State-Level Culture and Workplace Diversity Policies: Evidence from US Firms

For example, countries such as Saudi Arabia and Iran punish same-sex relationships with the death penalty. https://​www.​amnestyusa.​org/​the-state-of-lgbt-rights-worldwide/​ (accessed on Nov 01, 2019).

To improve the workplace environment and eliminate discrimination against LGBT employees, the UN High Commissioner for Human Rights released new standards of conduct in 2017 (Hossain et al. 2019). These standards (including respect, elimination, and prevention of discrimination, support, and a stand for LGBT individuals) advocate for an open and broadminded workplace environment in which employees are not discriminated based on their sexual orientation or gender identity.

Over the last two decades, there has been a substantial increase in calls for supporting and protecting LGBT rights in different countries, including Australia, Europe, UK, and the US (Lloren and Parini 2017; Pichler et al. 2018).

Williamson (2000) suggests an analytical framework of economic and social analysis that consists of four levels where the informal institution is the most fundamental level (Level 1).

Refer to Vandello and Cohen (1999) for differences in state-level culture.

Scholars ponder whether the constructs of individualism and collectivism are orthogonal dimensions or opposite ends of a single dimension. Triandis (1989) states that both interpretations of the individualism–collectivism construct can be correct, based on the context of research. Existing empirical studies interpret the individualism–collectivism construct in a unidimensional manner (e.g., Chen et al. 2015; Gorodnichenko and Roland 2017; Hofstede 1980; Li et al. 2013). Therefore, we treat individualism and collectivism as opposite ends of a unidimensional scale as suggested by Vandello and Cohen (1999).

De Mooij and Hofstede (2010) show that individuals view themselves as more important in decision making than the group.

This is somewhat relevant to the masculinity vs. femininity dimension of Hofstede’s (2001) cultural values and indicates that organizational cultures that score high on femininity are likely to support workplace diversity policies.

Partial support to this argument is provided by Cook and Glass (2016), who find that firms with female directors in their boardrooms are more likely to introduce workplace diversity policies than firms with no women in their boardrooms.

The HRC is the leading organization on reporting of LGBT progress within the US (Cook and Glass 2016). The Corporate Equality Index (CEI) score includes adoption of sexual orientation non-discrimination policies, adoption of gender-identity non-discrimination policies, in addition to having domestic partner benefits and transgender policies from 2011 to 2014. The HRC survey covers large firms in the US (i.e., Standard and Poor’s 500, Forbes’ list of the 200 largest privately held firms, and the Fortune 500 publicly traded firms). The HRC reports during the sample period are based on similar scoring criteria. However, we acknowledge the limitations associated with the CEI measure due to certain perceptions about the HRC, state-level culture index, and data availability relating to small firms.

We do not report the VIF table for the sake of brevity.

Lardaro (1993) suggests that multicollinearity may be a concern if VIF exceeds 10.

Industry-wise detail of the sample is presented in Table 11 in the appendix.

We also use multiple matched firms by modifying the difference in propensity scores to 1.0% and 0.5% in value (Atif et al. 2019a, b); nevertheless, our un-tabulated results are consistent.

Mean difference between the treatment and control group is based on the average treatment effect on the treated (ATT).

The system GMM estimations are performed using Roodman (2009) Stata module ‘xtabond2.′ Refer to Roodman (2009) and Pathan (2009) for details on dynamic panel data estimations.