Spillovers_in_Ownership_Interlocking_Networks.pdf

1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Gender diversity spillovers through networks Ownership and Interlocking Hubert Marek Drazkowski Gabriela Contreras Joanna Tyrowicz Szczecin, April 2025 Gender diversity spillovers through networks 1 / 26

2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Questions and suggestions and grilling are welcomed during the presentation (apparently we have 40 minutes) Gender diversity spillovers through networks 2 / 26

3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Society’s dilemma ... economic dillema • 68% of the firms do not have any women on boards (executive or supervisory in corporate Europe) (4% do not have any men) (Drazkowski et al. 2024) • Reasons: childcare, ∆ psychology ... historical biases, cognitive biases • At least, slightly more women bolster the firm performance (10 p.p. ∆ GBD ⇒ 1% ∆ OPRE) (Tyrowicz et al. 2024) ⇒ Decision makers want to promote diversity (e.g., EU 2022-2026 Directive) • Policy: “Let’s impose quotas on boards in listed firms” • (1) Will it ripple? (2) Is there a mechanism? (3) What is the effect? (4) What is a better policy? New measurement medium: firms’ ownership networks Gender diversity spillovers through networks 3 / 26

11. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Our contribution Past literature studied spillovers • Mostly within firms and much less in interlocked boards • Almost all studies focused on public firms • Empirical evidence is mixed • Theories are not integrated • No tests between theories Novelty 1 Unseen data: 2.5 million of private firms with managers in networks from 29 European countries (ownership and interlocking) 2 New medium: Ownership networks 3 New opportunities: New test space between theories: hierarchy (ownership) vs imitation (interlocks) Research questions 1 Are there gender diversity spillovers through ownership networks? 2 Are the spillovers driven by power through hierarchy or by imitation of peers? ... Gender diversity spillovers through networks 4 / 26

14. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References 1 Literature 2 Data (GBDD and ownership) 3 Methods (panel linear model) 4 Results Gender diversity spillovers through networks 5 / 26

16. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Gender diversity spillovers through networks 7 / 26

17. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References In firm spillovers (correlations) In firm - horizontal (supervisory → executive) • + (mostly public) (Matsa and Miller 2011, Cook and Glass 2015, Gould et al. 2018, Guldiken et al. 2019, Kirsch and Wrohlich 2020) • 0/− no effect, replacement, negative (Farrell and Hersch 2005, Smith and Parrotta 2018, Garcia-Blandon et al. 2023, Maida and Weber 2022, Schoonjans et al. 2024) • + in visible vs 0 in not visible (Drazkowski et al. 2023) • + when power vs 0 when no power (Bozhinov et al. 2021) In firm - vertical (higher ↓ lower ranks) • + positive (Bossler et al. 2020, Kunze and Miller 2017) Gender diversity spillovers through networks 8 / 26

23. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Networks - interlocking members and organizational hierarchy • Personal networks matter a lot in job search (Granovetter 1973, 1995, Holzer 1988, Montgomery 1991, Mortensen and Vishwanath 1994, Hensvik and Skans 2016, Kramarz and Skans 2014, Dustmann et al. 2016, Brown et al. 2016, von Essen and Smith 2023) • Social network theory (imitation under risk, peers contrast, supply network) (Gimeno et al. 2022, Mateos de Cabo et al. 2024) • Strategy and corporate culture spill in the ownership network e.g., (Ghoshal and Bartlett 1990, Yoshikawa et al. 2019) • Resource dependency theory Mizruchi (1996) What else? Gender diversity spillovers through networks 9 / 26

29. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Ownership networks - main selling point • There is power and directionality • Encapsulates in-ﬁrm mechanisms and personal networks • Across diﬀerent environments ⇒ Perfect ground for testing theories Gender diversity spillovers through networks 10 / 26

30. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Ownership networks - main selling point • There is power and directionality • Encapsulates in-ﬁrm mechanisms and personal networks • Across diﬀerent environments ⇒ Perfect ground for testing theories Gender diversity spillovers through networks 10 / 26

32. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Managers - sample • GBDD (Drazkowski et al. 2024) • Gender attributed • Corporate bodies: managerial (executive) ∪ supervisory (non-executive) positions ⊆ boardroom (boards) • Time given • Ought to have boards, at least 2 people, at least 2 years, joins with networks sample Table 1: Sample descriptives • # of unique obs. # of ﬁrm/person-years. Firms 2,421,247 8,920,469 Listed 9,145 47,664 People 8,739,302 27,610,247 Men 6,667,682 21,340,294 Women 1,988,946 5,987,913 Total 8,656,628 27,328,207 Women % in total attributed 22.98 21.91 Gender diversity spillovers through networks 12 / 26

35. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Ownership - sample • Owners: 1,593,648 • Subsidiaries: 1,899,92 Table 2: Quantiles for Degree, In-Degree, and Out-Degree for direct connections Quantile Peers Parents Subsidiaries 25% 2 1 1 50% 2 1 1 75% 3 1 2 90% 6 2 4 95% 11 3 8 Gender diversity spillovers through networks 13 / 26

36. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Ownership - sample • Owners: 1,593,648 • Subsidiaries: 1,899,92 Table 2: Quantiles for Degree, In-Degree, and Out-Degree for direct connections Quantile Peers Parents Subsidiaries 25% 2 1 1 50% 2 1 1 75% 3 1 2 90% 6 2 4 95% 11 3 8 Gender diversity spillovers through networks 13 / 26

37. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Ownership - construction • Subsidiaries = aﬃliated, but ̸= branches • Direct and indirect % of ownership (use direct) • Cannot assert all the links (sampling bias?) • Asynchronous information on ownership (interpolation) • Qualitative links (use weights and indicators) • Stocklisted daughters excluded Gender diversity spillovers through networks 14 / 26

46. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Interlocks sample • Three consecutive years, at least two board members, exclude listed ﬁrms as receivers • Number of ﬁrms in interlocking boards network 888,213 Gender diversity spillovers through networks 15 / 26

47. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Interlocks sample • Three consecutive years, at least two board members, exclude listed ﬁrms as receivers • Number of ﬁrms in interlocking boards network 888,213 Gender diversity spillovers through networks 15 / 26

48. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Interlocks - construction • Unique identiﬁers for persons based on names and surnames, personal information, company information • Incomplete links Table 3: Comparison of edges in two networks by year raw data year Ownership edges Intersected edges Interlocking edges 2010 261,081 68,809 10,025,763 2011 245,790 67,558 11,825,801 2012 458,686 72,040 15,020,010 2013 609,539 97,084 17,894,085 2014 778,999 254,359 32,104,839 2015 644,137 212,210 24,510,663 2016 272,625 169,353 29,436,047 2017 316,184 195,129 25,873,370 2018 471,425 260,763 28,154,505 2019 446,580 265,992 25,787,610 2020 146,592 83,801 5,979,976 • After restriction of at least 3 consecutive years 2,749,049 nodes left Gender diversity spillovers through networks 16 / 26

49. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Interlocks - construction • Unique identiﬁers for persons based on names and surnames, personal information, company information • Incomplete links Table 3: Comparison of edges in two networks by year raw data year Ownership edges Intersected edges Interlocking edges 2010 261,081 68,809 10,025,763 2011 245,790 67,558 11,825,801 2012 458,686 72,040 15,020,010 2013 609,539 97,084 17,894,085 2014 778,999 254,359 32,104,839 2015 644,137 212,210 24,510,663 2016 272,625 169,353 29,436,047 2017 316,184 195,129 25,873,370 2018 471,425 260,763 28,154,505 2019 446,580 265,992 25,787,610 2020 146,592 83,801 5,979,976 • After restriction of at least 3 consecutive years 2,749,049 nodes left Gender diversity spillovers through networks 16 / 26

51. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Panel data features 1 Firm fixed effects: • subsidiary ∼ in subsidiary variation, differences in parents across • parent ∼ in parent variation, differences between subsidiaries across • parent-subsidiary ∼ in pair variability 2 Controls (for parents and subsidiaries) • Time fixed effects ∼ global shocks • Industry fixed effects ∼ environment relative diversity • HHI ∼ competetiveness dimension • Board size ∼ direct influence on probability 3 Weighting • Adjacency indicator or • Point % ownership weights • Normalize to subsidiaries equalized Gender diversity spillovers through networks 18 / 26

64. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Are there spillovers? 1 Q1: Are there spillovers in ownership networks? • subsidiary i in t • parent j in t-1 • i ← j GBDi,t ∼ GBDj,t−1 + controlsi,t + controlsj,t−1 + ϵi,j,t (1) 2 Q2: Imitation vs hierarchy? • (k, i) : k ∼ i nodes connected by path • GapI (i) = GBDi,t−2 − 1 |{k:k∼i|} ∑ k:k∼i GBDk,t−2 • GapO(i) = GBDi,t−2 − 1 |{k:k→i|} ∑ k:k→i GBDk,t−2 ∆GBDi,t ∼ GapI (i) + controlsi,t + ϵi,j,t (2) ∆GBDi,t ∼ GapI (i) + GapO (i) + controlsi,t + ϵi,j,t (3) ∆GBDi,t ∼ 1(Gap(i) > 0)Gap(i) + 1(Gap(i) < 0)Gap(i) + controlsi,t + ϵi,j,t (4) Gender diversity spillovers through networks 19 / 26

67. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Ownership vs interlocking: diﬀerent vibe • Average gap for interlocks: -0.05 p.p. • # > 0: 1,523,293; # < 0: 1,526,275 • Average gap for ownerships: -0.02 p.p. • # > 0: 842,912; # < 0: 614,752 Gender diversity spillovers through networks 20 / 26

68. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Ownership vs interlocking: diﬀerent, but the same Gender diversity spillovers through networks 21 / 26

70. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Table 4: Spillovers in Ownership: presence ∼ presence Woman in boards in t Subsidiary FE Parent FE Parent & subsidiary FE (1) (2) (3) Panel A: parent-subsidiary link per se Woman in parent’s boardroom in t − 1 (β) 0.083*** 0.13*** 0.11*** (0.000) (0.000) (0.000) ÷ of firms w/ 1+ woman (Subsidiary) 0.47 0.47 0.47 # of observations 3,460,667 3,527,196 3,417,196 # of firms (Parent) 682,626 658,017 644,536 # of firms (Subsidiary) 808,807 906,690 801,017 Panel B: parent-subsidiary link adjusted for ownership share Woman in parent’s boardroom in t − 1 (β) 0.10*** 0.14*** 0.11*** (0.000) (0.000) (0.000) ÷ of firms w/ 1+ woman (Subsidiary) 0.47 0.47 0.47 # of observations 3,274,224 3,340,739 3,236,418 # of firms (mother) 664,191 643,751 630,932 # of firms (Subsidiary) 785,191 878,938 778,526 • The same qualitative interpretation for shares ∼ shares Gender diversity spillovers through networks 23 / 26

71. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Table 5: Gaps ∆t,t−1÷ (1) (2) (3) Ownership Gap t − 2 -0.040*** -0.13*** -0.13*** Sample size 2,647,056 Interlocking Gap t − 2 -0.027*** -0.095*** -0.089*** Sample size 2,749,049 Intersection Gap interlocking t − 2 -0.018*** -0.086*** -0.082*** Gap ownership t − 2 -0.029*** -0.076*** -0.088*** Sample size 1,208,513 Year FE board size Firm FE Sector FE & hhi TOAS ∪ OPRE ∪ EMPL Gender diversity spillovers through networks 24 / 26

72. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Table 6: Asymmetric gaps ∆t,t−1÷ (1) (2) (3) Ownership Gap > 0 t − 2 -0.062*** -0.18*** -0.18*** Gap < 0 t − 2 -0.024*** -0.083*** -0.085*** Sample size 2,647,056 Interlocking Gap > 0 t − 2 -0.039*** -0.15*** -0.14*** Gap < 0 t − 2 -0.016*** -0.057*** -0.053*** Sample size 2,749,049 Year FE board size Firm FE Sector FE & hhi TOAS ∪ OPRE ∪ EMPL Gender diversity spillovers through networks 25 / 26

73. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References What do we learn from this? 1 Policy makers need to make informed decisions on the social and economic dillema • ought to know mechanics of GBD spillovers (channel, capacity, eﬀect) • identify mechanisms, predict outcome, implement and calibrate policies 2 We are ﬁrst to present them in ownership network medium 1 There is a positive correlation between parents’ and subsidiaries’ diversities in the network 2 Hierarchy and peers seem to have similar size of correlations 3 These convergences work in both directions, but with asymmetric strength 3 This fertile ground presents great new opportunities for theoretical - qualitative and modeling, as well as empirical research for the future Gender diversity spillovers through networks 26 / 26

81. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Bossler, M., Mosthaf, A. and Schank, T.: 2020, Are female managers more likely to hire more female managers? evidence from Germany, ILR Review 73(3), 676–704. Bozhinov, V., Joecks, J. and Scharfenkamp, K.: 2021, Gender spillovers from supervisory boards to management boards, Managerial and Decision Economics 42(5), 1317–1331. Brown, C., Daly, A. and Liou, Y.-H.: 2016, Improving trust, improving schools: Findings from a social network analysis of 43 primary schools in england, Journal of Professional Capital and Community 1(1), 69–91. Cook, A. and Glass, C.: 2015, Diversity begets diversity? the effects of board composition on the appointment and success of women CEOs, Social Science Research 53, 137–147. Drazkowski, H., Timmermans, B. and Tyrowicz, J.: 2023, Gender board diversity spillovers and the public eye, GRAPE Working Paper 90. Drazkowski, H., Tyrowicz, J. and Zalas, S.: 2024, Gender board diversity across Europe throughout four decades, Nature Scientific Data 11(1), 567–590. Dustmann, C., Glitz, A., Schönberg, U. and Brücker, H.: 2016, Referral-based job search networks, The Review of Economic Studies 83(2), 514–546. Farrell, K. and Hersch, P.: 2005, Additions to corporate boards: The effect of gender, Journal of Corporate Finance 11, 85–106. Garcia-Blandon, J., Argilés-Bosch, J. M., Ravenda, D. and Castillo-Merino, D.: 2023, Direct and spillover effects of board gender quotas: Revisiting the Norwegian experience, Business Ethics, the Environment & Responsibility 32(4), 1297–1309. Gender diversity spillovers through networks 26 / 26

82. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Ghoshal, S. and Bartlett, C. A.: 1990, The multinational corporation as an interorganizational network, The Academy of Management Review 15(4), 603. Gimeno, R., Mateos de Cabo, R., Grau, P. and Gabaldon, P.: 2022, Network diffusion of gender diversity on boards: A process of two-speed opposing forces, PLoS One 17(11), e0277214. Gould, J. A., Kulik, C. T. and Sardeshmukh, S. R.: 2018, Trickle-down effect: The impact of female board members on executive gender diversity, Human Resource Management 57(4), 931–945. Granovetter, M.: 1995, Coase revisited: Business groups in the modern economy, Industrial and corporate change 4(1), 93–130. Granovetter, M. S.: 1973, The strength of weak ties, American journal of sociology 78(6), 1360–1380. Guldiken, O., Mallon, M. R., Fainshmidt, S., Judge, W. Q. and Clark, C. E.: 2019, Beyond tokenism: How strategic leaders influence more meaningful gender diversity on boards of directors, Strategic Management Journal 40(12), 2024–2046. Hensvik, L. and Skans, O. N.: 2016, Social networks, employee selection, and labor market outcomes, Journal of Labor Economics 34(4), 825–867. Holzer, H. J.: 1988, Search method use by unemployed youth, Journal of labor economics 6(1), 1–20. Kirsch, A. and Wrohlich, K.: 2020, More women on supervisory boards: Increasing indications that the effect of the gender quota extends to executive boards, DIW Weekly Report 10(4/5), 44–49. Kramarz, F. and Skans, O. N.: 2014, When strong ties are strong: Networks and youth labour market entry, Review of Economic Studies 81(3), 1164–1200. Gender diversity spillovers through networks 26 / 26

83. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References Kunze, A. and Miller, A. R.: 2017, Women helping women? Evidence from private sector data on workplace hierarchies, Review of Economics and Statistics 99(5), 769–775. Maida, A. and Weber, A.: 2022, Female leadership and gender gap within ﬁrms: Evidence from an Italian board reform, ILR Review 75(2), 488–515. Mateos de Cabo, R., Gimeno, R., Gabaldón, P. and Grau, P.: 2024, The board gender diversity imitation game: Uncovering the resistant boards that refuse to play, Corporate Governance: An International Review . Matsa, D. A. and Miller, A. R.: 2011, Chipping away at the glass ceiling: Gender spillovers in corporate leadership, American Economic Review 101(3), 635–39. Mizruchi, M. S.: 1996, What do interlocks do? an analysis, critique, and assessment of research on interlocking directorates, Annual Review of Sociology 22(1), 271–298. Montgomery, J. D.: 1991, Social networks and labor-market outcomes: Toward an economic analysis, The American economic review 81(5), 1408–1418. Mortensen, D. T. and Vishwanath, T.: 1994, Personal contacts and earnings: It is who you know!, Labour economics 1(2), 187–201. Schoonjans, E., Hottenrott, H. and Buchwald, A.: 2024, Welcome on board? appointment dynamics of women as directors, Journal of Business Ethics 192(3), 561–589. Smith, N. and Parrotta, P.: 2018, Why so few women on boards of directors? Empirical evidence from Danish companies in 1998–2010, Journal of Business Ethics 147(2), 445–467. Tyrowicz, J., Bech, K. and Zalas, S.: 2024, Revisiting gender board diversity and ﬁrm performance, FAME|GRAPE Working paper . Gender diversity spillovers through networks 26 / 26

84. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Data (GBDD and ownership) Methods (panel linear model) Results References von Essen, E. and Smith, N.: 2023, Network connections and board seats: are female networks less valuable?, Journal of Labor Economics 41(2), 323–360. Yoshikawa, T., Shim, J. W., Kim, C. H. and Tuschke, A.: 2019, How do board ties aﬀect the adoption of new practices? the eﬀects of managerial interest and hierarchical power, Corporate Governance: An International Review 28(1), 2–22. Gender diversity spillovers through networks 26 / 26

Spillovers_in_Ownership_Interlocking_Networks.pdf

More Related Content

Similar to Spillovers_in_Ownership_Interlocking_Networks.pdf (20)

More from GRAPE (20)

Recently uploaded (20)

Spillovers_in_Ownership_Interlocking_Networks.pdf