CC BY
1DOI: 10.29141/2658-5081-2024-25-1-2 EDN: OAVMWX JEL classification: J16, O32, M12
Melek gil Beykent University, Istanbul, Turkey
Yildiz Yilmaz Guzey Beykent University, Istanbul, Turkey
Technology and gender: Understanding the changing dynamics of female unemployment in the G7 countries
Abstract. Technology has profound effects on labour market dynamics. Numerous studies have highlighted the unique opportunities and challenges that technological change presents to specific demographic groups. The aim of the study is to examine the long-term impact of technological changes within organisations on female unemployment at a macro level. From the perspective of the contingency approach and economic growth theories, the research conducts a panel cointegration analysis employing CCE-MG and AMG long-term panel cointegration estimators. The data on the female unemployment rate, the percentage of R&D expenditure in GDP and the ICT patents in total patents in the G7 countries for 1985-2020 is sourced from the OECD statistics. The analysis indicates the presence of an effect of technological change on the female unemployment rate as well as the national variations in their relationship. In particular, in three countries out of seven (Germany, the UK, the USA), there is the relationship between technological advancements and female unemployment. For the most part, increases in the percentage of R&D expenditure and ICT patents augment the female unemployment, though in Germany a rise in R&D expenditure leads to a decrease in it. The results will contribute to understanding the impact of technology-driven changes in organisations on gender-based labour inequality. The research highlights the complexity of the impact of technological advancements on the female employment and underscores the need for shaping related government policies by considering each country's specific conditions.
Keywords: technology; female unemployment; digitalisation; R&D; gender inequality; contingency approach.
For citation: Ql M., Guzey Y. Y. (2024). Technology and gender: Understanding the changing dynamics of female unemployment in the G7 countries. Journal of New Economy, vol. 25, no. 1, pp. 26-49. DOI: 10.29141/2658-5081-2024-25-1-2. EDN: OAVMWX.
Article info: received August 15, 2023; received in revised form November 8, 2023; accepted December 1, 2023
Introduction
Technological advancements, in interaction with economic and societal structures, have evolved from the primitive agricultural era to the industrialisation process and, ultimately, to today's high technology stage. Currently, we are undergoing a period of the scientific and technological revolution that affects virtually all economic sectors. It directly impacts on various areas, from production organisation to the distribution of jobs and incomes.
The contingency theory aims to elucidate the interaction between environmental variables and changes in organisational structure [Lee, Luthans, Olson, 1982, p. 553]. This theory posits that the organisational structure will vary depending on factors such as the environment in which an organisation operates [Burns, Stalker, 1994], the technology employed, organisation size, and strategy [Chandler, 1962]. While technology and environmental factors are treated as independent variables, internal organisational factors are considered dependent variables.
Technological advancements shape the development of societies and significantly impact on the business environment. To sustain their presence in this variable environment, businesses must constantly adapt. Technological innovations are crucial factors that enhance production quality, create new industries, and job opportunities. The key to competition lies in adapting to innovations and technological changes. Businesses must manage technology and human resources harmoniously to achieve real competitive advantages. The implementation of new technologies within a business leads to substantial changes in jobs, organisational structures, and the workforce. To understand the effects of technological advancements, it is necessary to examine social interactions and interpretations. The social construction of technology (SCOT) approach has also considered issues such as how technologies are defined and utilized. It emphasises that adapting to technology begins with understanding how these technologies are constructed. Businesses should manage technology not only as a physical tool but also taking into account the social and organisational context of this technology. Technological change can lead to the reshaping of jobs, changes in organisational structures, and the evolution of skill requirements in the workforce.
As a result of technological advancements, some jobs disappear while new tasks emerge, leading to changes in the workforce structure within organisations. It is essential to handle the organisational dimension of the change arising from the use of new technologies with sensitivity. However, as is the case in any environment, attitudes, both positive and negative, toward innovation and change will naturally arise in business life. These attitudes can be observed at the individual level as well as at the group and even the entire organisational level. Resisting change can be considered a negative attitude. Among the reasons for resisting change are fears of technological unemployment, concerns about changes in jobs and wages, anxieties about acquiring new skills, and the lack of support for changes in social relationships and structure.
The management and organisation of businesses are closely related not only to internal dynamics but also to external factors. These external factors include economic conditions, competitive environment, legal regulations, and social dynamics. In this context, the success and sustainability of businesses are determined by how they adapt to these external factors and interact with them.
Female labour force operates within a socio-cultural framework, particularly shaped by societal and cultural factors. Society's gender norms influence the roles of women in the business world and their positions within organisations. These norms can shape women's entry into the labour market, promotion opportunities, efforts to combat wage inequality, and their experiences within workplace culture. Women experience a complex interplay of both their personal choices and the opportunities and barriers presented by society and organisations.
For organisations, fully assessing the potential of female labour force and combating gender inequality is closely associated with their ability to understand and adapt to the socio-cultural context. The position of women within the organisation depends on whether the organisation is fully open to the capabilities, contributions, and leadership potential of women. Therefore, the management and leadership of business entities must develop a more equitable and inclusive approach taking into account both their internal dynamics and the socio-cultural context in which female labour force operates. This is critical not only for female employees but also for the long-term success of organisations.
The impact of environmental adaptation and technology investments in organisational structure on factors such as gender inequality, women's career development, and job satisfaction may become more pronounced over time. Therefore, conducting long-term research to examine their effects on female employees would be highly appropriate.
The purpose of this study is to examine the long-term impact of technological developments or technology investments on female unemployment at the macro level. To achieve this goal, panel cointegration analysis covering the years 1985-2020 has been conducted for the G7 countries. In the analysis, R&D investments and the number of ICT patents are taken as indicators of technological investments in the G7 countries. The analysis section of the research presents preliminary test results, model outcomes, and interpretations. In the final section, the findings of the study are summarised, and recommendations are provided.
Theoretical background
Contingency approach. In the late 1950s and early 1960s, the first modernist organisation theorists attempted to determine the most appropriate organisational structure by using the principles proposed by classical organisation theory and neoclassical approaches. Organisation theory, until this period, had largely overlooked environmental conditions while focusing on organisational structure and management. Therefore,
this new approach aimed to take a step towards understanding how organisations interacted with external and internal factors and how they were shaped accordingly [Shafritz, Ott, Jang, 2015, p. 127].
Contingency theory views organisational structure as a structure that takes shape depending on various internal and external conditions. The organisational structure of a business is influenced by various external factors (market conditions, government policies, etc.) and various internal factors (the technology used, the quality of employees, the organisation's purpose, etc.) [Pennings, 1987, pp. 225-226]. In contingency theory, the most emphasised independent factors are the environment, technology, organisational size, and organisational strategy [Donaldson, 2015, p. 610]. Contingency theory is the most suitable approach for dealing with an uncertain environment due to its flexibility in formulating, developing, and implementing strategies [Prajogo, 2016, p. 243].
Many researchers have made significant contributions to the contingency theory. Let us discuss some of the most important in the context of the research problem.
Chandler [1962] argues that organisations need to develop strategies to adapt to environmental changes, and these new strategies require new organisational structures. As Fligstein sums up, Chandler emphasises that environmental changes form the basis for structural changes [Fligstein, 2008, pp. 242-245].
Burns and Stalker have stated that businesses facing market and technological changes will produce the most favourable results with an organic organisational structure [Burns, Stalker, 1994, pp. 96-98].
According to the research by Lawrence and Lorsch, the organisational structure varies depending on the market's demands and the rate of technological environmental changes [Lawrence, Lorsch 1967, p. 1].
Hopkins and Woodward empirically demonstrated the necessity of alignment between the technology used and the organisational structure for an organisation to be successful [Hopkins, Woodward, 1966, p. 288].
Newman [1971] has examined the relationship between organisations and technology. He asserts that how organisations utilize technology significantly determines the structure and processes of the organisation. Newman also states that the adoption of technology will lead to substantial changes in the workforce structure. Changes in the number of employees, alterations in the hierarchical structure of the organisation, and shifts in the expectations regarding the skill levels and education of employees are seen as outcomes of the organisation's technology strategies [Newman, 1971, pp. 22-23].
Organisations are in the effort to adapt to environmental conditions and technological developments. This theory assists organisations in understanding how they can adapt to environmental dynamics to gain a competitive advantage and sustain their viability. The contingency theory provides an important theoretical framework for how organisations can respond to changing conditions and optimise their structures.
Technology-organisation interaction. In the field of organisational science, technology began to be addressed as a variable with the development of the contingency approach. This issue was not considered by classical and behavioural theory thinkers. For instance, classical theorists like Weber and Fayol argued that organisations could be managed with the same principles without taking into account that technical systems of organisations could differ. Taylor was more concerned with the techniques used during task execution, and primarily focused on improving the operational efficiency of workers through these techniques and could not relate these techniques to the entire system [Hissom, 2009, pp. 6-8].
In the 21st century, businesses' main objectives are to boost efficiency and ensure sustainability. Embracing technological innovations is crucial for achieving these goals and remaining competitive in knowledge-based economies [Dodgson, Gann, Coopmans, 2008, p. 2]. In the 21st century, it is essential for every business to accurately understand the role of information technology not only within its own organisation but also within the broader society where the organisation needs to compete [Tansey, 2003, p. 174].
Effectively dealing with global competition necessitates a level of flexibility that many organisations lack. Fortunately, the advent of information technology (IT) provides a means for organisations to cultivate the required adaptability. IT has far-reaching implications for individual employees, teams, and entire organisations. It transforms various facets of an organisation, including its structure, products, markets, and manufacturing processes. Additionally, it enhances the value of intangible assets such as knowledge and competencies, fosters a more democratic work environment by facilitating information flow among employees, increases work flexibility by enabling remote work [Hellriegel, Slocum, 1976, p. 512].
Businesses must maintain flexible structures to adapt to changing environments, with technology influencing key aspects like hierarchy, workforce size, and qualifications [Mintzberg, 1989, p. 7]. Technology can reshape organisations in several ways, enhancing their competitiveness, efficiency, and adaptability. This includes digitizing processes, creating flexible structures, enabling remote work, utilizing collaboration tools, evolving workforce skills, and fostering data-driven decision-making. Tech-focused roles may require advanced digital skills, potentially reducing middle management positions, increasing the need for specialized, non-routine tasks, and prompting skill upgrades [Tansey, 2003, p. 174]. It is important for organisations to rapidly adapt to technological changes. Technological change inherently introduces complexity into any system. However, this complexity is often short-lived, followed by a period of stability. Changes in organisational structure and mechanisms are in line with tracking technological change [Hawthorne, 1978, p. 102].
The social status of women and women in the workforce. For many years, just like in societal life, in the economic life, men and women have been perceived
as representatives of different roles. There are various socio-cultural factors such as education, urbanisation, marital status, and economic conditions that influence women's labour force participation [Psacharopoulos, Tzannatos, 1989, p. 198]. Although women have been part of the labour force for a long time, they have mainly been employed in secondary roles in the workforce. Throughout history, women have faced numerous discriminatory practices in the business world. These discriminatory behaviours have hindered women's progress in their careers, access to career opportunities, and utilization of professional training opportunities [Stamar-ski, Son Hing, 2015, pp. 13-14].
Data related to women's presence in the workforce show that there have been certain turning points in different societies at different times. For instance, wars have been contributory factors that have changed women's labour force participation. During the First and Second World Wars, as men went to war, women became more involved in the workforce. During these periods, women made significant contributions to the production and service sectors by working in various industries [Goldin, 1991, p. 743].
The dramatic changes in women's social and economic status can only be understood by looking back into the past. In the early years of the 20th century, women working in factories and offices were mostly single women. In the late 1950s, married women were allowed to enter the labour force. Subsequently, married women paved the way for the rise of the modern career woman by engaging in jobs that required long-term education and offered real advancement opportunities [Costa, 2000, p. 102]. Globalisation has opened doors for women in the workforce. Advancements in technology, communication, and global trade have broadened women's job opportunities, even in emerging sectors like information technology. Additionally, studies reveal that the expansion of global textile supply chains has notably boosted female employment, particularly in Asian nations. For example, a study by Karim [2014, p. 88] revealed a significant increase in the number of women workers in textile factories in Bangladesh over the past decade. This serves as a concrete example of globalisation's impact on women's employment. At present, it is appropriate to say that we are going through a change process as significant as the one caused by the industrial revolution. Inventors, entrepreneurs, scientists, curious innovators, and amateur researchers will utilize all their advantages to create technologies that will amaze, entertain, and work on our behalf [Brynjolfsson, Mcafee, 2018, pp. 23-25]. Digitalisation is expected to not only change lifestyles and work patterns but also considerably influence women's work patterns and employment rates in societies [Sorgner et al., 2017, p. 22]. Furthermore, the changes in technology and the shifts in employees' preferences and expectations will necessitate businesses to focus on how to respond to these developments [Goldin, 2015, p. 33].
Technology and women. For individuals to regard themselves as integral to a genuine civilization, they must actively and conscientiously shape the era of technology
[Schwab, 2016, p. 106]. In the contemporary world, rapidly evolving technology is substantially impacting on the lifestyles and work modes of individuals. This impact has led to the emergence of an entirely new culture, seamlessly integrating technological advancements into daily life and professional spheres. However, this development is being accompanied by deep concerns about the potential of this new culture to exacerbate existing social and economic inequalities, including gender hierarchies. In the first half of the 20th century, the supply of educated workers exceeded technological changes, leading to lower inequality. However, in the second half of the century, technological advancements outpaced the supply of educated workers, resulting in a sharp increase in inequality [Goldin, Katz, 2007, p. 8]. Advancements in information and communication technologies have also raised the question of whether technology has a gender, in other words, whether there is an imbalance in the design and utilization of technology that favours one gender over the other [Varol, 2014, p. 220].
Assumptions about gender in technology's development, production, and use can be complex. The idea of technologies inherently having "masculine" traits is linked to the way tasks have been historically labeled as either male or female-dominated, evident in fields like medicine, industry, and information technology. It is rooted in the patriarchal structure of technology, reflecting society's predominantly male-oriented relationship with it. Since its inception, technology has been influenced by notions of masculinity, with men often associated with culture and science, while women are linked to nature and intuition, shaping technology accordingly. In many societies, women have learned through socialisation that computer technologies are predominantly associated with men. Gender and technology literature highlights how many technologies have been deeply influenced by societal gender norms. The perception of tech jobs as lacking in social interaction and work-life balance can be a barrier for women entering this field [Aksoy, 2012, s. 408]. Discussions and studies often revolve around whether technology offers new opportunities for female workers and how technology inherently influences the dynamics of technology-worker relationships in the professional sphere. These discussions emphasise the interplay between "technology" and "society", underscoring the prominent role of societal influence in technology's development and use. This context raises fundamental questions about whether science, which is believed to be socially shaped, truly shapes technology, thus influencing subsequent technologies, or if, over the long term, it is society that ultimately molds technology [Savci, 1999, ss. 125-126].
Limited access to vocational training can hinder women from developing their skills fully, pushing them towards the secondary labour force instead of the primary one in businesses. This situation often results in lower job security for female labour, with many occupying low-paid and precarious positions [Nielsen, 1990, pp. 233-234]. It is believed that tasks that require repetition and are not considered heavy labour will be performed more effectively by female workers. Light tasks are defined as those
that can be easily learned and do not require physical strength, emphasising the importance of employing female labour force in such roles. In an era where science and technology have taken the place of industrialisation, marking the beginning of a new phase in the labour market, the preference for female workers is attributed to women's tendencies to accept jobs that are precarious, low-paying, and irregular [Beneria, Roldan, 1987]. However, it is expected that as the demand for high skills increases, the perceived inadequacy of women's skills may become a significant barrier in the labour market. The impact of new technologies on women's employment can only be more clearly understood through research conducted at the macro level.
From technology's outset, a male-centric view linked masculinity to reason and femininity to emotion, fostering gender-biased associations. This culture positioned men in culture and science, women in nature and intuition. Despite tech's evolution, gender stereotypes endured. Computer culture did not break these stereotypes: women found computer tech male-dominated, battling anxiety and inadequacy. UNESCO aims to address digital skills disparities and technological biases through an emphasis on STEM education for girls. While digital skills are essential for future job prospects, gender inequalities persist, resulting in a digital divide that disproportionately impacts on girls and women1. Promoting female engagement in digital sectors, particularly in STEM fields, is crucial to narrowing this gap. UNESCO's 2017 report highlights a low enrollment of females in STEM disciplines, indicating potential challenges for future job prospects and labour market involvement2. Gender-based discrimination continues to hinder global STEM education at all levels. Ensuring gender equality in science and innovation is fundamental for achieving fairness. The inclusion of diverse perspectives is believed to enhance employment opportunities regardless of gender3. Despite women's increasing educational achievements, their representation in STEM fields remains inadequate [Castillo, Grazzi, Tacsir, 2014, p. 22].
Research on the impact of technological advancements on employment and unemployment. There are numerous studies that have examined the relationship between technological development and unemployment at the micro and sectoral levels. However, there is relatively limited research conducted at the macro level. R&D investments often serve as indicators of technological progress, typically supported by patent statistics. Findings obtained from micro-level research vary depending on the business or sector, thus, generalising becomes challenging. Therefore, conducting macro-level investigations becomes crucial for obtaining comprehensive results. While technological changes in organisations often occur rapidly, observing the effects
1 UNESCO. (2020). Global Education Monitoring Report 2020. A new generation: 25 years of efforts for gender equality in education. Paris. P. 16.
2 UNESCO. (2017). Cracking the code: Girls' education in science, technology, engineering and mathematics (STEM). Paris. P. 20.
3 European Commission. (2009). Gender stereotyping in Germany. Germany. P. 2.
of these changes requires long-term studies. Thus, conducting long-term research is essential to understand the impact of technological change on female employment.
Ger^eker, Ozmen and Mucuk investigated the relationship between R&D activities and unemployment across the G7 countries in the 1990-2016 period and found a bidirectional causality relationship between R&D and unemployment in Germany, France, Italy, and Japan [Ger^eker, Ozmen, Mucuk, 2019, s. 413]. Feldmann examines the relationship between technology and unemployment in industrialised countries and concludes that unemployment increases during the adoption phase of technology [Feldmann, 2013, p. 1099]. Abbasabadi and Soleimani find that digital technology indices initially increase unemployment and then decrease it after passing a certain threshold [Abbasabadi, Soleimani, 2021, p. 1]. Garcia-Murillo, MacInnes and Bauer also investigated the impact of developments in information and communication technologies on employment and drew attention to the differences across countries [Garcia-Murillo, MacInnes, Bauer, 2018, p. 1863].
Studies on technological advancements' impact on unemployment from a skill-based perspective reveal concerns over the potential disappearance of low-skilled or routine jobs. Weiss and Garloff examined 14 developed countries and found the negative effects of technology on the employment of unskilled workers [Weiss, Garloff, 2011, p. 811]. Frey and Osborne [2013] analysed the expected effects on US labour markets by examining the probability of computerisation, wages and education levels. They estimated that 47 % of total US employment could be at risk from new information technologies [Frey, Osborne, 2013, p. 41]. Bessen echoed these concerns, pointing to the potential for technology to eliminate or transform certain occupations [Bessen, 2019, p. 291]. Goldin and Katz have demonstrated the complex impact of technological progress on unemployment, as automation and productivity gains can reduce labour demand in some areas while creating new opportunities that require different skills [Goldin, Katz, 2008, p. 741]. Bertulfo, Gentile and de Vries explore the decline in labour demand that technological change will bring about in developing Asian countries [Bertulfo, Gentile, de Vries, 2019, p. 23]. Grigoli, Koczan and Topalova proved that technological advances negatively affect routine work participation, pushing workers out of the labour force [Grigoli, Koczan, Topalova, 2020, p. 20]. Conversely, demand for non-routine cognitive-based jobs, which often require higher education, increased.
McClure revealed that emerging job technologies like robots and artificial intelligence disproportionately impact on specific workforce groups in the USA, with women, non-white individuals, and those with lower education levels exhibiting technophobia towards adaptation [McClure, 2017, p. 139]. The study indicated lack of information access and educational technology as sources of this fear. Acemoglu and Restrepo emphasised that rapid automation's progression, replacing tasks with robots, would lead to job and wage inequalities, particularly affecting women, non-white individuals, and those with lower education levels [Acemoglu, Restrepo, 2020, p. 356].
Robinson et al. highlighted inequalities stemming from technological developments, tied to lifestyle, gender, race, and class disparities [Robinson et al., 2016, p. 569]. Fossen and Sorgner considered technological advancements like digitalisa-tion and AI, revealing their significant labour market impact and emphasising gender-based inequalities, particularly affecting women in terms of job transformation [Fossen, Sorgner, 2018, p. 19]. Mehta, Awasthi and Mehta [2021] and Sorgner et al. [2017] examined technological advancements' effect on female employment, underscoring vulnerability of women in low-skilled jobs to automation, leading to unemployment risks in manufacturing and service sectors. Mehta, Awasthi and Mehta also emphasised enhancing women's digital literacy for expanded job opportunities [Mehta, Awasthi, Mehta, 2021, p. 427].
Materials and methods
The study aims to test the impact of technological advancements on women's employment and answer the following research questions: "Do technology investments impact on women's labour force participation?" and "Is the effect of technological investments on women's labour force participation diverse among different countries?".
Women's historical disadvantage in employment is examined alongside the transformative influence of technological advancements and investments on the business landscape, prompting inquiry into their impact on women's employment. A thorough literature review explores the intersection of technological developments, investments, and women's employment rates, revealing a spectrum of studies with evidence of both positive and negative effects on women's employment.
The study investigates the impact of technological advancements and investments on women's unemployment, specifically focusing on the G7 countries: Canada, France, Germany, Italy, Japan, the UK, and the USA. The research operates at a macro level to analysee the correlation between technological developments, technology investments, and women's unemployment. The G7 nations, accounting for about 65 % of global capital and representing approximately 45 % of worldwide production [Turker, 2018, s. 142], hold considerable significance for this study.
The study employed panel data analysis to investigate the relationship between technological developments, technology investments, and female unemployment rates in the G7 countries. The dependent variable was the female unemployment rate, while the independent variables included the percentage of ICT patents within total patents and the percentage of R&D expenditure within GDP. Annual data spanning from 1985 to 2020 were utilized for the analysis. Stata 17 and Eviews 12 software packages were used for analysis, with data sourced from the OECD official website1.
Table 1 displays the variables associated with the countries included in the panel analysis along with their abbreviations. The study further conducted logarithmic
1 OECD. https://stats.oecd.org/.
transformations and first-order differences on the variables, denoted by the prefix ln for logarithmic transformation and the symbol f for first-order difference.
Table 1. Variables used and their abbreviations
Variable Name Symbol
Dependent variable Long-term female unemployment rate W_Unemp
Logarithm of long-term female unemployment rate \nW_Unemp
First-order difference of the logarithm of long-term female unemployment rate flnW_Unemp
First independent variable Percentage of ICT patents in total patents ICT_Patent
Logarithm of the percentage of ICT patents in total patents lnICT_Patent
First-order difference of the logarithm of the percentage of ICT patents in total patents flnICT_Patent
Second independent variable Percentage of R&D in GDP RD_GDP
Logarithm of the percentage of R&D in GDP lnRD_GDP
First-order difference of the logarithm of the percentage of R&D in GDP flnRD_GDP
Research results
Cross-section dependence test. The results from cross-section dependence analysis inform the selection of suitable unit root and cointegration tests for the subsequent analysis stages. When cross-section dependence exists among units, the use of firstgeneration tests can yield deviations and unreliable outcomes. In such cases, second-generation unit root tests are recommended. Hence, cross-section dependence tests serve as a crucial initial step in test selection [Agir, Ozbek, Turkmen, 2020, s. 77].
The study covers the time span from 1985 to 2020, encompassing 36 years (T), and involves the G7 countries as the cross-sectional units (N). To assess cross-section dependence, the Breusch - Pagan LM test [Breusch, Pagan, 1980, p. 251], Pesaran's LMS test [Pesaran, 2004, p. 23], Baltagi, Feng, and Kao's LMBC test [Baltagi, Feng, Kao, 2012, p. 172], and Pesaran's CD test [Pesaran, 2004, p. 23] were employed. The test hypotheses are as follows:
H0: There is no cross-section dependence;
H1: There is cross-section dependence.
Rejecting H0 indicates that cross-section dependence exists among the units constituting the panel [Pesaran, 2004, pp. 23-24]. In scenarios where cross-section dependence is absent, first-generation unit root tests are employed; conversely, in the presence of cross-section dependence, second-generation unit root tests are employed [Hurlin, Mignon, 2006, p. 8]. The outcomes of the cross-section dependence tests are presented in Table 2.
The outcomes of the cross-section dependence tests in Table 2 are consistent with one another. With probability values below 0.05, the H0 implying the absence of cross-section dependence has been rejected. This indicates the identification of cross-section dependence (p = 0.000) across all G7 country series.
Table 2. Cross-section dependence test results for the G7 countries
Variable Test Statistic Standard deviation p-value
W_Unemp Breusch - Pagan LM 156.39 21 0.000
Pesaran scaled LMS 20.89 - 0.000
Bias-corrected scaled LMBC 20.79 0.000
Pesaran CD 6.59 0.000
RD_GDP Breusch - Pagan LM 198.69 21 0.000
Pesaran scaled LM 27.42 - 0.000
Bias-corrected scaled LM 27.32 0.000
Pesaran CD 5.81 0.000
ICT_Patent Breusch - Pagan LM 520.61 21 0.000
Pesaran scaled LM 77.09 - 0.000
Bias-corrected scaled LM 76.99 0.000
Pesaran CD 22.6 0.000
Testing for homogeneity in panel data. The choice of tests in panel data analysis is influenced by whether the units are homogeneous or heterogeneous. When analysing diverse time series of countries, variations in macro units due to distinct unit characteristics often lead to heterogeneity. In the analytical framework, the pooled mean group (PMG) estimator and the common correlated effects (CCE) estimator assume series heterogeneity and can be employed. Notably, the impact exhibited by the variables we are employing can significantly differ among various cross-sections. Hence, it is more appropriate to employ estimates that account for coefficient heterogeneity. The hypotheses evaluated by the slope homogeneity test are as follows:
H0: Slope coefficients are homogeneous;
Hf: Slope coefficients are not homogeneous.
In the context of this study, which examines the correlation between female unemployment and R&D as well as patent rates in ICT, the delta homogeneity test developed by Pesaran and Yamagata [2008, pp. 64-65], was employed to ascertain the variability of slope coefficients. The obtained results are presented in Table 3.
As seen in Table 3, the probability values of the homogeneity tests conducted are less than 0.05, leading to the rejection of the H0. It is concluded that the regression model for the G7 countries shows heterogeneous constant and slope parameters for all groups.
Table 3. Homogeneity test results
Test Statistic p-value
Ä 16.131* 0.000
Ä adj 17.140* 0.000
Note: The superscript * indicates that the slope coefficients are heterogeneous at the significance level of 1 %.
Panel unit root tests (stationarity analysis). Cross-sectional dependence exists among the scrutinised units, where modifications in one unit influence other units, therefore, second-generation unit root tests are utilized. In line with Yildirim, Mercan and Kostakoglu, the prominent second-generation unit root tests encompass the one developed by Bai and Ng, MADF by Taylor and Sarno, SURADF by Breuer, Mc-known, and Wallace, CADF by Pesaran, and PANKPSS by Carrion-i-Silvestre and co-authors [Yildirim, Mercan, Kostakoglu, 2013, s. 88].
For the variables utilized as dependent and independent factors in this study, previous assessments have established the presence of cross-sectional dependence. In instances of identified cross-sectional dependence, Pesaran's second-generation unit root test, denoted as the CADF test developed in 2006, is employed to assess the stationarity of the series. In this context, Pesaran [2007, pp. 275-279] expanded upon the standard Dickey - Fuller (DF) or augmented Dickey - Fuller (ADF) regressions within the panel unit root test, introducing the Cross-Sectionally Augmented Dickey - Fuller (CADF) test. The CADF test yields consistent outcomes for both individually augmented ADF (hereafter CADF) statistics and their simple averages. This method's reliability extends to cases with relatively small numbers of cross-sectional units (N) and time series dimensions (T), encompassing scenarios where T is either greater or smaller than N (T > N and N > T) [Pesaran, 2007, pp. 266-267]. After conducting the CADF test, which assumes that cross-sectional units are affected at different levels by time effects and considers spatial autocorrelation, the obtained test statistic values are compared with Pesaran's CADF critical table values to conduct the stationarity test. The hypotheses evaluated by the test are as follows:
H0: There is a unit root;
H1: There is no unit root.
If the CADF test statistic values are greater than the CADF critical values, the H0 is rejected. Table 4 presents the results of the Pesaran's CADF panel unit root test for female unemployment rates, R&D rates, and ICT patent rate variables.
When the test statistic values in Table 4 are compared with the critical values, it is observed that for fixed models, the level is non-stationary at the 95 % confidence interval, but it becomes stationary after taking the first-order differences. According to the results of the Pesaran CADF panel unit root test, the analysis performed by taking
the logarithm of variables shows that lnW_Unemp, lnRD_GDP, and lnICT_Patent data are stationary in their first-order differences at the 95 % confidence interval.
Table 4. Results of the panel unit root test
Variable Level Difference (variation) Result
t-bar Zt-bar p-value t-bar Zt-bar p-value
W_Unemp -1.442 -1.380 0.5960 - - - -
RD_GDP -0.290 -0.298 0.999
ICT_Patent -1.171 -1.158 0.8359
lnW_Unemp -1.310 -1.263 0.7372 -3.096 -2.706 0.000* I(1)
ln RD_GDP -0.499 -0.482 0.9994 -3.836 -3.1786 0.000* I(1)
ln ICT_Patent -1.429 -1.399 0.5687 -4.978 -3.624 0.000* I(1)
Notes: 1. The superscript * denotes that since probability is less than 0.05, H0 is rejected. 2. For fixed models, there are values associated with the variable. 3. The t-bar statistic of the t-bar test is the inverse normal test statistic of the Zt-bar test. 4. The critical values of the CADF table for fixed models are determined at levels of -2.57, -2.33, and -2.21 for error margins of 1 %, 5 %, and 10 %, respectively.
Cointegration test. Tests on the panel dataset confirm cross-sectional dependence among the G7 countries. The units in the dataset display heterogeneity, while the first-order differences of both the dependent and independent variables are identified as stationary in the analysis. Macro data time series often exhibit non-stationarity, potentially resulting in spurious regression. To mitigate this, differencing the series and conducting regressions is a common approach. However, such differencing might overlook long-term relationships in extended analyses. To address this concern, cointegration tests are recommended. These tests assess the presence of long-term equilibrium relationships among variables, enabling the estimation of such relationships and preventing loss of pertinent information [Pedroni, 2004, p. 597].
The current research employs the ECM panel cointegration test introduced by Westerlund [2007] to explore the existence of long-term relationships within the series. This test computes four distinct panel cointegration test statistics via the error correction model. Two of these statistics furnish group mean statistics, while the remaining two offer panel statistics [Westerlund, 2007, p. 712]. To ascertain the presence of a cointegration relationship among the series, the computed statistics from the Westerlund ECM test are juxtaposed with critical values. While evaluating the series comprising the panel, cross-sectional dependence is examined in relation to bootstrap values [Westerlund, 2007, p. 721]. The hypotheses evaluated by the test are as follows:
H0: There is no cointegration among the units;
Hf: There is cointegration among the units.
A probability value below 0.05 leads to the rejection of the H0. The outcomes of the Westerlund ECM cointegration test are presented in Table 5.
Table 5. Results of Westerlund ECM cointegration test
Dataset Test Statistic Asymptotic p-value Bootstrap p-value
Group G_tau -2.993 0.000* 0.000*
G_alpha -13.364 0.041* 0.022*
Panel P_tau -7.152 0.011* 0.031*
P_alpha -13.211 0.021* 0.000*
Note: Calculated for the fixed model.
According to Table 5, all four tests for the fixed model used in our study are statistically significant, both in terms of asymptotic and bootstrap distributions. Considering the presence of cross-sectional dependence, the results of the tests conducted using the bootstrap distribution should be considered. The objectives of the bootstrap method include constructing confidence intervals, minimising error predictions, reducing standard deviations, and making parameter estimates more reliable [Efron, Tibshirani, 1993, p. 15]. Since the tests conducted using the bootstrap distribution are statistically significant, the hypothesis stating no cointegra-tion among the variables is rejected. In other words, it is said that there is a cointegration relationship between the variables for at least one of the countries forming the panel.
Estimation of long-term cointegration coefficients using CCE-MG and AMG analysis. At this juncture, the conducted tests have uncovered the heterogeneity of the panel dataset, the presence of cross-sectional dependence among the countries, and the existence of a cointegration relationship. When these prerequisites are fulfilled, the estimation of the model's long-term cointegration vector is undertaken through the utilisation of the CCE (common correlated effects) estimator. When estimating the coefficients in the presence of a long-term cointegration relationship, the CCE estimator is consistent [Pesaran, 2004, p. 2]. The CCE model calculates the estimated regression coefficients separately for each cross-sectional unit in panel data analysis [Erataç, Baççi Nur, 2013, s. 222].
The model employs two distinct alternative estimators, namely CCE-MG (common correlated effects mean group) and AMG (augmented mean group), both designed to accommodate cross-sectional dependence, for estimating the long-term regression coefficients of the explanatory variables. CCE-MG and AMG serve as co-integration estimators tailored for heterogeneous datasets that acknowledge cross-sectional dependence. These cointegration estimators furnish insights into the direction and magnitude of the relationship between the variables. The outcomes of the CCE-MG and AMG long-term panel cointegration estimators, which illustrate the influence of R&D as a percentage of GDP on female unemployment rates in the G7 countries, are displayed in Table 6.
Table 6. CCE-MG and AMG long-term panel cointegration estimators for the impact of R&D on female unemployment
Country CCE-MG AMG
coefficient standard deviation p-value coefficient standard deviation p-value
Canada 0.033 0.301 0.91 0.255 0.310 0.41
France 0.464 0.592 0.43 0.555 0.503 0.27
Germany -1.189 0.474 0.01*** -1.545 0.402 0.00***
Italy 1.141 0.607 0.06* 0.905 0.432 0.481
Japan -0.225 0.595 0.70 -0.162 0.539 0.76
UK 0.691 0.451 0.126 0.686 0.359 0.05**
USA 0.888 0.495 0.07* 1.734 0.437 0.00***
Panel 0.342 0.332 0.303 0.113 0.471 0.51
Note: The superscripts *, **, and *** represent the significance at the 10 %, 5 %, and 1 % levels, respectively.
Significant variations in the impact of the R&D rate on the female unemployment rate have been identified at the country level through the utilisation of CCE-MG and AMG estimators. Notably, this association demonstrates statistical significance for Germany, Italy, the UK, and the USA. Alterations in the R&D rate in Germany exhibit a noteworthy impact on the female unemployment rate, with a 1 % increment in R&D leading to a substantial 1.18 % reduction in female unemployment (p < 0.01). Conversely, in Italy and the USA, a distinctive pattern emerges, showing a positive and statistically significant relationship at the 10 % level between R&D and the female unemployment rate. Within Italy, a 1 % augmentation in the R&D rate corresponds to a 1.14 % rise in female unemployment, while the USA experiences a 0.89 % increase. Evidently, the repercussions of augmented R&D rates do not exert a uniform impact on female unemployment across all countries. This diversity underscores the influence of multifarious factors such as a nation's economic composition, R&D domains, technological and industrial advancement, overall employment framework, and policy formulation. Upon comprehensive analysis encompassing all G7 nations in a panel setting, a significant link between R&D rate and female unemployment rate does not emerge. This underscores the notion that the effect of R&D on female unemployment manifests a heterogeneous structure across diverse countries.
Country-specific assessments of the relationship between R&D rates and female unemployment rates, conducted using the AMG cointegration estimator, exhibit significant associations for Germany, the UK, and the USA. In Germany, akin to the findings from the CCE-MG estimator, the relationship's direction remains negative. Specifically, a 1 % rise in R&D corresponds to a substantial 1.55 % decrease in female unemployment. Contrarily, in the USA, the relationship takes an opposing trajectory,
where a 1 % increase in R&D results in a 1.73 % escalation in female unemployment (p < 0.05). Divergent from the CCE-MG analysis, the AMG estimator underscores a positive and notable correlation for the UK. According to the AMG estimator's outcomes, a 1 % augmentation in the R&D rate in the UK translates to a 0.69 % upsurge in female unemployment (p = 0.05). The comprehensive panel analysis, however, does not substantiate any statistically significant relationship. Therefore, evaluations were made on a country-specific basis.
The CCE-MG and AMG long-term panel cointegration estimators, examining the impact of ICT patents' share in total patents on female unemployment rates, are presented in Table 7. When CCE-MG long-term cointegration estimators are calculated, no significant relationship is found between the number of ICT patents and the female unemployment rate, both at the panel level and country-specific levels. However, in the case of the AMG cointegration estimator, a significant relationship between the variables is observed for Germany and the UK (p < 0.05).
Table 7. CCE-MG and AMG long-term panel cointegration estimators for the impact of
ICT patent on female unemployment
Country CCE-MG AMG
coefficient standard deviation p-value coefficient standard deviation p-value
Canada 0.099 0.898 0.27 0.033 0.072 0.65
France -0.928 0.142 0.51 -0.127 0.118 0.28
Germany 0.389 0.331 0.24 0.554 0.279 0.04**
Italy -0.003 0.135 0.98 -0.083 0.092 0.36
Japan -0.207 0.357 0.56 0.046 0.336 0.89
UK -0.274 0.199 0.17 -0.304 0.138 0.02**
USA -0.302 0.368 0.41 -0.146 0.274 0.594
Panel -0.684 0.098 0.48 -0.766 0.074 0.29
Note: The superscripts *, **, and *** represent the significance at the 10 %, 5 %, and 1 % levels, respectively.
According to the results of the AMG long-run cointegration estimator, in Germany, a 1 % increase in the ICT patents' share leads to a 0.55 % increase in the female unemployment rate. In the case of Germany, while the relationship between the R&D rate and the female unemployment rate is negative, the relationship between the ICT patents and the female unemployment is positively significant. Similarly, a significant relationship between these variables is observed for the UK (p < 0.05). A 1 % increase in the ICT patents leads to a 0.30 % decrease in the female unemployment rate. While an increase in the R&D rate in the UK leads to an increase in female unemployment, an increase in the number of ICT patents leads to a decrease in female unemployment.
Table 8 presents the CCE-MG and AMG long-term panel cointegration estimators for the effect of R&D rate and ICT patent rate as a percentage of total patents on female unemployment rates.
When examining the CCE-MG long-term panel cointegration estimators, they are not statistically significant at the panel level. However, when evaluated at the country level, there is a statistically significant relationship between R&D rates and female unemployment for Germany (p < 0.05). In Germany, a 1 % increase in R&D rates leads to a 1.01 % decrease in the female unemployment rate. This suggests that R&D positively impact on female employment in Germany.
Table 8. CCE-MG and AMG long-term panel cointegration estimators for the effect of R&D
and ICT patent on female unemployment
Country CCE-MG AMG
variable coefficient standard deviation p-value variable coefficient standard deviation p-value
Canada RD-GDP 0.164 0.282 0.56 RD-GDP 0.096 0.251 0.70
ICT_Patent 0.124 0.927 0.18 ICT_Patent 0.025 0.074 0.73
France RD-GDP 0.071 0.574 0.90 RD-GDP -0.354 0.423 0.40
ICT_Patent -0.020 0.575 0.89 ICT_Patent -0.133 0.119 0.24
Germany RD-GDP -1.011 0.473 0.03** RD-GDP -1.549 0.379 0.00***
ICT_Patent 0.343 0.271 0.20 ICT_Patent 0.538 0.225 0.01***
Italy RD-GDP 0.893 0.654 0.17 RD-GDP 0.459 0.429 0.28
ICT_Patent -0.014 0.137 0.92 ICT_Patent -0.076 0.092 0.41
Japan RD-GDP -0.020 0.632 0.97 RD-GDP 0.026 0.610 0.97
ICT_Patent -0.205 0.396 0.60 ICT_Patent 0.039 0.365 0.915
UK RD-GDP 0.562 0.442 0.20 RD-GDP 0.595 0.344 0.08*
ICT_Patent -0.243 0.200 0.22 ICT_Patent -0.277 0.135 0.04**
USA RD-GDP 0.631 0.536 0.23 RD-GDP 1.191 0.488 0.01***
ICT_Patent -0.224 0.330 0.49 ICT_Patent -0.163 0.258 0.52
Panel RD-GDP 0.236 0.239 0.32 RD-GDP 0.162 0.337 0.63
ICT_Patent -0.412 0.086 0.63 ICT_Patent -0.92 0.62 0.14
Note: The superscripts *, **, and *** represent the significance at the 10 %, 5 %, and 1 % levels, respectively.
When analysing the AMG long-term panel cointegration estimators for the G7 countries, a statistically significant relationship is observed for Germany, the UK, and the USA (p < 0.05). However, no statistically significant relationship is found between these variables when considering the panel data of all G7 countries together.
For Germany, both R&D rates and ICT patent rates have statistically significant effects on the female unemployment rate. A 1 % increase in R&D rates reduces the female unemployment rate by 1.55 %, while a 1 % increase in ICT patent rates raises
the female unemployment rate by 0.54 %. These results are quite similar to the separate estimations obtained for each variable.
In the case of the UK, some differences are observed compared to Germany. A 1 % increase in R&D rates raises the female unemployment rate by 0.59 %, while a 1 % increase in ICT patent rates decreases the female unemployment rate by 0.28 %. The direction of the effects of the variables is similar to the results obtained from separate estimations.
In the USA, the CCE-MG long-term panel cointegration estimators show no statistically significant relationship between R&D rates and female unemployment rates (p > 0.05). However, the AMG long-term panel cointegration estimators reveal a statistically significant relationship (p < 0.05), indicating that a 1 % increase in R&D rates leads to a 1.19 % increase in the female unemployment rate.
Conclusion
The contingency theory posits that organisations are in an interaction with their environments, and because of this interaction, structural changes within organisations can occur. For organisations to maintain their competitive advantage, it is essential to adapt to technological changes. These changes can affect how organisations conduct their operations, deliver products and services, communicate, and carry out business processes. Consequently, these structural changes driven by technology can have significant implications for the workforce.
The G7 countries play a significant global role with their high-income economies and technological advancements. However, differences in countries' economic structures, rates of industrial transformation, gender equality policies, labour market discrimination policies, education levels, skill development, socioeconomic conditions, and work-life balance vary significantly. These disparities lead to variations in how technological changes affect women's employment opportunities across different countries. In many organisations, technology often replaces low-skilled jobs, increasing the demand for a highly skilled workforce. Thus, the extent to which technology is embraced in different countries can affect the balance between skilled and unskilled labour. Findings suggest that technology investments tend to disproportionately impact on low-skilled female workers, potentially leading to unemployment among women. Encouraging skill development and adaptability to technology among women is crucial. Existing research in the literature has demonstrated the potential impact of automation on employment and unemployment, particularly in the USA [Frey, Osborne, 2017, p. 254], Europe [Bowles, 2014], and Japan [David, 2017, p. 77]. The findings of our study align with previous research. It was found that at least in three countries out of seven (Germany, the UK and the USA) there is an effect of technological change on the female unemployment rate, though the nature of this effect varies.
In advanced economies, there is a decrease in low-skilled jobs but different outcomes in high-skilled employment. The ability to rapidly adapt to technological advancement and digitalisation is supported by a competent and skilled workforce. Therefore, conducting more comprehensive studies to examine the various effects of technological developments on female unemployment in the G7 countries is recommended. Precise data collection with different methodologies across labour force sectors is necessary. Evaluating gender equality policies in the G7 countries is crucial in measuring their impact on female employment in the technology sector and unemployment risks. Comprehensive efforts are required for education and skill development programs to enhance women's ability to adapt to technology and participate in the workforce. Initiatives to increase female participation in STEM fields are highly significant. Assessing the consequences of flexible working models on work-life balance and female workforce participation is vital. Finally, researching the cultural and societal influences on female workforce participation enriches the understanding of the effects of technological developments on female employment.
In conclusion, the impact of technological advancements on female employment is a complex issue that yields different results in different countries. Therefore, policies related to female employment should be tailored to each country's specific conditions, and international collaboration should be encouraged. Promoting women's education, skill development, creating gender-neutral working environments, and enforcing legislation will contribute to gender equality and justice in female employment.
References
Abbasabadi H. M., Soleimani M. (2021). Examining the effects of digital technology expansion on Unemployment: A cross-sectional investigation. Technology in Society, vol. 64, 101495. https://doi.org/10.1016Zj.techsoc.2020.101495.
Acemoglu D., Restrepo P. (2020). Unpacking skill bias: Automation and new tasks. AEA Papers and Proceedings, no. 110, pp. 356-361. https://doi.org/10.1257/pandp.20201063.
Agir H., Ozbek S., Turkmen S. (2020). The relationship between financialization and income inequality within the framework of the financial Kuznets curve approach: An econometric estimate on E7 countries. Bilgi Ekonomisi ve Yonetimi Dergisi = Journal of Knowledge Economy and Knowledge Management, vol. 5, no. 2, pp. 71-84. https://dergipark.org.tr/en/ pub/beyder/issue/58428/833969. (In Turkish)
Aksoy B. (2012). Information technologies and new labour relations. Ege Akademik Baki§ = Ege Academic Review, vol. 12, no. 3, pp. 401-414. https://dergipark.org.tr/tr/download/ article-file/559248. (In Turkish)
Baltagi B. H., Feng Q., Kao C. (2012). A Lagrange multiplier test for cross-sectional dependence in a fixed effects panel data model. Journal of Econometrics, vol. 170, no. 1, pp. 164-177. https://doi.org/10.1016Aj.jeconom.2012.04.004.
Beneria L., Roldan M. (1987). The crossroads of class and gender. Industrial homework, subcontracting and household dynamics in Mexic city. Chicago: The University of Chicago Press. 211 p.
Bertulfo D. J., Gentile E., Vries G., de (2019). The employment effects of technological innovation, consumption, and participation in global value chains: Evidence from developing Asia (Asian Development Bank Economics Working Paper Series no. 572). 35 p. http://dx.doi. org/10.2139/ssrn.3343998.
Bessen J. (2019). Artificial intelligence and jobs: The role of demand. In: Agrawal A., Gans J., Goldfarb A. The economics of artificial intelligence: An agenda (pp. 291-307). Cambridge, MA: University of Chicago Press.
Bowles J. (2014). The computerisation of European jobs. https://www.bruegel.org/blog-post/computerisation-european-jobs.
Breusch T. S., Pagan A. (1980). The Lagrange multiplier test and its applications to model specification in econometrics. Review of Economic Studies, vol. 47, no. 1, pp. 239-253. htt-ps://doi.org/10.2307/2297111.
Brynjolfsson E., Mcafee A. (2018). The second machine age (Akilli Teknolojiler Devrinde $ali§ma, ilerleme ve Refah). istanbul. 336 p.
Burns T., Stalker G. M. (1994). The management of innovation. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780198288787.001.0001.
Castillo R., Grazzi M., Tacsir E. (2014). Women in science and technology: What does the literature say? (Technical note no. IDB-TN-637). New York: Inter-American Development Bank. https://repositorio.minedu.gob.pe/handle/20.500.12799/3376.
Chandler A. D. (1962). Strategy and structure: Chapters in the history of the American industrial enterprise. Beard Books. 480 p.
Costa D. L. (2000). From mill town to board room: The rise of women's paid labor. Journal of Economic Perspectives, vol. 14, no. 4, pp. 101-122. https://doi.org/10.1257/jep.14.4.101.
David B. (2017). Computer technology and probable job destructions in Japan: An evaluation. Journal of the Japanese and International Economies, vol. 43, pp. 77-87. https://doi. org/10.1016/j.jjie.2017.01.001.
Dodgson M., Gann D., Coopmans C. (2008). Playful technologies: Creativity, innovation and organization. Paper presented at DRUIDs 25th Anniversary Conference on Entrepreneur-ship and Innovation - Organisations, Institutions, Systems and Regions (Copenhagen, CBS, Denmark, June 17-20, 2008). 21 p.
Donaldson L. (2015). Structural contingency theory. In: International encyclopedia of the social & behavioral sciences (pp. 609-614). https://doi.org/10.1016/b978-0-08-097086-8.73110-2.
Efron B., Tibshirani R. (1993). An introduction to the bootstrap. New York: Chapman & Hall. 456 p. https://doi.org/10.1201/9780429246593.
Erata§ F., Ba§^i Nur H. (2013). The relationship between external debt and economic growth: The example of "emerging markets economies". Marmara Universitesi iktisadi ve idari Bilimler Dergisi = Marmara University Journal of Economic and Administrative Sciences, vol. 35, no. 2, pp. 207-223. https://dergipark.org.tr/tr/pub/muiibd/issue/494/4372. (In Turkish)
Feldmann H. (2013). Technological unemployment in industrial countries. Journal of Evolutionary Economics, vol. 23, no. 5, pp. 1099-2013. https://doi.org/10.1007/s00191-013-0308-6.
Fligstein N. (2008). Chandler and the Sociology of Organizations. Business History Review, vol. 82, no. 02, pp. 241-250. https://doi.org/10.1017/s0007680500062747.
Fossen F. M., Sorgner A. (2018). The effects of digitalization on employment and entre-preneurship. IZA - Institute of Labor Economics. 31 p. https://conference.iza.org/confer-ence_files/MacroEcon_2018/sorgner_a21493.pdf.
Frey C. B., Osborne M. (2013). The future of employment: How susceptible are jobs to computerisation? (Working paper of Oxford Martin Programme on Technology and Employment). 79 p. https://www.oxfordmartin.ox.ac.uk/downloads/academic/future-of-employment.pdf.
Frey C. B., Osborne M. A. (2017). The future of employment: How susceptible are jobs to computerisation? Technological Forecasting and Social Change, vol. 114, pp. 254-280. https:// doi.org/10.1016/j.techfore.2016.08.019.
Garcia-Murillo M., MacInnes I., Bauer J. M. (2018). Techno-unemployment: A framework for assessing the effects of information and communication technologies on work. Telematics and Informatics, vol. 35, no. 7, pp. 1863-1876. https://doi.org/10.1016/j.tele.2018.05.013.
Ger^eker M., Ozmen i., Mucuk M. (2019). The empirical analysis of the causality between R&D expenditures and unemployment: The case of G7 countries. Marmara Universi-tesi iktisadi ve idari Bilimler Dergisi = Marmara University Journal of Economic and Administrative Sciences, vol. 41, no. 2, pp. 413-431. https://doi.org/10.14780/muiibd.665104. (In Turkish)
Goldin C. (1991). The role of World War II in the rise of women's employment. American Economic Review, vol. 81, no. 4, pp. 741-756.
Goldin C. (2015). How to achieve gender equality in pay. Milken Institue Review. https:// www.milkenreview.org/print-edition/third-quarter-2015.
Goldin C., Katz L. F. (2007). The race between education and technology the evolution Of U.S. educational wage differentials, 1890 to 2005 (NBER Working Paper no. 12984). Cambridge, MA: NBER. 48 p. https://doi.org/10.3386/w12984.
Goldin C., Katz L. F. (2008). The race between education and technology. Belknap Press for Harvard University Press.
Grigoli F., Koczan Z., Topalova P. (2020). Automation and labor force participation in advanced economies: Macro and micro evidence. European Economic Review, no. 126, 103443. https://doi.org/10.1016/j~.euroecorev.2020.103443.
Hawthorne E. P. (1978). The management of technology. UK: Mc Graw-Hill Book Co. 208 p.
Hellriegel D., Slocum J. W. (1976). Organizational behaviour. St. Paul, MN: West Publishing Company.
Hissom A. (2009). Introduction to management technology. Management (Kent Staty University), vol. 26, no. 4.
Hopkins T. K., Woodward J. (1966). Industrial organization: Theory and practice. Administrative Science Quarterly, vol. 11, no. 2, pp. 284-289. https://doi.org/10.2307/2391249.
Hurlin C., Mignon V. (2006). Second generation panel unit root tests. Science Humaines et Sociales. https://shs.hal.science/halshs-00159842/.
Karim M. (2014). Effects of globalization on working women in the private sector of Bangladesh: A study of the garment industry. Thai Journal of Public Administration, vol. 12, no. 1, pp. 67-96.
Lawrence P., Lorsch J. (1967). Differentiation and integration in complex organizations. Administrative Science Quarterly, vol. 12, no. 1, pp. 1-47. https://doi.org/10.2307/2391211.
Lee S. M., Luthans F., Olson D. L. (1982). A management science approach to contingency models of organizational structure. Academy of Management Journal, vol. 25, no. 3, pp. 553-566. https://doi.org/10.2307/256079.
McClure P. K. (2017). "You're Fired," says the robot: The rise of automation in the workplace, technophobes, and fears of unemployment. Social Science Computer Review, vol. 36, no. 2, pp. 139-156. https://doi.org/10.1177/0894439317698637.
Mehta B. S., Awasthi I., Mehta N. (2021). Women's employment and digital technology: A regional analysis in India. Indian Journal of Human Development, vol. 15, no. 3, pp. 427-442. https://doi.org/10.1177/09737030211064138.
Mintzberg H. (1989). Mintzberg on management: Inside our strange world of organizations. New York: The Free Press. 418 p.
Newman W. H. (1971). Strategy and management structure. Academy of Management Proceedings, vol. 1971, no. 1, pp. 7-24. https://doi.org/10.5465/ambpp.1971.4980739.
Nielsen J. M. (1990). Feminist research methods: Exemplary readings in the social sciences. Boulder, CO: Westview Press. 262 p.
Pedroni P. (2004). Panel cointegration: Asymptotic and finite sample properties of pooled time series tests with an application to the PPP hypothesis. Econometric Theory, vol. 20, no. 3, pp. 597-625. https://doi.org/10.1017/S0266466604203073.
Pennings J. M. (1987). Structural contingency theory: A multivariate test. Organization Studies, vol. 8, no. 3, pp. 223-241. https://doi.org/10.1177/017084068700800302.
Pesaran M. H. (2004). General diagnostic tests for cross section dependence in panels (IZA Discussion Paper no. 1240). 42 p. https://docs.iza.org/dp1240.pdf.
Pesaran M. H. (2007). A simple panel unit root test in the presence of cross section dependence. Journal of Applied Econometrics, vol. 22, issue 2, pp. 265-312. https://doi. org/10.1002/jae.951.
Pesaran M. H., Yamagata T. (2008). Testing slope homogeneity in large panels. Journal of Econometrics, vol. 142, no. 1, pp. 50-93. https://doi.org/10.1016/j.jeconom.2007.05.010.
Prajogo D. I. (2016). The strategic fit between innovation strategies and business environment in delivering business performance. Journal of Production Economics, vol. 171, no. 2, pp. 241-249. https://doi.org/10.1016/j.ijpe.2015.07.037.
Psacharopoulos G., Tzannatos Z. (1989). Female labor force participation: An international perspective. The World Bank Research Observer, vol. 4, no. 2, 187-201. https://doi. org/10.1093/wbro/4.2.187.
Robinson L., Cotten S. R., Ono H., Quan-Haase A., Mesch G., Chen W., ... Stern M. J. (2016). Digital inequalities and why they matter. Information, Communication & Society, vol. 18, no. 5, pp. 569-582. https://doi.org/10.1080/1369118X.2015.1012532.
Savci Î. (1999). Gender and technology. Ankara Universitesi SBF Dergisi = Ankara University SBF Journal, vol. 54, no. 1, pp. 123-142. https://doi.org/10.1501/SBFder_0000001939. (In Turkish)
Schwab K. (2016). The fourth industrial revolution. Geneva: World Economic Forum. 192 p.
Shafritz J. M., Ott J. S., Jang Y. S. (2015). Classics of organization theory. Boston: Cengage Learning. 496 p.
Sorgner A., Bode E., Krieger-Boden C., Aneja U., Coleman S., MishraV., Robb A. M. (2017). The effect of digitalization on the gender equality in the G20 economies: Women20 study. Kiel: Kiel Institute for the World Economy. http://hdl.handle.net/10419/170571.
Stamarski C. S., Son Hing L. S. (2015). Gender inequalities in the workplace: the effects of organizational structures, processes, practices, and decision makers' sexism. Frontiers in Psychology, vol. 6. https://doi.org/10.3389/fpsyg.2015.01400
Tansey S. D. (2003). Business, information technology and society. London, New York: Routledge. 272 p.
Turker O. (2018). The impact of government spendings on foreign trade: A sample of G-7 countries. Sosyal Bilimler Dergisi = The Journal of Social Science, vol. 5, no. 27, pp. 141-156. https://sobider.com/DergiTamDetay.aspx?ID=4454. (In Turkish)
Varol S. F. (2014). A utopic approach to the relationship of women with technology: Cy-berfeminism. The Journal of Academic Social Science Studies, no. 27, p. 219-234. https:// dx.doi.org/10.9761/JASSS2463. (In Turkish)
Weiss M., Garloff A. (2011). Skill-biased technological change and endogenous benefits: The dynamics of unemployment and wage inequality. Applied Economics, vol. 43, no. 7, pp. 811-821. https://doi.org/10.1080/00036840802599933.
Westerlund J. (2007). Testing for error correction in panel data. Oxford Bulletin ofEconomics and Statistics, vol. 69, no. 6, pp. 709-748. https://doi.org/10.1111/j.1468-0084.2007.00477.x.
Yildirim K., Mercan M., Kostakoglu S. F. (2013). Test for validity of purchasing power parity: Time series and panel data analysis. Eski§ehir Osmangazi Universitesi iiBF Dergisi = Eski§ehir Osmangazi University Journal of Economics and Administrative Sciences, vol. 8, no. 3, pp. 75-96. https://dergipark.org.tr/tr/pub/oguiibf/issue/5713/76478. (In Turkish)
Information about the authors
Melek gil, PhD student in the Business Management Doctoral Program. Beykent University, Istanbul, Turkey. E-mail: melekcil201@gmail.com
Yildiz Yilmaz Guzey, PhD, Professor of Business Administration Dept. Beykent University, Istanbul, Turkey. E-mail: yildizguzey@beykent.edu.tr
© gil M., Guzey Y. Y., 2024
