Racing to the Bottom? Chinese Foreign Direct Investment and Collective Labor Rights

AbstractStudies have discussed how and if foreign direct investment (FDI) may improve collective labor rights in host countries. These studies presume FDI is from developed democracies to developing countries. Far less discussed is whether this positive effect of FDI on labor rights holds when FDI originates from a developing economy with oppressive labor conditions. The rapid rise of China as a new source of FDI provides an opportunity to examine this question. Using a time-series cross-sectional data set covering 123 countries from 2003 to 2017, I test how Chinese FDI is associated with collective labor rights of host countries in both developing and developed regions and whether the effect of Chinese FDI differs from that of FDI from other countries. I find that FDI from China is negatively associated with collective labor rights in host countries, while non-Chinese FDI does not show such a negative association. The negative association between Chinese FDI and collective labor rights is more pronounced in host countries in developing regions than in developed regions.Existen diversos estudios que han analizado cómo y si la inversión extranjera directa (IED) puede contribuir a mejorar los derechos laborales colectivos en los países receptores. Estos estudios parten de la base de que la IED procede de democracias desarrolladas y de que se destina a países en desarrollo. Sin embargo, se ha analizado en mucha menor medida si este efecto positivo de la IED sobre los derechos laborales se mantiene cuando la IED procede de una economía en desarrollo con condiciones laborales opresivas. El rápido ascenso de China como nueva fuente de IED constituye una oportunidad para estudiar esta cuestión. Comprobamos, mediante el uso de un conjunto de datos transversales de series cronológicas que abarca 123 países de 2003 a 2017, cómo influye la IED china en los derechos laborales colectivos de los países receptores, tanto en regiones en desarrollo como en regiones desarrolladas, y si el efecto de la IED china difiere del de la IED de otros países. Los resultados muestran que la IED procedente de China se asocia de forma negativa con los derechos laborales colectivos en los países receptores, mientras que la IED no china no tiene un efecto tan negativo. El impacto negativo de la IED china sobre los derechos laborales colectivos es más pronunciado en los países receptores de las regiones en desarrollo que en países receptores de regiones desarrolladas.Des études ont porté sur comment et si les investissements directs à l’étranger (IDE) peuvent améliorer le droit du travail dans les pays hôtes. Ces études supposent que les IDE proviennent des démocraties développées et sont dirigés vers des pays en voie de développement. On s’est beaucoup moins intéressé au maintien de cet effet positif des IDE sur le droit du travail quand ils proviennent d’un pays en développement aux conditions de travail oppressives. La montée rapide de la Chine en tant que nouvelle source d’IDE fournit une opportunité d’analyse de cette question. À l’aide d’ensembles de données chronologiques transversaux couvrant 123 pays de 2003 à 2017, j’évalue l’influence des IDE chinois sur le droit du travail des pays hôtes, tant dans les régions développées qu’en développement, et si celle-ci diffère de l’influence des IDE d’autres pays. J’observe que les IDE de la Chine entraînent des effets négatifs sur le droit du travail des pays hôtes, quand les IDE d’autres pays n’ont pas de tels effets négatifs. Les conséquences négatives des IDE chinois sur le droit du travail sont plus marquées chez les pays hôtes de régions en développement que de régions développées.Keywords: Chinacollective labor rightsFDIlabor standardsMNCs AcknowledgmentThe author is grateful to four anonymous reviewers and the editors of International Interactions for their helpful comments and suggestions. The author also thanks Fangjin Ye for sharing data. The author received valuable feedback from Chelsea Chou, Wen-Chin Wu, and other participants at the workshop at the Institute of Political Science, Academia Sinica (March 2022), Joint Area Centers Symposium Series at the Center for East Asian & Pacific Studies (October 2021), and the Faculty-Student workshop at the School of Labor and Employment Relations (April 2023) at the University of Illinois at Urbana-Champaign The data that support the findings of this study are available in the International Interactions Dataverse.Notes1 See Lo (Citation2017) and Myint and Leee (Citation2017).2 Developed countries are identified as countries classified as “high-income countries” by the World Bank.3 GDP per capita is excluded from the model due to the multicollinearity. Adding GDP per capita as control does not change the main findings of the analysis.4 It is a dummy variable coded 1 if the host country i ratified a BIT with China in year t and remains 1 in the subsequent year unless it is terminated. The data for BIT with China is retrieved from the International Investment Agreement database from the UNCTAD.5 It is coded 1 if the host country i signed onto the BRI with China (or BRI memorandum) in year t and remains 1 in the subsequent year. The data for BRI is retrieved from the Council on Foreign Relations. See https://www.cfr.org/blog/countries-chinas-belt-and-road-initiative-whos-and-whos-out (Last Accessed October 4, 2022).6 State visit is a dummy variable coded 1 if the Chinese president or premier visited country i in year t. The data for state visits are retrieved from Wang and Stone (Citation2023).7 Note that the years 2013 and 2014 are dropped due to the lack of collective labor right scores from either of the sources I used.8 Taiwan dummy is excluded due to collinearity. In the first stage regression, the instrument variable strongly predicts Chinese OFDI stock and the F-statistics of the excluded instrument is above 10. See Supplementary Table A2 for first-stage regression results.9 A country is identified as a developing country if it is classified as a non-high-income country for more than 12 years.10 See Supplementary Table A2 for first-stage regression results. The F-statistics of excluded instruments in first-stage regression is 12.08, which is above 10, and is above the critical value of 8.96 for a maximum bias in the instrument relative to OLS of less than 15% (Stock and Yogo Citation2005).11 The data for bilateral FDI stock are obtained from the UNCTAD. I identify developed countries as countries classified by the World Bank as high-income countries in 2003 and the rest as developing countries. I used data compiled by Cheibub, Gandhi, and Vreeland (Citation2010) to identify democracies.