Volume 26 Issue 1 Article 1 March 2024 Gender and Age Wage–Productivity Gaps in Intangible and Non- Gender and Age Wage–Productivity Gaps in Intangible and Non- Intangible Work Occupations Intangible Work Occupations Tanja Istenič University of Ljubljana, School of Economics and Business, Ljubljana, Slovenia Tjaš a Redek University of Ljubljana, School of Economics and Business, Ljubljana, Slovenia Daš a Farč nik University of Ljubljana, School of Economics and Business, Ljubljana, Slovenia, dasa.farcnik@ef.uni-lj.si Follow this and additional works at: https://www.ebrjournal.net/home Part of the Labor Economics Commons Recommended Citation Recommended Citation Istenič , T., Redek, T., & Farč nik, D. (2024). Gender and Age Wage–Productivity Gaps in Intangible and Non- Intangible Work Occupations. Economic and Business Review, 26(1), 1-12. https://doi.org/10.15458/ 2335-4216.1332 This Original Article is brought to you for free and open access by Economic and Business Review. It has been accepted for inclusion in Economic and Business Review by an authorized editor of Economic and Business Review. ORIGINAL ARTICLE Gender and Age Wage–Productivity Gaps in Intangible and Non-Intangible Work Occupations T anja Isteniˇ c, Tjaša Redek, Daša Farˇ cnik * University of Ljubljana, School of Economics and Business, Ljubljana, Slovenia Abstract The paper focuses on gender- and age-related wage–productivity gaps in intangible and non-intangible work occupa- tions using the 2017 Slovenian linked employer–employee microdata for privately owned rms. Comparing employees based on age, gender and occupation, our results show that, in general, there are wage gaps in favour of men, with the exception of individuals aged 50 or older who belong to the intangible capital group, where the wages of men and women are almost equal. There are also signicant wage gaps in favour of older workers, with the exception of women in non-intangible occupations, where those aged 30–49 and those aged 50C earn almost the same. Comparing the productivity of workers using value added decomposition method and based on age, gender and occupation, in general we nd that gender and age gaps are more pronounced. For example, women tend to be more productive than men, with the exception of men under the age of 30 in non-intangible work occupations. Similarly, older workers tend to be less productive than their younger counterparts, with the exception of women aged 30–49 compared to women under 30 in non-intangible work occupations. Moreover, age-related wage productivity gaps are higher for intangible than for non-intangible worker occupations. Keywords: Ageing, Gender, Wage gap, Productivity gap, Slovenia JEL classication: J24, J31 Introduction T he ageing of the population and the associated increase in the old-age dependency ratio are forc- ing governments to undertake various reforms to increase labour force participation, especially among those aged 55–64 and women (Directorate-General for Economic and Financial Affairs, 2021). In line with this trend, there is growing interest among re- searchers in analysing the impact of the inclusion of older workers and women in the labour market, in particular in what this means for individual and na- tional productivity. It turns out that the variability of productivity of individuals in different age groups is not clear. Some studies show a decline in the produc- tivity of individuals at older ages (e.g., the extensive literature review by Gabriele et al., 2018; Lee et al., 2018; Skirbekk, 2004). However, productivity declines are smaller or non-existent for older workers whose work tasks require experience or verbal skills (Skir- bekk, 2008). In addition, ageing may have a positive impact on labour productivity if older workers are employed in industries with a high ICT share of the capital stock (Lee et al., 2020). In addition, robotics technology can mitigate the negative effects of age- ing on productivity growth (Park et al., 2021). Thus, efcient resource allocation combined with lifelong learning can help maintain the productivity of older workers (Lee et al., 2022). Although younger workers tend to be paid below their marginal productivity, while older workers are paid above their marginal productivity (Lazear, 1979), the conclusion that older workers are paid above their productivity is also empirically unsupported in many cases. On the other hand, research shows that women are less productive than men, but they are also paid Received 21 July 2023; accepted 27 November 2023. Available online 5 March 2024 * Corresponding author. E-mail address: dasa.farcnik@ef.uni-lj.si (D. Farˇ cnik). https://doi.org/10.15458/2335-4216.1332 2335-4216/© 2024 School of Economics and Business University of Ljubljana. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/). 2 ECONOMIC AND BUSINESS REVIEW 2024;26:1–12 less (e.g., Ilmakunnas & Maliranta, 2005). The ques- tion arises, however, to what extent the gender pay gap results from the gender productivity gap and not from other dimensions, such as labour market dis- crimination (e.g., Castilla, 2008). Even though the traditional division of labour be- tween men and women persists, over time they have increasingly chosen similar elds of study and oc- cupations (England, 2010). Therefore, gender gaps in various areas such as work and decision making are gradually narrowing. Recent data shows that gender balance in decision making is visibly increasing. In 2017–2018, the area of power contributed 81% to the overall increase in the Gender Equality Index in the EU. It is also evident that in the EU, young men and women, as the most digitally literate generation, ben- et equally from their digital skills. However, at older ages, the gender gap is wider, with men being more digitally literate than women (European Institute for Gender Equality [EIGE], 2020). Although the information and communication technology (ICT) sector and the demand for ICT pro- fessionals are growing overall, and the gender gap in elds of study and occupations is narrowing, only 20% of graduates and 18% of employees in ICT- related elds are women in the EU. The gender gap is even more pronounced for scientists and engineers in high-tech sectors in the EU. Technological progress, in particular articial intelligence, offers a number of opportunities for European society, but at the same time poses some challenges. For example, the lack of gender diversity in the development of articial in- telligence technologies may lead to potentially unfair treatment of women in the future (EIGE, 2020). For the future, we therefore expect the working population to become older and to have a higher proportion of women. In addition, workers will tend to do less physical and more innovative work. In line with these demographic and economic trends, as well as the fact that previous literature shows that the wage–productivity gap needs to be analysed considering age and gender, plus the type of work performed by individuals, the main objective of this paper is to examine the age and gender differences in wages and productivity for non-intangible and intan- gible work occupations. Intangible work occupations represent innovative work types, including: research and development (R&D), organizational, and ICT capital work types (as dened in Piekkola, Bloch, Ry- balka, & Redek, 2021). The role of intangible capital on individuals’ productivity and wages is analysed based on Slovenian linked employer–employee data from 2017. The role of intangible capital on productivity has been previously measured for the Slovenian transi- tion by Prašnikar (2010), Verbiˇ c and Polanec (2014), Piekkola, Bloch, Redek, and Rybalka (2021), and Bavdaž et al. (2021), while wage and productivity differentials, considering age and gender, have been studied by Vodopivec (2014). However, the research has so far not addressed the wage–productivity nexus with respect to age and gender and the intangible capital. The novelty of this paper, besides using more recent data, is to estimate the role of intangible capital on the wage–productivity nexus, considering age and gender. In particular, we seek to answer the following research questions: (i) How large are gender- and age- related wage gaps in intangible and non-intangible work occupations? (ii) How large are the gender- and age-related productivity gaps in intangible and non- intangible work occupations? (iii) How large are the gender- and age-related wage-productivity gaps in intangible and non-intangible work occupations? Our analysis may also be useful for policymakers in other postsocialist countries that face rapid population age- ing and are generally characterized, like Slovenia, by relatively small gender gaps in labour income and rel- atively high dependence on public transfers at older ages, partly due to the relatively low retirement age of the elderly (see, e.g., Sambt et al., 2021). The paper begins with a comprehensive litera- ture review on age- and gender-specic differences in wages and productivity and their relationships. We then present the methodological framework and the data used, followed by a presentation of the results of the gender and age-related wage and productivity gaps and the gender and age-related wage–productivity gaps. The nal chapter makes conclusions. 1 Age–productivity–wage nexus A person’s productivity varies throughout their life for a number of reasons: the length of their work ex- perience, their cognitive and physical abilities, their motivation, the match between worker and task, etc. (Skirbekk, 2008). The assumption that productivity declines with age goes back to the human capital models of Ben-Porath (1967) and Mincer (1958). As people get older, they become more skilled and ex- perienced, so their productivity increases. However, productivity begins to decline after a certain period of prime working age (Mincer, 1958). As discussed in the following, although the initial theoretical work assumes that older workers are less productive, the empirical evidence does not show a clear relationship between productivity and the age of individuals. On the contrary, some empirical studies show that older workers may even be more productive than their younger counterparts. ECONOMIC AND BUSINESS REVIEW 2024;26:1–12 3 Changes in the productivity of individuals over their life cycles are examined in detail in Skirbekk (2004). The author reviews a number of empirical studies and concludes that previous studies mainly show a decline in the work performance (i.e., pro- ductivity) of individuals at older ages. A decline is particularly evident among individuals over the age of 50. Consistent with human capital models, Skirbekk (2008) shows, based on a review of nu- merous articles, that productivity increases at the beginning of working life, then stabilizes, and of- ten decreases with age. A decline in productivity at older ages is particularly evident in work tasks where problem-solving, learning, and speed are important. In contrast, productivity declines are smaller or ab- sent for those older workers whose work tasks are related to experience or verbal skills. Similar conclu- sions are also found in the comprehensive literature review by Mahlberg et al. (2006). However, Mahlberg et al. (2006) additionally claim that the productivity of older workers may be biased upwards, as older workers who choose to stay in the labour market are likely to be more productive than those who leave it. The negative relationship between productivity and individual age is also evident in some recent em- pirical studies, for example, Hu (2016) and Gabriele et al. (2018). Based on Chinese data, Hu points out that experience can be a barrier to increasing the pro- ductivity of older workers. This is especially true in today’s information age, where workers’ knowledge and experience are outdated. Moreover, based on Ko- rean rm-level data, Lee et al. (2018) show a negative relationship between the proportion of workers over 50 and value added per worker. As the authors claim, a similar situation has also been shown in most stud- ies based on European data. However, the authors continue to show a positive relationship between the proportion of workers over 50 and value added per worker in large manufacturing rms that are in a high-risk or growing environment. Moreover, based on the Austrian matched employer–employee panel, Mahlberg et al. (2013a) conclude that rm productivity is not negatively associated with the share of older workers. Next, the authors claim that older workers are not overpaid relative to their productivity. Mahlberg et al. (2013b) show that the relationship between age and productivity varies considerably by region and industry, with the latter being even more important. Similarly, using German data for the period 1986– 2006, Gordo and Skirbekk (2013) show that workers in their 50s have adapted well to technological change and have actually made larger gains in cognitively demanding tasks than younger workers in their 30s. A meta-analysis of 418 empirical studies examining the consistency of common stereotypes about older workers (i.e., their lower motivation, lower willingness to participate in training and career development, lower willingness to change, and lower condence) shows that the only stereotype consistent in past research is that older workers are actually less willing to participate in training and career development (Ng & Feldman, 2012). On the other hand, Skirbekk (2004) emphasizes that lower productivity contrasts with the empiri- cally observed wage increase at older ages. Therefore, to better understand age differences in labour in- come, researchers should go beyond age differences in productivity—it is also necessary to consider in- stitutional factors and/or market rigidities. This is consistent with Lazear’s (1979) “alternative theory,” according to which young workers are paid below their marginal productivity, while older individu- als are paid above their marginal productivity. This also explains why the aging population poses a ma- jor threat to the overall labour force. Reducing the wages of young workers below their marginal pro- ductivity will not be enough to cover the wages of older workers above their marginal productivity. Thus, population aging must lead to a reduction in the wages of the older age group (Lazear, 1990). Sim- ilarly, Casanova (2013) empirically demonstrates that there is no downward sloping wage–age prole for older people in the US. She nds that the wage of a typical fully employed male increases slightly after age 50. The downward-sloping age proles of wages and earnings often found in empirical research are due to an increased share of part-time employment in old age. Furthermore, Ilmakunnas and Maliranta (2005), using Finnish plant-level data, conclude that the wage–productivity gap increases with age, re- ecting strong seniority effects. In contrast, using a matched worker–rm panel data set of Dutch man- ufacturing rms, van Ours and Stoeldraijer (2011) nd little evidence of an age-related pay–productivity gap. Similarly, by comparing the age–wage and age– productivity proles, Cardoso et al. (2011) show that productivity increases up to age 50–54, while wages peak at lower ages—that is, the age of 40–44. They argue that wages rise in line with productivity gains at younger ages, while wage increases lag behind pro- ductivity gains at prime working ages. It follows that older workers are worth their pay. In contrast, Cataldi et al. (2011) show that, while older workers are indeed less productive than younger workers, the relative productivities across age groups do not show statisti- cally signicant differences between ICT and non-ICT rms, although they show that the upward age–wage prole seems to be somewhat steeper in ICT rms. Re- gardless of the ICT environment, however, the results 4 ECONOMIC AND BUSINESS REVIEW 2024;26:1–12 show that young workers are paid below, and older workers are paid above their marginal productivity. 2 Gender–productivity–wage nexus According to human capital models, women have traditionally participated less continuously in the labour market than men because of their family roles, resulting in their lower productivity and lower wages (Becker, 1985; Blau & Kahn, 2007; Mincer & Polachek, 1974). Moreover, women tend to choose less risky and consequently lower-paying occupations (Blau & Kahn, 2007) and also face income discrimination in the labour market (see, e.g., Castilla, 2008). However, the gender wage gap has narrowed over time (Blau & Kahn, 2007). According to an empirical study by Mandel and Semyonov (2014), this reduc- tion is particularly due to less frequent discrimination in the workplace. Moreover, centralised wage-setting institutions in Europe have worked to reduce the gender wage gap in industry. Considering the occu- pational gender segregation, this has also helped to reduce the gender wage gap (Kahn, 2014). Over time, men and women have also chosen more similar elds of study and occupations (England, 2010), which has further narrowed the gender pay gap. Using Finnish plant-level data, Ilmakunnas and Maliranta (2005) show that the share of female work- ers is negatively related to productivity, although this productivity difference is not fully reected in pay. However, this result may depend on the methodol- ogy used. Moreover, Hellerstein et al. (1999) conclude that the marginal product of women is lower than that of men, but that they are also paid signif- icantly less than men. In contrast to Ilmakunnas and Maliranta, they conclude that the wage gap be- tween men and women is much larger than the productivity gap. They also show that this conclu- sion holds most strongly for women who are not managers, who work in rms where many women are employed, and in larger rms. Zhang and Dong (2008), using Chinese rm-level data, also reach a similar conclusion. They nd that there is a signif- icant negative relationship between wages and the proportion of female employees, but also nd that the marginal productivity of female employees is signicantly lower than that of male employees. In ex- amining the gender wage–productivity gap between state-owned enterprises and private rms, they also nd that the wage gap in state-owned enterprises is smaller than the productivity gap, while the opposite is true for private rms. From this, they conclude that women in the state sector receive wage premiums, while women in the private sector experience wage discrimination. Using Canadian data, Dostie (2011) nds that pro- ductivity, measured by value added, is not signif- icantly different from wages on average. However, when looking at differences by age and gender, Dostie further nds that the productivity of women over the age of 54 continues to rise and exceed their wages, and this difference offsets the decline in productiv- ity for older men with university degrees. However, young men (under 35) earn less than their produc- tivity, while there is little difference between the productivity and wages of young women. In con- trast, using data from New Zealand, Sin et al. (2017) show that the relative wage–productivity gap be- tween genders increases with age and tenure, but only after the age of 40. Moreover, they nd that the productivity–wage gap is larger for highly skilled workers, when there is less product market competi- tion, and in more competitive markets. In this context, there is much academic literature analysing the gen- der productivity gap. Much of the research suggests that women’s underrepresentation in science actually results from the existing gender productivity gap (for a meta-analytic review, see Astegiano et al., 2019). However, Garnero et al. (2014), using Belgian data from 1999–2006, show that gender and age differences tend to be detrimental to rm productivity in general, but that gender diversity is benecial in technology- and knowledge-intensive rms because it can fos- ter complementarities and a more enjoyable working environment and hence increases productivity. The opposite has been found for more traditional sec- tors, in which gender diversity negatively impacts productivity. 3 Methodological framework The main objective of this paper was to investi- gate the gender- and age-related wage–productivity gap in intangible and non-intangible work occupa- tions. Specically, there were ve research objectives, namely (i) to examine the gender wage gap control- ling for age and type of occupation, (ii) to examine the age-related wage gap controlling for gender and type of occupation, (iii) to examine the gender pro- ductivity gap and (iv) to examine the age-related productivity gap controlling for either age or gen- der and type of occupation, and (v) to examine the wage–productivity gap separately by gender and age controlling for type of occupation. The methodological framework relied on Dostie (2011) and rst dened 12 different groups of workers based on their age (under 30, 30–49, and 50C), gender (men and women), and the type of work performed (non-intangible and intangible work occupations). Intangible work occupations include organizational ECONOMIC AND BUSINESS REVIEW 2024;26:1–12 5 Table 1. Individuals (occupations) belonging to different types of intangible capital work occupations. Organizational capital work R&D capital work ICT capital work Business services and administration managers; Production managers in agriculture, forestry and sheries; Manufacturing, mining, construction, and distribution managers; Professional services managers; Finance professionals; Administration professionals. Physical and earth science professionals; Mathematicians, actuaries and statisticians; Life science professionals; Engineering professionals (excluding electrotechnology); Electrotechnology engineers; Architects, planners, surveyors and designers; Medical doctors; Nursing and midwifery professionals; Other health professionals; Physical and engineering science technicians; Life science technicians and related associate professionals; Medical and pharmaceutical technicians; Research and development managers. Information and communications technology professionals; Information and communications technicians; Information and communications technology service managers. Source: Authors’ own work based on Piekkola, Bloch, Rybalka, and Redek (2021). capital work, R&D capital work, and ICT capital work as dened by the Globalinto methodology (for more details, see, e.g., Piekkola, Bloch, Rybalka, & Redek, 2021). All other employees were assumed to perform non-intangible capital work. Table 1 provides indi- viduals who t into one of these three categories of intangible capital work occupations. Second, we calculated the mean wage (i.e., gross wages excluding employers’ social contributions) for each of these groups. Employees who did not work full time were weighted differently—in pro- portion to their hours worked per week. We then compared the means with respect to 1) gender and 2) age, controlling for different types of work in each case. To analyse the differences, we calculated the absolute and relative differences in mean wages between the different groups of individuals. The rel- ative gender wage gap was calculated as the ratio between the average wage of men and women, sep- arately for different age groups and types of work (Equation 1). The relative gender wage gap was calculated as: Gender wage gapD ¯ w m ¯ w f ¯ w m 100 (1) where ¯ w m is the mean wage for men, and ¯ w f is the mean wage for women. Gender wage gaps were therefore reported as percentages and dened as the absolute difference between the mean wages of men and women, relative to the mean wage of men. Further, the relative age gap was calculated as the ratio between the average wage of the group of older workers (i.e., either 30–49 or 50C) and younger workers (i.e., either workers under 30 or 30–49) (Equa- tion 2). Such a comparison between age groups (older vs. younger workers) sheds light on how wages change over the life cycle of individuals. Finally, we compared the mean differences using the t-test for independent samples. The relative age wage gap was calculated as: Age wage gapD ¯ w older ¯ w younger ¯ w older 100 (2) where ¯ w older is the mean wage of the group of older employees (i.e., either 30–49 or 50C), and ¯ w younger is the mean wage of the group of younger workers (i.e., either workers under 30 or 30–49). This means that the age wage gap was calculated once comparing the rst (under 30 years old) and the second (30–49 years old) age group and then the second (30–49 years old) and the third (50C years old). In this paper, productivity per employee is mea- sured using real total value added of a rm. While the register includes data on wages at the individ- ual level, (real) value added (i.e., productivity) is recorded only at the rm level. To distribute the total value added of the rm among its employees, differ- ent methodologies can be used. For example, Dostie (2011) assumes that groups of workers (dened by, for example, age and gender) are perfectly substitutable with the same marginal product that includes the ra- tio of the number of employees of a specic group over the total number of employees and inserts this in a production function. Following this methodology, it is assumed that workers have the same marginal product across rms (Hellerstein et al., 1999). How- ever, as Skirbekk (2004), for example, shows, the productivity of workers of different ages varies. To consider that and to relax the assumption that groups of workers are perfectly substitutable, we aimed to decompose the value added within a rm to different groups of workers. In order to do so, we followed the methodological framework used in the estima- tion of the National Transfer Accounts (Population Division, 2013), where some variables (such as private consumption), reported at the household level, are distributed among household members using a re- gression method without a constant term. Therefore, in this paper, the value added of a rm is distributed among groups of workers in this way. Specically, we 6 ECONOMIC AND BUSINESS REVIEW 2024;26:1–12 regressed each rm’s total value added on the share of workers belonging to a particular group, again con- trolling for differences in workers’ working hours. The beta coefcients (reported in Appendix, Table A1) then served as weights for the distribution of the rm’s total value added across its employee groups (i.e., 12 different types of employees) and, nally, em- ployee value added. The weights were calculated as: weight i D b i x i P 12 i b i x i (3) where i is each of the groups, and i2 (1; 12);b i is the estimated regression coefcient of each group, and x i is the number of employees in each group. Finally, value added for each employee was calculated as: V A i D TV A real weight i (4) where V A is the value added per employee in each group, and TV A real is the real total value added. As with wages, we compared the average value added of the groups using absolute and relative dif- ferences. However, due to the equal distribution of value added among the rm’s employees belonging to the same group, the productivity differences were not tested with a t-test. Finally, we calculated the gender wage– productivity gap by dividing the relative gender wage gap and the relative gender productivity gap. In this way, we captured how much of the wage gap resulted from productivity differences rather than from other factors, such as labour market discrimination. Similarly, we calculated the age-related wage–productivity gap, but using the age-related differences in wages and productivity. In this paper, we use linked employer–employee data (LEED) from 2017, the latest available data point in the analysis period, provided by the Statistical Ofce of the Republic of Slovenia (SORS). In order to analyse gender- and age-related wage and pro- ductivity differentials, we restricted our analysis to privately owned rms for which all required data were available. Ultimately, we obtained a sample size of 516,068 employees, of which about 36% were men aged 30–49, followed by women in the same age group (about 22%). Men aged 50 or older made up 17% of the total sample, and women in the same age group made up 10%. The youngest age group made up the smallest percentage of the total sample. Men under the age of 30 made up 11% of the total sample, while women in the same age group made up 5% of the total sample. The majority of individu- als belonged to the non-intangible work occupations (about 84%), while 16% of the total sample belonged to the intangible work occupations (12% men and 4% women) (see Table 2 for details). Table 2. Number of observations by type of workers. Type of work Men Women Row total Under 30 years old Non-intangible 48,476 23,541 72,017 Intangible 8683 2998 11,681 30–49 years old Non-intangible 149,224 98,637 247,861 Intangible 35,700 14,287 49,987 50C years old Non-intangible 69,347 43,836 113,183 Intangible 15,885 5454 21,339 Total 327,315 188,753 516,068 Source: SORS (2022), own calculations. 4 Gender and age-related wage gap Table 3 shows the results of absolute and relative gender wage differentials by age and type of work. In general, individuals who performed non-intangible work earned less than individuals who performed intangible work, regardless of age. Table 3 shows mainly signicant wage differences in favour of men, ranging from 8% to 32%. The only exception is women aged 50 or more in the intangible work group, where men and women received roughly the same wage on average. Gender gaps were larger for workers in the non-intangible work group, with the largest gender gaps for those aged 50 or older. On the other hand, gender gaps were smaller in intangible occupations and decreased steadily with age. Table 4 shows the results of absolute and relative age wage differences by gender and type of work. On average, older workers received higher wages, regardless of gender and type of work. The largest relative differences across age groups were between workers under the age of 30 and workers between the ages of 30 and 49. These age differences were par- ticularly pronounced in intangible work occupations, Table 3. Absolute and relative gender wage gaps, by age and type of work, Slovenia, 2017. Mean Mean wage— wage— Absolute Relative men women difference difference Type of work (w m ) (w w ) (w m w w ) (w m =w w ) Under 30 years old Non-intangible 13,236 10,867 2,369 1.22 Intangible 18,280 16,655 1,625 1.10 30–49 years old Non-intangible 17,686 14,873 2,813 1.19 Intangible 28,785 26,740 2,045 1.08 50C years old Non-intangible 20,174 15,328 4,846 1.32 Intangible 31,755 32,224 469 0.99 Notes: Wages in euros per year. is signicant at 10%, is signicant at 5%, and is signicant at 1%. Source: SORS (2022), own calculations. ECONOMIC AND BUSINESS REVIEW 2024;26:1–12 7 Table 4. Absolute and relative age wage gaps, by gender and type of work, Slovenia, 2017. Mean wage—younger Mean wage—older Absolute difference Relative difference Type of work group (w y ) group (w o ) (w o w y ) (w o =w y ) Men, under 30 vs. 30-49 years old Non-intangible 13,236 17,686 4,450 1.34 Intangible 18,280 28,785 10,505 1.57 Men 30–49 vs. 50C years old Non-intangible 17,686 20,174 2,488 1.14 Intangible 28,785 31,755 2,970 1.10 Women, under 30 vs. 30–49 years old Non-intangible 10,867 14,873 4,006 1.37 Intangible 16,655 26,740 10,085 1.61 Women 30–49 vs. 50C years old Non-intangible 14,873 15,328 455 1.03 Intangible 26,740 32,224 5,484 1.21 Notes: Wages in euros per year. is signicant at 10%, is signicant at 5%, and is signicant at 1%. Source: SORS (2022), own calculations. where women aged 30 to 49 received wages that were on average 61% higher than for those under 30. The situation is similar for men, but with somewhat smaller differences between age groups. The wage differences between workers aged 30 to 49 and those aged 50 and over, while still mainly statistically sig- nicant, were smaller regardless of gender and work type—accounting for up to 21% in favour of those aged 50 and over. The age difference was smallest for older women in the non-intangible work type. 5 Gender and age-related productivity gap Our results show that, in general, workers who belonged to intangible work occupations were, on average, more productive than their counterparts who belonged to non-intangible work occupations, regardless of gender and age. The only exception is women aged 50C, for whom productivity was about the same regardless of work type (see Table 5). Our results also show that men were generally less pro- ductive than women, with a 10–23% difference in favour of women. The only exception is for workers under 30 in non-intangible occupations, where men’s productivity was 12% higher than women’s. The gen- der productivity gap was smaller in intangible work occupations but tended to increase with age. This could be a consequence of positive selection bias, i.e., that only those with higher wages remained in the labour market. Table 6 shows the absolute and relative age– productivity gaps by gender and type of work. Our results show that older workers were always less pro- ductive than their younger counterparts, regardless of the age groups, gender, and type of work. The only exception is women aged 30–49 who belonged to non-intangible work occupations, whose produc- tivity was 18% higher than that of their peers under 30. Although older workers (aged 30–49 or 50C) were generally less productive than their younger counterparts (aged under 30 or 30–49), productivity differences narrowed with age. This is true regardless of gender or work type. Table 6 also shows that age- related productivity differences were always smaller for non-intangible work occupations—when compar- ing productivity differences between 30–49- and 50C- year-olds, the difference was only 3% for men and 4% for women, always in favour of 30–49-year-olds. On Table 5. Absolute and relative gender productivity gaps, by age and type of work, Slovenia, 2017. Mean productivity—men Mean productivity—women Absolute difference Relative difference Type of work (P m ) (P w ) (P m P w ) (P m =P w ) Under 30 years old Non-intangible 38,504 34,498 4,006 1.12 Intangible 68,217 75,628 7,411 0.90 30–49 years old Non-intangible 31,465 40,694 9,229 0.77 Intangible 43,692 49,465 5,773 0.88 50C years old Non-intangible 30,677 39,250 8,573 0.78 Intangible 31,288 38,812 7,524 0.81 Note: Productivity in euros per year. Source: SORS (2022), own calculations. 8 ECONOMIC AND BUSINESS REVIEW 2024;26:1–12 Table 6. Absolute and relative age productivity gaps, by gender and type of work, Slovenia, 2017. Mean productivity— Mean productivity— Absolute difference Relative difference Type of work younger group (P y ) older group (P o ) (P o P y ) (P o =P y ) Men, under 30 vs. 30–49 years old Non-intangible 38,504 31,465 7,039 0.82 Intangible 68,217 43,692 24,525 0.64 Men 30–49 vs. 50C years old Non-intangible 31,465 30,677 788 0.97 Intangible 43,692 31,288 12,404 0.72 Women, under 30 vs. 30–49 years old Non-intangible 34,498 40,694 6,196 1.18 Intangible 75,628 49,465 26,163 0.65 Women 30–49 vs. 50C years old Non-intangible 40,694 39,250 1,444 0.96 Intangible 49,465 38,812 10,653 0.78 Note: Productivity in euros per year. Source: SORS (2022), own calculations. the other hand, age-related productivity differences were higher for intangible work occupations, espe- cially when comparing younger age groups (30–49 vs. under 30), where the productivity of older workers was 36% and 35% lower than that of younger workers (men and women, respectively). 6 Gender- and age-related wage–productivity gap Table 7 shows the gender wage–productivity gap, calculated by dividing the relative gender wage gap and the relative gender productivity gap, separately by age and type of work. Such an indicator was in- tended to show how much of the gender wage gap can be explained by the gender productivity gap. Our results show that while women were generally paid less than men, they were generally more productive than men, leading to the gender wage–productivity gap, which was always in favour of men. This is true even for non-intangible workers under the age of 30, where men were actually more productive Table 7. Gender wage–productivity gaps, by age and type of work, Slovenia, 2017. Gender Gender Gender wage— wage productivity productivity gap gap gap (wage gap/ Type of work (w m =w w ) (P m =P w ) productivity gap) Under 30 years old Non-intangible 1.22 1.12 1.09 Intangible 1.10 0.90 1.22 30–49 years old Non-intangible 1.19 0.77 1.54 Intangible 1.08 0.88 1.22 50C years old Non-intangible 1.32 0.78 1.68 Intangible 0.99 0.81 1.22 Source: SORS (2022), own calculations. than women, but the wage gap was still larger than the productivity gap, resulting in a positive gender wage–productivity gap in favour of men. Table 7 also shows that the gender wage–productivity gap for non-intangible occupations increased with age, from 1.09 for those under the age of 30 to 1.68 for those aged 50C. Moreover, the gender wage–productivity gap was higher for non-intangible occupations in the 30–49 and 50C age groups than for intangible occupa- tions, where the gender wage–productivity gap was a constant 1.22 regardless of age. Table 8 shows the wage–productivity gap by age. Regardless of age, gender, and type of work, the higher wages received by the older group were never offset by their higher relative productivity. In gen- eral, the age-related wage–productivity gap even exceeded the age-related wage gap, meaning that older workers earned more than their younger coun- terparts despite being less productive. However, the Table 8. Age wage–productivity gaps, by gender and type of work, Slove- nia, 2017. Age Age Age wage– wage productivity productivity gap gap gap (wage gap/ Type of work (w o =w y ) (P o =P y ) productivity gap) Men, under 30 vs. 30–49 years old Non-intangible 1.34 0.82 1.64 Intangible 1.57 0.64 2.46 Men 30–49 vs. 50C years old Non-intangible 1.14 0.97 1.17 Intangible 1.10 0.72 1.54 Women, under 30 vs. 30–49 years old Non-intangible 1.37 1.18 1.16 Intangible 1.61 0.65 2.45 Women 30–49 vs. 50C years old Non-intangible 1.03 0.96 1.07 Intangible 1.21 0.78 1.54 Source: SORS (2022), own calculations. ECONOMIC AND BUSINESS REVIEW 2024;26:1–12 9 age-related wage–productivity gap decreased with age, being less pronounced when comparing the age groups 50C and 30–49 than when comparing age groups 30–49 and under 30. Age-related wage– productivity gaps also tended to be smaller for women then for men. In the case of women, when comparing individuals aged 30–49 and 50C belong- ing to non-intangible work occupations, the gap equalled 1.07 only, resulting from relatively low wage and productivity gaps for this group of workers. In contrast, age-related wage–productivity gaps were al- ways relatively high in the case of intangible work occupations (as compared to non-intangible ones); for example, the gap was approximately 2.5 for both gen- ders belonging to the intangible capital group when comparing workers aged below 30 and 30–49. This means that workers aged 30–49 performing intangible work occupations did not only earn much more than their counterparts aged below 30, but they tended to also be much less productive than their younger counterparts. 7 Discussion and conclusion We found signicant wage differentials in favour of men, ranging from 8% to 32%, with the exception of women of 50 or older belonging to the intangible capital group, where men’s and women’s wages are roughly equal. This result can be explained by a pos- itive selection bias revealed in Mahlberg et al. (2006). Gender differences are larger for non-intangible work occupations than for intangible occupations, and for the latter the differences actually decrease with age. We also found that older workers, regardless of the type of capital and gender, receive higher wages. Sim- ilar ndings have been made by other authors, such as Skirbekk (2004) and Casanova (2013). The largest relative differences between age groups are between workers up to the age of 30 and workers between the ages of 30 and 49. The age-related wage gap is generally smaller for non-intangible occupations. Our results also show that women are more pro- ductive than men, with the exception of young men (under the age of 30) who are in the non-intangible type of work occupations. The gender productivity gap generally increases with age, leading to a 19% (for intangible occupations) or 22% (for non-intangible occupations) lower productivity for men aged 50C than for women. Since women in Slovenia tend to retire earlier than men, this may also be the result of a positive selection bias. In a similar vein to Hu (2016), Gabriele et al. (2018), and Lee et al. (2018), our results also show that older workers (aged 30–49 or 50C) tend to be less productive than their younger counterparts (aged under 30 or 30–49), with the sole exception of younger women in the non-intangible group. However, age-related productivity differences narrow with increasing age. In summary, the gender wage–productivity gaps show that women are paid less than men in most groups, but that they are generally more productive than men, which implies that the gender wage– productivity gaps are actually larger than the gender wage gaps in most cases. The only exception is men under 30 in non-intangible occupations, where the gender wage–productivity gap is still in favour of men, but they tend to be more productive than women. Although these results are subject to the productivity measure used in this study, they are important for policymakers seeking to reduce the gender gap in the labour force. Regardless of age, gender, and type of work, the higher wages received by older groups are never offset by their higher rel- ative productivity. The result is that older workers, although they earn more than their younger counter- parts, are generally less productive. This leads to an age-related wage–productivity gap in favour of older workers, regardless of gender or work type. Although the age-related wage–productivity gap declines with age, it is particularly problematic in the current era of population ageing, when rms employ a growing share of older workers. It is important to keep in mind, however, that our work provides a single-year analysis, and therefore the generalizability of the re- sults may be limited. Future research could therefore rely on additional analytical methods based on longer time series. Acknowledgements The work was co-funded by H2020 GLOBAL- INTO project, which was supported by the European Union’s Horizon 2020 The mechanisms to promote smart, sustainable and inclusive growth under grant agreement No. 822259. This work was also co-funded by the following Slovenian Research Agency grants: P5-0128, J5-4575, V5-2264, V5-2267. The paper was prepared using linked employer-employee datasets, provided by the Statistical Ofce of the Republic of Slovenia under the contract No. 960/150/2019. This analysis would not be possible without their expert support, in particular the User relations section of the Data publication and communication division. References Astegiano, J., Sebastián-González, E., & Castanho, C. de T. (2019). Unravelling the gender productivity gap in science: A meta- analytical review. Royal Society Open Science, 6(6), Article 181566. https://doi.org/10.1098/rsos.181566 10 ECONOMIC AND BUSINESS REVIEW 2024;26:1–12 Bavdaž, M., Piekkola, H., Roth, F., Caloghirou, Y., & Redek, T. (2021, June 9). Investicije v znanje, neotipljivi kapital in izzivi rasti prihodnosti: kako se primerjate z najboljšimi in kaj nas uˇ ci pandemija? [Investments in knowledge, intangible capital, and future growth challenges: How do you compare to the best, and what can we learn from the pandemic?] [Web event]. Alumni EF. Becker, G. S. (1985). Human capital, effort, and the sexual division of labor. Journal of Labor Economics, 3(1), S33–S58. Ben-Porath, Y. (1967). The production of human capital and the life cycle of earnings. Journal of Political Economy, 75(4), 352–365. https://doi.org/10.1086/259291 Blau, F. D., & Kahn, L. M. (2007). The gender pay gap: Have women gone as far as they can? Academy of Management Perspectives, 21(1), 7–23. https://doi.org/10.5465/AMP.2007.24286161 Cardoso, A. R., Guimarães, P ., & Varejão, J. (2011). Are older workers worthy of their pay? An empirical investigation of age–productivity and age–wage nexuses. De Economist, 159(2), 95–111. https://doi.org/10.1007/s10645-011-9163-8 Casanova, M. (2013). Revisiting the hump-shaped wage prole: Im- plications for structural labor supply estimation. University of California, Los Angeles. Castilla, E. J. (2008). Gender, race, and meritocracy in organizational careers. American Journal of Sociology, 113(6), 1479–1526. https:// doi.org/10.1086/588738 Cataldi, A., Kampelmann, S., & Rycx, F. (2011). Productivity–wage gaps among age groups: Does the ICT environment matter? De Economist, 159(2), 193–221. https://doi.org/10.1007/s10645-011 -9162-9 Directorate-General for Economic and Financial Affairs (2021). The 2021 Ageing Report: Economic and budgetary projections for the 28 EU member states (2019–2070). European Commission, Publica- tions Ofce of the European Union. https://doi.org/10.2765/ 84455 Dostie, B. (2011). Wages, productivity and aging. De Economist, 159(2), 139–158. https://doi.org/10.1007/s10645-011-9166-5 England, P . (2010). The gender revolution: Uneven and stalled. Gender and Society, 24(2), 149–166. https://doi.org/10.1177/ 0891243210361475 European Institute for Gender Equality. (2020). Gender Equality In- dex 2020: Digitalisation and the future of work. https://eige.europa .eu/publications/gender-equality-index-2020-report Gabriele, R., Tundis, E., & Zaninotto, E. (2018). Ageing workforce and productivity: The unintended effects of retirement regula- tion in Italy. Economia Politica, 35(1), 163–182. https://doi.org/ 10.1007/s40888-017-0079-x Garnero, A., Kampelmann, S., & Rycx, F. (2014). The heterogeneous effects of workforce diversity on productivity, wages, and prof- its. Industrial Relations: A Journal of Economy and Society, 53(3), 430–477. https://doi.org/10.1111/irel.12064 Gordo, L. R., & Skirbekk, V . (2013). Skill demand and the compar- ative advantage of age: Jobs tasks and earnings from the 1980s to the 2000s in Germany. Labour Economics, 22, 61–69. https:// doi.org/10.1016/j.labeco.2012.09.003 Hellerstein, J. K., Neumark, D., & Troske, K. R. (1999). Wages, pro- ductivity, and worker characteristics: Evidence from plant-level production functions and wage equations. Journal of Labor Eco- nomics, 17(3), 409–446. https://doi.org/10.1086/209926 Hu, Y. (2016). Implications of population ageing for the Chi- nese productivity. Economic and Political Studies, 4(4), 454–467. https://doi.org/10.1080/20954816.2016.1251728 Ilmakunnas, P ., & Maliranta, M. (2005). Technology, labour char- acteristics and wage–productivity gaps. Oxford Bulletin of Eco- nomics and Statistics, 67(5), 623–645. https://doi.org/10.1111/ j.1468-0084.2005.00134.x Kahn, L. M. (2014). New evidence on gender and the labour market: A symposium. Industrial & Labor Relations Review, 67(2), 283– 286. https://doi.org/10.1177/001979391406700201 Lazear, E. P . (1979). Why is there mandatory retirement? Journal of Political Economy, 87(6), 1261–1284. https://doi.org/10.1086/ 260835 Lazear, E. P . (1990). Job security provisions and employment. The Quarterly Journal of Economics, 105(3), 699–726. https://doi.org/ 10.2307/2937895 Lee, B., Park, J., & Yang, J.-S. (2018). Do older workers really reduce rm productivity? The Economic and Labour Relations Review, 29(4), 521–542. https://doi.org/10.1177/1035304618811008 Lee, J.-W., Kwak, D. W., & Song, E. (2022). Can older workers stay productive? The role of ICT skills and training. Journal of Asian Economics, 79, Article 101438. https://doi.org/10.1016/j.asieco .2021.101438 Lee, J.-W., Song, E., & Kwak, D. W. (2020). Aging labor, ICT capital, and productivity in Japan and Korea. Journal of the Japanese and International Economies, 58, Article 101095. https://doi.org/10 .1016/j.jjie.2020.101095 Mahlberg, B., Freund, I., Crespo Cuaresma, J., & Prskawetz, A. (2013a). Ageing, productivity and wages in Austria. Labour Economics, 22, 5–15. https://doi.org/10.1016/j.labeco.2012.09 .005 Mahlberg, B., Freund, I., Crespo Cuaresma, J., & Prskawetz, A. (2013b). The age–productivity pattern: Do location and sector afliation matter? The Journal of the Economics of Ageing, 1–2, 72–82. https://doi.org/10.1016/j.jeoa.2013.08.002 Mahlberg, B., Skirbekk, V ., Freund, I., Winkler-Dworak, M., Lindh, T., Malmberg, B., Jans, A.-C., Skans, O. N., & Andersson, F. (2006). The impact of population ageing on innovation and produc- tivity growth in Europe (Research Report 28). Vienna Institute of Demography. Mandel, H., & Semyonov, M. (2014). Gender pay gap and employ- ment sector: Sources of earnings disparities in the United States, 1970–2010. Demography, 51(5), 1597–1618. https://doi.org/10 .1007/s13524-014-0320-y Mincer, J. (1958). Investment in human capital and personal income distribution. Journal of Political Economy, 66(4), 281–302. Mincer, J., & Polachek, S. (1974). Family investments in human capital: Earnings of women. Journal of Political Economy, 82(2), S76–S108. https://doi.org/10.1086/260293 Ng, T. W. H., & Feldman, D. C. (2012). Evaluating six common stereotypes about older workers with meta-analytical data. Per- sonnel Psychology, 65(4), 821–858. https://doi.org/10.1111/peps .12003 Park, C.-Y., Shin, K., & Kikkawa, A. (2021). Aging, automation, and productivity in Korea. Journal of the Japanese and Interna- tional Economies, 59, Article 101109. https://doi.org/10.1016/ j.jjie.2020.101109 Piekkola, H., Bloch, C., Redek, T., & Rybalka, M. (2021). Pro- ductivity growth and the intangible divide. In P . Domadenik Muren, M. Koman, & T. Redek (Eds.), Achieving growth that matters: A new economic paradigm (pp. 279–313). ˇ Casnik Finance, 2021. Piekkola, H., Bloch, C., Rybalka, M., & Redek, T. (2021). Intangibles from innovative work—Their valuation and technological change (H2020 Globalinto Project Report, Deliverable 5.3). University of Vaasa. https://globalinto.eu/wp-content/uploads/2023/ 02/D5.3-IntangiblesValuationTechnicalChange_2022_10_03 .pdf Population Division. (2013). National Transfer Accounts manual: Measuring and analysing the generational economy. United Nations, Department of Economic and Social Affairs. https://www.un.org/en/development/desa/publications/ measuring-and-analysing-the-generational-economy.html Prašnikar, J. (Ed.). (2010). The role of intangible assets in exiting the crisis. ˇ Casnik Finance. Sambt, J., Hammer, B., & Isteniˇ c, T. (2021). The European Na- tional Transfer Accounts: Data and applications. Economic and Business Review, 23(3), 184–193. https://doi.org/10.15458/2335 -4216.1287 Sin, I., Stillman, S., & Fabling, R. (2017). What drives the gender wage gap? Examining the roles of sorting, productivity differences, and discrimination (IZA Discussion Paper No. 10975). IZA Institute of Labor Economics. https://docs.iza.org/dp10975.pdf Skirbekk, V . (2004). Age and individual productivity: A litera- ture survey. Vienna Yearbook of Population Research, 2, 133–153. https://doi.org/10.1553/populationyearbook2004s133 Skirbekk, V . (2008). Age and productivity capacity: Descriptions, causes and policy options. Ageing Horizons, 8, 4–12. https:// www.ageing.ox.ac.uk/download/47 ECONOMIC AND BUSINESS REVIEW 2024;26:1–12 11 SORS - Statistical Ofce of the Republic of Slovenia. (2022). Linked employer-employee datasets provided upon request under the contract No. 960/150/2019. van Ours, J. C., & Stoeldraijer, L. (2011). Age, wage and productivity in Dutch manufacturing. De Economist, 159(2), 113–137. https:// doi.org/10.1007/s10645-011-9159-4 Verbiˇ c, M., & Polanec, S. (2014). Innovativeness and intangibles in transition: The case of Slovenia. Economic Research/Ekonomska Istraživanja, 27(1), 67–85. https://doi.org/10.1080/1331677X .2014.947109 Vodopivec, M. (2014). Equal pay for equal work? Eastern European Economics, 52(5), 87–110. https://doi.org/10.1080/001287755 .2014.1004266 Zhang, L., & Dong, X.-Y. (2008). Male–female wage discrimina- tion in Chinese industry: Investigation using rm-level data. Economics of Transition, 16(1), 85–112. https://doi.org/10.1111/ j.1468-0351.2007.00317.x 12 ECONOMIC AND BUSINESS REVIEW 2024;26:1–12 Appendix Table A1. Regression results of value-added decomposition. Variable (in share) Coefcient Men, under 30, non-intangible 524.558*** (150.854) Men, 30–49, non-intangible 401.866*** (62.327) Men, 50C, non-intangible 381.095*** (97.689) Women, under 30, non-intangible 489.376** (229.692) Women, 30–49, non-intangible 561.655*** (90.783) Women, 50C, non-intangible 550.322*** (144.476) Men, under 30, intangible 824.532** (325.191) Men, 30–49, intangible 500.591*** (105.764) Men, 50C, intangible 323.455** (142.814) Women, under 30, intangible 959.749 (600.242) Women, 30–49, intangible 550.145*** (181.828) Women, 50C, intangible 399.311 (260.051) Number of observations 44,923 Standard errors in parentheses. *** p < :01, ** p < :05, * p < :1.