaction with management and culture for these companies that persists well into the post-announcement period.

This translates to an 8% decline. Since satisfaction falls, the authors anticipate some employees may leave the company. Indeed, they observe that reviews for announcing employers are 4% more likely to be written by former employees after these announcements.

The increase in dissatisfaction is concentrated among reviews for male-dominated jobs, suggesting that men dislike these announcements more than women. The free-response text workers write in their reviews provide suggestive evidence that this decline might be driven by misalignment of political views: The word “woke” shows up about 325% more often in the ‘Cons’ section of reviews (where workers detail the negative aspects of their workplaces) for announcing companies compared with those for non-announcing companies.

Higher wages offered

Using wage data from Indeed job postings, the authors also found that pay at firms that did make post-Dobbs announcements increased 4% relative to similar jobs at non-announcing firms, even on top of the potential costs of offering additional healthcare benefits. This wage increase is separate from, and cannot explain, the increased job seeker interest in postings from announcing firms.

The study also revealed that the increase in posted wages among announcing firms was larger at firms that experienced deeper declines in senior management ratings. This suggests, although it does not prove, that at least some companies that made post-Dobbs announcements and received negative feedback from existing employees may have offered bigger raises to offset deteriorating satisfaction with the firm’s political stance.

These findings highlight the challenges employers face when navigating politically polarizing issues. While the announcements caused current employees, especially those in male-dominated jobs, to give their companies lower evaluations, job seekers expressed more interest in working for them, especially those in more Democratic-leaning states. This suggests that even when companies wade into divisive issues, this can help them hire workers whose political views align with those of the company.

For a more detailed summary, see also the Indeed Hiring Lab.

What we associate with first names and how it helps explain discrimination

by

Names are often one of the first things we learn about people. They can influence our first impression and how we process subsequent information about them. Names can also serve as the basis for discriminatory behavior. For example, research shows that resumes with distinctively Black names receive 50% fewer callbacks from employers compared to identical resumes with distinctly white names. In a new IZA discussion paper, Martin Abel and Rulof Burger aim to unpack underlying reasons contributing to name-based discrimination.

Figure 1: “Out of 10 people.., how many are… ?“

What do people associate with names?

Figure 1 shows race associations for 30 first names from a nationally representative sample comprising 1,500 participants from all 50 U.S. states. To accurately elicit beliefs about race, the researchers asked people how many out of 10 people belong to different groups and offered monetary incentives for correct responses. It turned out that when people perceive a name as Black, they also expect this worker to have lower levels of education, productivity, and non-cognitive skills. For example, perceiving a name to be Black (compared to white) is associated with a 47% decrease in the predicted likelihood of holding a master’s degree and a 46% decrease in perceived trustworthiness. Most of this racial bias persists when only looking at the effect of variation in race association between employers for the same name.

What’s in a name? Less than people think!

In a next step the authors compared the perceived racial gap with actual data obtained from 2,400 workers who performed the same task. The perceived gap in productivity of 25.2% is nearly three times larger than the actual gap of 9%. This finding aligns with recent studies that demonstrate people’s tendency to form beliefs about minorities that exaggerate between-race differences.

Name associations contribute to hiring discrimination.

Participants were then presented with pairs of workers and tasked with making hiring decisions. Each time they chose the more productive worker, they received a small payment. The results indicate that workers whose names are perceived as Black face a 30 percentage points hiring disadvantage. Notably, Figure 2 illustrates substantial variation in discrimination across different employer groups, with higher levels of discrimination observed among male, older, white, and conservative employers. In contrast, the study found no correlation between discrimination and employers’ educational attainment, the level of racial diversity in their zip code, or their prior experience with hiring.

Figure 2: Race discrimination across employer groups

 Employers use race as a decision heuristic

Hiring managers allocate very limited time to review each application, raising concerns that they may resort to heuristics or mental shortcuts based on race. Indeed, the researchers observe that greater differences in race perceptions among candidates lead to shorter decision times and increase confidence in their hiring decisions, even after accounting for variations in perceived productivity. Requiring employers to make decisions within a 2-second timeframe, which prompts instinctive decision-making, increases the racial gap in hiring by an additional 25%. Estimates from a neuro-psychological model of decision-making (DDM) show that under time pressure, people lower the information threshold and base their decision more on race perceptions, which are salient and thus easier to retrieve.

Who is most affected by time pressure?

Support for race-based affirmative action (AA) is the single biggest predictor of hiring discrimination (see Figure 2, top right). It also affects the responses to time pressure. Among white participants who oppose AA, the substantial hiring race gap of over 40% remains unaffected by time pressure. In contrast, those in favor of AA reduce discrimination from 34% to below 20% when provided with sufficient time. Results show this group overrides their instinctive, race-based assessments of workers and replaces them with more relevant productivity beliefs.

Name associations can hinder learning from new information and own memory

How much names affect people’s long-term economic outcomes remains an open question. Recent research suggests that names become less influential once people have access to more information about a person. However, other studies also indicate that (racial) biases can prevent people from collecting new information about others, leading them to refrain from interviewing a candidate, renting to certain individuals, or extending credits. The new study finds that biased beliefs can also impede individuals’ ability to learn from their own memory. Instead of seeking more nuanced information, many individuals overly rely on race associations when making decisions.

What does job applicants’ body art signal to employers?

by

Scientific research confirms the influence of attractiveness on job prospects. While previous studies mainly examined factors beyond personal control, such as physical beauty and obesity, a new IZA discussion paper focuses on the importance of another external characteristic: body art, namely piercings and tattoos. How does visible body art affect job opportunities, and what perceptions do employers hold regarding candidates with body art?

To address these questions, Stijn Baert, Philippe Sterkens and Jolien Herregods conducted a large-scale experiment involving U.S. recruiters. They asked the recruiters to evaluate fictitious job candidates whose AI-generated photos varied in whether they had visible small or large tattoos or piercings. Additionally, the photos (see examples below) were manipulated to compare the effect of body art with another aspect of physical appearance – obesity.

Recruiters were not only tasked with advising colleagues on whether to invite the candidates for job interviews but also to identify any potential stigmas associated with body art, including eagerness to work with such candidates, personality traits, and productivity drivers.

Differences in perceived personality traits and productivity

The findings revealed that recruiters perceive job candidates with body art as less pleasant to collaborate with, both in terms of recruiters’ personal preferences for collaboration and their perceptions of other employees’ and clients’ preferences for collaboration. Regarding personality, candidates with body art are seen as less honest, less emotionally stable, less agreeable, and less conscientious overall. Notably, the stigma of lower emotional stability only applies to men with body art. On the other hand, job candidates with body art are also seen as more extroverted and open to new experiences. However, in terms of direct productivity drivers, they are perceived as less manageable.

These perceptions ultimately result in lower hireability for men with body art but not for women. Compared to candidates who are obese, those with body art score better overall in terms of hireability and rated personality, and similarly in terms of rated taste for collaboration, but worse in terms of rated direct productivity drivers.

Considering these findings, body art wearers (and overweight individuals) and those assisting them in the labor market, such as public employment agency officers, should be mindful of the stigmas they may encounter when applying for jobs and the contexts in which they might face higher thresholds for success.

The authors suggest that individuals with body art would benefit from signaling qualities like honesty, stability, and conscientiousness during job applications. For instance, examples of conscientiousness, a personality trait found to be highly valued by employers, could be highlighted in the cover letter.