Offshoring, Welfare, and Flexibility in the Context of US Protectionism

Abstract

We reexamine the effects of offshoring in the US economy in the aftermath of the 2008 crisis. We use a matching model with endogenous adjustment of educational skills while distinguishing between offshoring of high and low-skill activities. We first show that offshoring leads to a restructuring of the economy through a change in the wage premium where overall welfare is improved. Moreover, in a policy exercise, we show that, if offshoring were to be opposed by a protectionist agenda, the resulting welfare losses could be counterbalanced by increased labor flexibility.

Keywords:

• Offshoring
• skills
• welfare
• wage premium
• flexibility

JEL classifications:

• F66
• J64
• F16
• F17

Acknowledgements

Thanks are extended to an anonymous referee and the editor of the journal for helpful comments and suggestions. We are also thankful to the participants of the XX Conference on International Economics, Granada, Spain, for their helpful feedback.

Disclosure Statement

No potential conflict of interest was reported by the authors.

Notes

1 See, for example, Berman et al. (1994, 1998), Acemoglu (2002), Card and DiNardo (2002), or Chusseau et al. (2008), who frame their analysis within the broader subject of skill-biased technological change.

2 Feenstra and Hanson (1996a) also work with the nonproduction workers share of the industry wage bill (as do Berman et al., 1994), which they calculate from the NBER database. Their last value for year 1990 stands at around 43 % whereas our first value for 1995 is close to 44.6 %. They also proxy offshoring by an intermediate imports index, although their analysis ends in the 1990s – the rationale is that the higher the volume of intermediate trade the higher the offshoring intensity, as inputs previously produced in the home country are now being imported back from low-wage countries.

3 See the errata to their 1996 paper, which revises the original estimated contributions downwards (from 31%).

4 The change in this variable over the 1979–1990 period was in the order of 39%.

5 A description of how the model works for one type of worker can be found in Rogerson et al. (2005) and Williamson (2010).

6 Albrecht and Vroman (2002), Agnese and Hromcová (2016), and Hromcová and Agnese (2019), analyze the cases where high-skill workers accept low-skill jobs. In these papers, two types of equilibria are considered: the equilibrium with cross-skill matching (CSM) and the equilibrium with ex post segmentation (EPS). CSM is reached when high-skill workers and low-skill vacancies are matched, whereas EPS kicks in when these potential matches do not meet. Both equilibria, CSM and EPS, can potentially coexist. Due to the characteristics of our study, we focus on the EPS equilibrium only, and the description of the economy is in line with this premise.

7 Appendix 2 shows the definitions of the variables along with their sources, and Appendix 3 goes over the classification of manufacturing industries based on their R&D intensities – this classification is then employed to distinguish between high and low-skill offshoring measures.

8 This stems from assuming a translog cost function, as suggested earlier by Berman et al. (1994).

9 We have worked with the deflators of value added and input prices (and their lags) as instruments. Further notice that we entertain a lag of the dependent variable as to introduce some dynamics, and also that the value added variable enters the specifications in differences, while the offshoring regressors only enter with one period lag.

10 As discussed in the introduction, the resulting change in the nonproduction wage share during the eighties was found to be within the 19–21% interval (see Feenstra & Hanson, 1996a, and errata, 1996b). Feenstra and Hanson (1999) find a slightly lower effect as they work with other variables in their specifications – offshoring explains a 15% increase in the relative wage of nonproduction workers while expenditures in high-technology capital such as computers, for example, can explain as much as 35%.

11 See, among others, Ebenstein et al. (2014), Görg and Görlich (2015), Hummels et al. (2013), and the earlier contributions by Feenstra and Hanson (1996a, 1999).

Additional information

Funding

Financial support from the Spanish Ministry of Education and Science [grant number RTI2018-095799-B-I00] is gratefully acknowledged.

Notes on contributors

Pablo Agnese

Pablo Agnese, UIC Barcelona, Lecturer, holds a PhD in economics (2010) from the Autonomous University of Barcelona, Spain. His research interests include offshoring and welfare, trade, and cryptonomics and the Blockchain. He currently holds a Lecturer position with tenure at UIC Barcelona, Spain, and prior to that he held positions at FH Düsseldorf, Germany, and IESE Business School and Pompeu Fabra University, Spain.

Jana Hromcová

Jana Hromcová, NEOMA Business School, Adjunct Professor, received her PhD in Economics from Universitat Autònoma de Barcelona in 1999. Her research interests are in macroeconomics, defense, and labor economics. She has been lecturing at NEOMA Business School, France, since 2019. Previously she held positions at Universitat Autònoma de Barcelona, Universitat de Girona, Universidad Carlos III de Madrid, and Universidad de Alicante, Spain.