In this presenation Pamela Campa (Assistant Professor at SITE) discussed the under-representation of women in power and ongoing inequality of women in politics.

1. Women in power
Pamela Campa
March 6, 2020

2. Women under-represented at the top of hierarchical governmental organizations

3. Women under-represented at the top of hierarchical governmental organizations
Dearth of women “in power”

4. Corporate sector: US

5. Corporate sector: Europe
Board members in largest listed companies

6. Corporate sector: Europe
CEOs in largest listed companies

7. Corporate sector: Sweden over time
Presidents of boards in largest listed companies

8. Gender Diversity Index 2019: Sweden
Figure: European Women on Boards Gender Diversity Index (2019)
Note:Index calculated for STOXX Europe 600

9. Dearth of women in powerful position...

10. Dearth of women in powerful position...
Problem not limited to the corporate sector; similar patterns in politics, academia

11. Women in power in politics
Women 15.4 % of mayors in EU-28 countries in 2019
30% of ministers in EU. Positive trend, with ups and downs
Ursula von der Leyen first female President of the European Commission, starting on Nov 2019
Commission established by the treaty of Rome, 1957
In Europe 5 heads of State and 3 heads of government are currently women
Never a woman as US President. Neither as Swedish prime minister

12. Three common arguments for why we should care about
dearth of women in power (see Bertrand 2019)

13. Three common arguments for why we should care about
dearth of women in power (see Bertrand 2019)
1. Right and fairness

14. Three common arguments for why we should care about
dearth of women in power (see Bertrand 2019)
1. Right and fairness
2. Selection/Performance

15. Three common arguments for why we should care about
dearth of women in power (see Bertrand 2019)
1. Right and fairness
2. Selection/Performance
3. Representativeness

16. Three common arguments for why we should care about
dearth of women in power (see Bertrand 2019)
1. Right and fairness
2. Selection/Performance
3. Representativeness

17. Right and fairness
Argument boils down to whether we believe that women have (or don’t have) same
opportunities as men to ascend to leadership

18. Right and fairness
Argument boils down to whether we believe that women have (or don’t have) same
opportunities as men to ascend to leadership
Discrimination hard to measure. Are we concerned only about taste-based discrimination, or
also about statistical discrimination? How can we ever assess whether two people are equal in
terms of potential to enhance the organization’s goals?

19. Discrimination hard to measure. But indirect evidence that women, at least in some
circumstances, face higher (at least perceived) costs than men when they ascend to leadership

20. Indirect evidence of higher costs for women in leadership
In a field experiment, single female students reported lower desired salaries and willingness to
travel and work long hours when they expected classmates to see their preferences. Other
groups’ response unaffected by peer observability. A second experiment indicates the effects
are driven by observability by single male peers. (Bursztyn et al. 2017)
Female supervisors more likely to suffer sexual harassment than female employees, in the US,
Sweden and Japan, especially in male-dominated industries and firms. Harassment of
supervisors also followed by more negative professional and social consequences (Folke et al.
2020).
Promotion to a top position - mayor or CEO - doubles the probability of divorce for women in
Sweden - not for men (Folke and Rickne 2020)

21. Other considerations/open questions about equality in opportunity
1. Some would argue that if process of selection is meritocratic we should not worry about gender
inequalities. But how well can we really measure performance? Research shows that women
are more risk-adverse and less competitive, and value flexibility in working hours more. This
may hamper their opportunities to be promoted (desire for flexibility appears particularly
important, see Goldin (2016)). What is the evidence that attitudes toward risk, competitiveness,
and inflexible hours further corporations’ goals?
2. Some of the behavioural gender differences appear to be determined more by nurture than by
nature (Gneezy et al., 2008). How do we think about equal opportunities when we socialize
boys and girls differently since childhood?
3. Gender inequalities may be self-reinforcing. Women in male-dominated teams are less
confident in their relative performance, less influential, more swayed by others in team
discussions, less likely to run and to be voted for leadership positions (Born et al., 2018). What
if past discriminations change opportunity set for women today, even in a world in which
discrimination is no longer prevalent?
4. What do we mean by meritocratic process? What if networks are important (as they appear to
be)?

22. Three common arguments for why we should care about
dearth of women in power (see Bertrand 2019)
1. Right and fairness
2. Selection/Performance
3. Representativeness

23. Selection/Performance
1. Does increased diversity improve:
Selection?

24. Selection/Performance
1. Does increased diversity improve:
Selection? ⇒

25. Selection/Performance
1. Does increased diversity improve:
Selection? ⇒ Performance?

26. Selection/Performance
1. Does increased diversity improve:
Selection? ⇒ Performance?
2. Other channels through which diversity can improve performance?

27. Diversity and Selection. The theoretical argument
Bertrand (2019): “All should agree that an economy that is tapping into a limited pool (men) to
find its leaders must be operating inside the efficiency frontier.”
Angela Merkel: “We can’t afford to do without the skills of women.”
Underlying assumption: talent equally distributed between women and men. Do we have
reasons to believe that it is not? If women were less talented than men, somehow inherently,
how do we explain massive variation in women’s achievement across countries and over time?
- Hsieh et al. (2019). In 1960, in the US 94% of doctors and lawyers were men. By 2010, the fraction was just 62%. Has distribution of talent across groups
changed? Likely not. Then, “a substantial pool of innately talented women and black men in 1960 were not pursuing their comparative advantage.”

28. Diversity and selection: what is the evidence?
Evidence from board quotas:
Board quotas in Norway: average quota of board members, as expressed by a human capital
index, increased throughout the period (Bertrand et al., 2018)
Board quotas in Italy: quotas associated with higher levels of education of board members
(Ferrari et al., 2018)

29. Diversity and selection: what is the evidence?
Evidence from electoral quotas:
Italy: ⇑ education - women more educated, they displace less educated men (Baltrunaite,
Bello, Casarico and Profeta (2014))

30. Diversity and selection: what is the evidence?
Evidence from electoral quotas:
Italy: ⇑ education - women more educated, they displace less educated men (Baltrunaite,
Bello, Casarico and Profeta (2014))
Sweden:⇑ competence of male politicians; resignation of male mediocre leaders (“the crisis of
the mediocre man”) (Besley, Folke, Persson and Rickne, 2017)

31. Diversity and selection: what is the evidence?
Evidence from electoral quotas:
Italy: ⇑ education - women more educated, they displace less educated men (Baltrunaite,
Bello, Casarico and Profeta (2014))
Sweden:⇑ competence of male politicians; resignation of male mediocre leaders (“the crisis of
the mediocre man”) (Besley, Folke, Persson and Rickne, 2017)
Spain, small municipalities:= education (Bagues and Campa, 2018)

32. Diversity and selection: what is the evidence?
(To the best of my knowledge) no evidence from Western democracies that electoral quota
decreases quality, in spite of the stubbornly persistent objection against gender quotas.
Note: not obvious what is a good measure of “quality”

33. Other potential channels of influence from diversity to performance
Important read: Iris Bhonet, What Works

34. Other potential channels of influence from diversity to performance
Important read: Iris Bhonet, What Works
Wooley et al. (2010): measure “collective intelligence” of randomly formed groups that engaged
in a number of tasks. Two important findings:
1. Collective intelligence better predictor of future performance than the sum of individual
abilities. Important to build good groups
2. Equal opportunities to speak, higher score on social sensitivity, and higher share of
women positively associated with collective intelligence ⇒ women may be an important
“glue” to connect different parts of a group

35. Other potential channels of influence from diversity to performance
Important read: Iris Bhonet, What Works
Wooley et al. (2010): measure “collective intelligence” of randomly formed groups that engaged
in a number of tasks. Two important findings:
1. Collective intelligence better predictor of future performance than the sum of individual
abilities. Important to build good groups
2. Equal opportunities to speak, higher score on social sensitivity, and higher share of
women positively associated with collective intelligence ⇒ women may be an important
“glue” to connect different parts of a group
Diversity of viewpoints likely important when a task involves collective problem-solving. When
coordination is needed, there may be a case for homogeneity. When listening and
bridge-building is important, having more women in the group likely helps. When the goal is
individual productivity, important to consider peer-effects
⇒ Positive girl effect in schools - both girls and boys perform better when there are more
girls in the classroom. Implications?

36. Diversity and performance. The evidence
Does having more women on top of organizations improve performance?

37. Diversity and performance. The evidence
Does having more women on top of organizations improve performance?
Evidence, based on correlations and introduction of quotas, is mixed. A clear relationship
between gender composition of the board and company performance has not been
convincingly established. Based on available findings, the “business argument” for quota in
boards is not necessarily strong.

38. Diversity and performance. The evidence
Does having more women on top of organizations improve performance?
Evidence, based on correlations and introduction of quotas, is mixed. A clear relationship
between gender composition of the board and company performance has not been
convincingly established. Based on available findings, the “business argument” for quota in
boards is not necessarily strong.
Women in politics: closely elected female mayors in Brazil engage less in corruption and
political patronage, and have lower probability to be re-elected (Brollo and Troiano, 2015).

39. Diversity and performance. Some considerations on the existing
evidence
Correlations are difficult to interpret

40. Diversity and performance. Some considerations on the existing
evidence
Correlations are difficult to interpret
but also studying the impact of quotas poses some problems: when quotas are introduced the
board becomes more gender diverse, but also more “new”; how long a team worked together
appears to matter for performance

41. Diversity and performance. Some considerations on the existing
evidence
Correlations are difficult to interpret
but also studying the impact of quotas poses some problems: when quotas are introduced the
board becomes more gender diverse, but also more “new”; how long a team worked together
appears to matter for performance
Technically very difficult to assess impact of more women in CEO positions, for several reason,
most important one perhaps being that they are too few

42. Diversity and performance. Some considerations on the existing
evidence
Correlations are difficult to interpret
but also studying the impact of quotas poses some problems: when quotas are introduced the
board becomes more gender diverse, but also more “new”; how long a team worked together
appears to matter for performance
Technically very difficult to assess impact of more women in CEO positions, for several reason,
most important one perhaps being that they are too few
What makes a group gender diverse? Important to consider critical mass theories and risk of
tokenism

43. Three common arguments for why we should care about
dearth of women in power (see Bertrand 2019)
1. Right and fairness
2. Selection/Performance
3. Representativeness

44. The representativeness argument

45. The representativeness argument
Are women’s preferences adequately represented in
male-dominated organizations?

46. The representativeness argument: evidence from politics
Women appear to have different preferences than men

47. The representativeness argument: evidence from politics
Women appear to have different preferences than men
Spain, survey: women more likely than men to report that unemployment, pensions, education,
the status of the health system, drugs, youth problems, violence against women, women’s
problems in general, and social issues are a main concern to them. Men are significantly more
concerned about housing, immigration, work conditions, politics, corruption, the status of
infrastructure, environmental degradation, the judiciary system and agriculture, hunting and
fishing (Bagues and Campa, 2018).

48. The representativeness argument: evidence from politics
Women appear to have different preferences than men
Spain, survey: women more likely than men to report that unemployment, pensions, education,
the status of the health system, drugs, youth problems, violence against women, women’s
problems in general, and social issues are a main concern to them. Men are significantly more
concerned about housing, immigration, work conditions, politics, corruption, the status of
infrastructure, environmental degradation, the judiciary system and agriculture, hunting and
fishing (Bagues and Campa, 2018).
Switzerland, reported voting behavior in referenda: women show less support for increasing
retirement age, nuclear energy, the military; more support for environmental protection, healthy
life-style, equal rights for women, assistance to disabled (Funk and Gathmann, 2015)

49. The representativeness argument: evidence from politics
Women appear to have different preferences than men
Spain, survey: women more likely than men to report that unemployment, pensions, education,
the status of the health system, drugs, youth problems, violence against women, women’s
problems in general, and social issues are a main concern to them. Men are significantly more
concerned about housing, immigration, work conditions, politics, corruption, the status of
infrastructure, environmental degradation, the judiciary system and agriculture, hunting and
fishing (Bagues and Campa, 2018).
Switzerland, reported voting behavior in referenda: women show less support for increasing
retirement age, nuclear energy, the military; more support for environmental protection, healthy
life-style, equal rights for women, assistance to disabled (Funk and Gathmann, 2015)
USA, extension of suffrage to women: immediate increases in state government expenditures
and revenue and more liberal voting patterns for federal representatives (Lott and Kenny, 1999)

50. Do differences in preferences translate into different policy
decisions?

51. 2 types of studies:
Gender quotas
Woman wins against man by narrow margin

52. 2 types of studies:
Gender quotas
Woman wins against man by narrow margin
Evidence is mixed

53. Gender and policy

54. Gender and policy
India: seat reservation for female head of villages; women prioritize more on areas where
female voters are more likely to bring complains, e.g. drinking water infrastructure
(Chattopadhyay and Duflo, 2004)

55. Gender and policy
India: seat reservation for female head of villages; women prioritize more on areas where
female voters are more likely to bring complains, e.g. drinking water infrastructure
(Chattopadhyay and Duflo, 2004)
India: women elected in the State Legislatures, policy changes; caste matters for type of
change (Clots-Figueras, 2011)

56. Gender and policy
India: seat reservation for female head of villages; women prioritize more on areas where
female voters are more likely to bring complains, e.g. drinking water infrastructure
(Chattopadhyay and Duflo, 2004)
India: women elected in the State Legislatures, policy changes; caste matters for type of
change (Clots-Figueras, 2011)
India: women elected in the State Legislatures, higher education in corresponding districts,
only in rural areas (Clots-Figueras, 2012)

57. Gender and policy
India: seat reservation for female head of villages; women prioritize more on areas where
female voters are more likely to bring complains, e.g. drinking water infrastructure
(Chattopadhyay and Duflo, 2004)
India: women elected in the State Legislatures, policy changes; caste matters for type of
change (Clots-Figueras, 2011)
India: women elected in the State Legislatures, higher education in corresponding districts,
only in rural areas (Clots-Figueras, 2012)
India: women elected in the State Legislatures, more public health facilities, increase in
antenatal care visits, institutional delivery, and breastfeeding (Bhalotra and Clots-Figueras,
2014)

58. India: seat reservation in States Legislatures, large and significant rise in documented crimes
against women - reporting ⇑ (Iyer, Mani, Mishra, and Topalova, 2015)

59. India: seat reservation in States Legislatures, large and significant rise in documented crimes
against women - reporting ⇑ (Iyer, Mani, Mishra, and Topalova, 2015)
Brazil: elected female mayors, less corruption and less political patronage, lower re-election
probability (Brollo and Troiano, 2015)

60. India: seat reservation in States Legislatures, large and significant rise in documented crimes
against women - reporting ⇑ (Iyer, Mani, Mishra, and Topalova, 2015)
Brazil: elected female mayors, less corruption and less political patronage, lower re-election
probability (Brollo and Troiano, 2015)
USA: elected female mayors, no impact on size and composition of expenditures and crime
rates (Ferreira and Gyourko, 2014)

61. India: seat reservation in States Legislatures, large and significant rise in documented crimes
against women - reporting ⇑ (Iyer, Mani, Mishra, and Topalova, 2015)
Brazil: elected female mayors, less corruption and less political patronage, lower re-election
probability (Brollo and Troiano, 2015)
USA: elected female mayors, no impact on size and composition of expenditures and crime
rates (Ferreira and Gyourko, 2014)
Spain: candidate gender quotas for municipal councillors, no significant changes in budget and
socio-economic indicators (Bagues and Campa, 2018)

62. Rich evidence from India that gender of policy-makers matters

63. Rich evidence from India that gender of policy-makers matters
Surprisingly little evidence from Western democracies.

64. Rich evidence from India that gender of policy-makers matters
Surprisingly little evidence from Western democracies.

65. Dearth of women in power. Issue in terms of equality,
efficiency, and representativeness.

66. Dearth of women in power. Issue in terms of equality,
efficiency, and representativeness.
What to do?

67. 1. Act on the pipeline
Evidence of gender differences in preferences over work arrangements. Explore opportunities for
flexible working hours, monitor performance carefully

68. 1. Act on the pipeline
Evidence of gender differences in preferences over work arrangements. Explore opportunities for
flexible working hours, monitor performance carefully
Family-relational considerations: role of policy - promote more egalitarian gender roles (paternity
leaves? childcare? role models?)

69. 2. “Stimulate demand”

70. 2. “Stimulate demand”
Focus on diversity in teams especially when goal is problem-solving. Important to promote diversity
when it is likely to pay off the most

71. 2. “Stimulate demand”
Focus on diversity in teams especially when goal is problem-solving. Important to promote diversity
when it is likely to pay off the most
Quota or not Quota?

72. What do we know about quotas?

73. What do we know about quotas?
They tend to increase women’s representation in boards or in political office if they are properly
designed (strategic positioning: “zipper” quota or double-preference system preferable)

74. What do we know about quotas?
They tend to increase women’s representation in boards or in political office if they are properly
designed (strategic positioning: “zipper” quota or double-preference system preferable)
Effects above and beyond the quota mandate? Plausible given recent evidence of “gendered
group dynamics”. Mixed evidence in politics from different contexts (India, Italy, Spain, Sweden)

75. What do we know about quotas?
They tend to increase women’s representation in boards or in political office if they are properly
designed (strategic positioning: “zipper” quota or double-preference system preferable)
Effects above and beyond the quota mandate? Plausible given recent evidence of “gendered
group dynamics”. Mixed evidence in politics from different contexts (India, Italy, Spain, Sweden)
Evidence from Norwegian board quota: little discernible impact on women in business beyond
direct effect on women who made it into the boardroom

76. What do we know about quotas?
They tend to increase women’s representation in boards or in political office if they are properly
designed (strategic positioning: “zipper” quota or double-preference system preferable)
Effects above and beyond the quota mandate? Plausible given recent evidence of “gendered
group dynamics”. Mixed evidence in politics from different contexts (India, Italy, Spain, Sweden)
Evidence from Norwegian board quota: little discernible impact on women in business beyond
direct effect on women who made it into the boardroom
Policy changes hard to measure. Also not obvious what should be the size of the mandated
increase in female representation to achieve policy change. Too little evidence from Western
democracies to draw conclusions

77. What do we know about quotas?
They tend to increase women’s representation in boards or in political office if they are properly
designed (strategic positioning: “zipper” quota or double-preference system preferable)
Effects above and beyond the quota mandate? Plausible given recent evidence of “gendered
group dynamics”. Mixed evidence in politics from different contexts (India, Italy, Spain, Sweden)
Evidence from Norwegian board quota: little discernible impact on women in business beyond
direct effect on women who made it into the boardroom
Policy changes hard to measure. Also not obvious what should be the size of the mandated
increase in female representation to achieve policy change. Too little evidence from Western
democracies to draw conclusions
No evidence that “quality” of politicians and boards deteriorates. On the contrary!

78. Quota or not Quota?

79. Why not. But..

80. Why not. But..
Important to understand what is the most efficient design,
given context - marginal increases? big push? risk of
tokenisms? critical mass?

81. Why not. But..
Important to understand what is the most efficient design,
given context - marginal increases? big push? risk of
tokenisms? critical mass?
In politics, quotas unlikely to be “enough” to increase
“substantial representation”. To be combined with other
measures, which also consider the supply side of the “market”
and role of parties.

82. Other potentially important factor: women seem to be more affected by negative feedback. What
happens on social media?

83. Monitor developments, collect more data and information (technology can help).

84. Monitor developments, collect more data and information (technology can help).
Do not become easily satisfied with positive developments. Progress not always linear.

85. A lot of work still ahead of us...

86. A lot of work still ahead of us...

87. Thank you!
pamela.campa@hhs.se Twitter: @PAMELACAMPA1