The quality of the relationship between employees and management

by Sunwoo Hwang, Assistant Professor of Finance at Korea University Business School and Biwon Lee, finance PhD candidate at WP Carey School of Business at Arizona State University

This paper asks whether it pays to invest in labour relations. Unlike the traditional view of labour as a production input like capital, this study recognises that employees may not be passive contributors but actively react to the perceived quality of their relationship with management. When employees’ perceptions are more positive than those of management, employees may exert greater effort than explicitly contracted through wages and non-wage benefits or choose to remain with their current employers instead of transitioning to similar firms.

Challenges in study

Addressing the research question faces 2 primary challenges. Firstly, the intangible and bilateral nature of labour-management relationships makes it difficult to measure their quality. Secondly, even if we can measure it, it is also difficult to randomly allocate varying levels of this intangible and study its causal effects on business outcomes. We overcome these obstacles by using data from surveys that pose the same set of questions to both employees and management about their relationship and exploiting within-sampling-unit variation in the intangible, which is random conditional on assumptions discussed and tested in the paper.

Key findings and implications

Using a proxy for labour-management relational quality perceived by employees relative to management (hereafter, relational quality) and a strategy for causal inference, we find a positive association between relational quality and a firm’s future profitability, productivity, and retention. Notably, this association is not observed when considering relational quality perceived by employees or management alone. These findings carry 2 implications. Firstly, relational quality, closely intertwined with corporate culture and influencing employee satisfaction, has the potential to generate long-term value and be viewed as relational capital. Secondly, bilateral intangible factors, such as labour relations, should be evaluated based on input from both parties. In the absence of data enabling such a bilateral evaluation, prior research and ESG rating agencies have evaluated these factors unilaterally, relying on either the perspectives of employees or management gathered through surveys or information disclosed by management through regulatory filings or websites.

Measuring relational quality

To quantify relational quality, we compute the difference between numerical responses given by employees and management on a 3-to-5-point scale for each survey question. We then aggregate these score differences by computing their principal components (PCs) and take the first PC that we show represents significant variation in these score differences as our proxy for relational quality. The proxy reflects the reality where employees form their perception taking managerial perception communicated within the organisation as a benchmark. It thereby captures the relative perception of employees, which may go unnoticed by management and potentially lead to unforeseen improvements in labour productivity or employee retention.

Research design

Our causal inference capitalises on variation in relational quality within each sampling unit. Since businesses are randomly sampled to constitute each unit for survey purposes, their selection is inherently random. However, there is a potential issue of omitted variable bias, where other characteristics within each business may correlate with relational quality and subsequently influence outcomes such as profitability, productivity, and retention. To this end, we show that observable characteristics do not significantly differ between businesses with high and low relational quality, mitigating concerns about omitted variable bias. Moreover, to address the possibility of reverse causality, where business outcomes might affect relational quality, we estimate the association between relational quality and future outcomes, after showing that the associations do not simply reflect their contemporaneous associations.

Validating the proxy for relational quality

We validate our proxy by demonstrating its strong correlations with labour relations events, which we believe have stronger effects on employees’ perceptions than those of management. The proxy is positively correlated with agreement between employees and management on wage issues, working conditions, and job security. Conversely, the proxy is negatively correlated with disagreement on a wage increase rate and actions against management. Furthermore, we show that these correlations are driven by employees’ perceptions.

The effects of relational quality on profitability, labour productivity, and employee retention

Using the proxy, we show that relational quality exhibits a positive and statistically significant association with future profitability. This finding aligns with prior research showing that firms with higher employee ratings enjoy greater stock valuation. While such increased valuation could be due to anticipated future cash flows or market mispricing, our results lend support to the former explanation. In contrast, when considering relational quality perceived by management alone, we find no significant association with future profitability.

This observation is in line with other research showing that management’s proclaimed values are not significantly linked to operating performance. Next, we decompose relational quality into 3 distinct dimensions of culture, policy, and dispute and show that the cultural aspect accounts for the positive association between relational quality and future business outcomes. Regarding channels, we find evidence supporting labour productivity and employee retention. Specifically, relational quality is positively associated with future labour productivity and negatively with future employee turnover, with the latter being due to an improvement in retention rather than recruiting.

Underlying mechanisms

Our analysis of the varying employee responses and the varying impacts of relational quality offer further insights into underlying mechanisms. Firstly, the observed improvement in retention is driven by inexperienced employees as opposed to experienced ones, suggesting that employees with less firm-specific skills and lower switching costs are more responsive to relational quality. Secondly, reinforcing this notion, our results are (resp. are not) more pronounced in businesses that invest in the human capital development of rank-and-file employees (resp. middle managers or high-skill workers). Lastly, our findings are stronger in privately held businesses, which typically face challenges in attracting and retaining employees compared to their public counterparts.

Implications for the divergence of ESG ratings

Lastly, we elicit the implications for ESG rating divergence. We find that management-perceived relational quality is positively and significantly correlated with the S (social) score, whereas relational quality or employee-perceived relational quality does not show such a correlation. This observation leads us to reinterpret the lack of significant associations between management-perceived relational quality and future outcomes; bilateral attributes based on one-sided evaluations, which constitute the S score, may exhibit insignificant associations with future business outcomes, thus providing unsatisfactory references for firms and investors seeking to capitalise on them.