Do NPM Strategies Lead to Negative Organizational Behavior? Lessons from the Differential Effects of Contracting Out on Voluntary Turnover

Gyeo Reh Lee

doi:10.20899/jpna.7.3.369-389

Abstract

While public sector organizations have increasingly utilized New Public Management (NPM) strategies as a means of increasing the values of the market, a growing body of literature suggests that market-based reforms may generate indirect costs associated with negative organizational behaviors in the public sector. Focusing on probable consequences of government contracting out for the public workforce, this study examines the relationship between contracting out and voluntary turnover relying on a panel data of U.S. federal agencies from 2010 to 2017. The results present that contracting activity is associated with voluntary quits in the opposite direction depending on the level of job satisfaction. This finding disentangles the previous discussion on the relationship between NPM strategies and employee behavior.

Highlights

Market-based reforms to make public organizations more business-like have been global phenomena for several decades
This study develops and tests a model that examines the independent effect of contracting out on employees’ voluntary turnover, and the relationship given the levels of job satisfaction in organizations
Among the types of turnover, this study focuses on voluntary turnover, in which employees hold higher human and social capital as compared to those who are involuntarily terminated from their positions by their employers for their poor performance or misconduct

Summary

Introduction

Market-based reforms to make public organizations more business-like have been global phenomena for several decades. This study develops and tests a model that examines the independent effect of contracting out on employees’ voluntary turnover, and the relationship given the levels of job satisfaction in organizations. This would help public managers find appropriate strategies to address probable effects of those practices. This study employs panel-data analyses along with eight years of data from the Federal Employee Viewpoint Survey, Fedscope, and the Federal Procurement Data System from 2010 to 2017 This longitudinal study allows better analyses in terms of causal inferences by allowing for more efficient estimation with increased variability, temporal priority of explanatory variables over outcomes, and reduced omitted variable bias, controlling for timeinvariant factors (Baltagi, 2005; Kennedy, 2008; Wooldridge, 2010; Lee et al, 2018). The paper concludes by offering suggestions for future research in the area

Literature Review

Method

Results and Discussion

Conclusion