Skip to main content

Advertisement

Log in

Services Input Intensity and US Manufacturing Employment Responses to the China Shock

  • Published:
Review of Industrial Organization Aims and scope Submit manuscript

Abstract

We present industry-level evidence that manufacturing sectors that make use of services as inputs more intensively are more robust to shocks from import competition. Specifically, the negative effect of the China shock on US manufacturing employment is lower for industries with high services input intensity. Furthermore, our analysis reveals significant heterogeneity in the impact of different types of services, which points towards a fruitful research agenda on the role of services as a determinant of firm competitiveness.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Similar content being viewed by others

Notes

  1. Data are from US Bureau of the Census, at https://www.census.gov/foreign-trade/statistics/historical/index.html and reflect balance-of-payments figures. Thus they do not include services that are sold by foreign affiliates of US multinationals, which are an important additional channel for international provision of services by US-owned companies (Francois and Hoekman 2010).

  2. Berlingieri (2015) finds that firms increase their services input intensity in order to manage coordination complexity (proxied by the number of contested export destination markets).

  3. On the one hand, offshoring reduces input prices and increases profits, which in turn potentially increases manufacturing production and labor demand. On the other hand, higher quality and cheaper service inputs may substitute for labor that is used in production, decreasing labor demand (Amiti and Wei 2006; Milberg and Winkler 2010b; Winkler 2010). Consistent with the theoretical ambiguity, empirical findings are mixed (Amiti and Wei 2005; 2006; Schöller 2007; Winkler 2010; Michel and Rycx 2012; Milberg and Winkler 2010a, 2015; Eppinger 2019). However, services offshoring tends to be associated with a higher demand for skilled labor at the firm level (Crinò 2010; Andersson et al. 2016).

  4. Technical coefficients in Fig. 1 capture the technical relationship between US industries that prevailed in the early 1990s. The coefficients comprise the elements of the square matrix A defined as \(A\equiv YM\), where Y is a dimension n square matrix of zeros, except along the main diagonal, that includes the inverse output of each industry and M is the intermediate demand matrix. For each services-manufacturing sector pair (sj), the technical coefficient is the element \(a_{sj}\) of A and represents the cost of the intermediate inputs from services sector s per dollar of total production of manufacturing sector j.

  5. We define local labor markets according to Autor et al. (2013) to be 722 non-overlapping commuting zones which represent areas with a high degree of labor mobility within and very little mobility across zones. Our empirical specification follows closely Section 6 of Acemoglu et al. (2016). We refer the reader to these papers for an in-depth discussion of identification and instrumentation strategies.

  6. The business services category includes professional services such as legal, accounting, management consulting, and engineering.

  7. The nine census divisions are identified by eight dummy variables that group together subsets of CZs.

  8. Our controls are not shown; they are the standard ones in the literature: the lags of local manufacturing employment share; college-educated share of population; foreign-born share of population; female employment share; share of routine occupations; and share of outsourcable occupations.

  9. Controlling for endogeneity due to potential unobserved demand shocks with the use of the instrumentation strategy is important as the OLS point estimates (not reported) of \(\beta\) and \(\mu\) are \(-0.0378\)*** and 0.379*, respectively, which significantly underestimate this impact.

References

Download references

Acknowledgements

The authors are grateful to Lawrence White, Andrea Ariu, Paola Conconi, Caroline Freund, Lorenzo Caliendo, Jeff Frieden, Beata Javorcik, Doug Nelson, Marcelo Olarreaga, Marco Sanfilippo, Asha Sundaram, Thierry Verdier and members of the Galbino Project for helpful comments and suggestions on earlier versions of this paper.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Matteo Fiorini.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix

Appendix

Table 5 Mapping between sectors 2-digit ISIC Rev. 3 codes and labels

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Bamieh, O., Fiorini, M., Hoekman, B. et al. Services Input Intensity and US Manufacturing Employment Responses to the China Shock. Rev Ind Organ 57, 333–349 (2020). https://doi.org/10.1007/s11151-020-09770-2

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11151-020-09770-2

Keywords

JEL Classification

Navigation