Offshoring via vertical FDI in a long-run Kaleckian Model

Abstract

This paper develops a two-country Kaleckian model in which “Northern” firms invest a fixed fraction of total investment in foreign affiliates in the low-wage “South” in order to offshore the production of intermediate goods over time. On the back of this setup follows an analysis of the macroeconomic implications of offshoring in the short and long run. Offshoring through vertical FDI is found to lead to a falling wage share and a simultaneously falling price level and rising markup in the North, whereas the effect on equilibrium capacity utilization may be positive or negative. Regardless of the effect on capacity utilization and firm profitability, we can show that the structural change implied by offshoring leads to lower rates of capital accumulation and employment in the North in the short run. The long-run effects on Northern employment and growth, on the other hand, depend crucially on the long-run accumulation rate of the Northern-owned multinational firms and on whether wages in the North and the South endogenously converge. The model appears well suited to shed light on many real-world macroeconomic phenomena, such as rising FDI flows, falling wage shares, rising markups in an era of low inflation, hysteresis, and secular stagnation.

Keywords:

• Distribution
• foreign direct investment
• offshoring
• stagnation

JEL CLASSIFICATIONS:

• F62
• F23
• O41
• E11
• E12

Acknowledgments

My thanks go to Eckhard Hein for his valuable comments on an earlier draft and to the other members of the Growth Regime Working Group of the IPE Berlin for their stimulating questions and comments. Of course, any remaining errors are mine alone.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data used in Figure 1 are derived from the public domain resources described in the text and reference list, and are also available from the author upon request.

Notes

1 Schröder (2020) allows profits to have a positive effect on investment under the assumption that the effect of profits on consumption and investment is smaller than the effect of the wage bill on consumption, i.e. domestic demand is wage-led by assumption. We reach the same qualitative result here by simply employing the neo-Kaleckian investment function.

2 Arguably, however, one could conceive of a firm that decides to engage in offshoring even if unit variable costs do not fall. For example, a firm that wishes to increase its mark-up by reducing labour union power may decide to offshore production even if a fall in unit labour costs does not result.

3 For a more elaborate definition of offshoring and its various distinctions, see OECD (2007).

4 This point is mentioned elsewhere in the literature, such as in Milberg and Winkler (2010) and Auvray and Rabinovich (2019), however, these authors argue that financialization redirects the increased profits from offshoring away from investment and towards shareholder value maximisation. While we take no issue with that explanation and its empirical relevance for many countries, it would be interesting, nonetheless, to understand whether the process of offshoring alone could lead to higher profits and lower domestic investment without invoking financialization.

5 Indeed, as Dunn (2005) argues, the theory of the firm advanced by John Kenneth Galbraith, grounded in uncertainty, power, and planning, helps explain why multinational corporations emerged in the first place.

6 Since productivity is the same abroad as at home but labour is cheaper abroad, one may wonder why Northern firms do not simply offshore all production. In fact, our model does not preclude the possibility. However, it is worth keeping in mind that time and finance constraints as well as fundamental uncertainty and perceptions of risk and affect the degree and pace of offshoring. Endogenous wages, as we will see in the final part, may also present a reason to not offshore all production, as might a number of other factors not considered here, such as productivity differentials, transport costs, and the natural geography of resources.

7 Alternatively, one can allow the foreign affiliate to apply the northern mark-up upon foreign affiliate unit labour costs and arrive at much the same outcome, so long as the mark-up is not applied a second time in the North. The main difference would then be that the model would have to account for net income receipts in the form of repatriated profits. Apart from this, the outcomes to be described in this paper are essentially the same, hence the more convenient notion that the foreign affiliate prices its output (i.e. the intermediate good) at cost price. Lastly, note that, by applying the mark-up only in the second stage of production in the North, the Northern multinationals are essentially engaged in profit shifting. In reality, this would have implications for tax revenues and public policy, but it does not matter for our purposes since our model does not include a government sector.

8 Of course, this is a large simplification stemming from the fact that our focus in this paper is on the effects of offshoring on the source of FDI rather than the recipient.

9 This scenario is not graphed in Figure 6, though it looks similar to Curve B.

10 More specifically, these are the non-zero values of the time parameter for which ${\epsilon }_N\left(\theta \right)={\epsilon }_{N,0}$ for Curves C and D, where the former is graphed using the values $u_0=0.8,$ $u_{\theta }=0,$ $g_0=0.1$ and $g_{\theta }=0.$05 and the latter $u_0=0.8,$ $u_{\theta }=0.05,$ $g_0=0.1$ and $g_{\theta }=0.$1. These values are purely for the sake of illustration, of course.

11 This finding complements, using a long-run model, the argument found in Woodgate (2021b), which is on a short-run basis.

12 Of course, it is very rare to see empirical work on demand regimes control for aspects of offshoring. Presumably, this is partly because the theoretical case for doing so has been hitherto underdeveloped and partly because data to construct offshoring control variable on a long-run basis may be hard to come by. This point is emphasised and elaborated upon in Woodgate (2021a).

Additional information

Notes on contributors

Ryan Woodgate

Ryan Woodgate is currently a research associate and lecturer at the Berlin School of Economics & Law and a PhD candidate at the Université Sorbonne Paris Nord. His work centers on the macroeconomic effects of globalization, intergovernmental policy competition, and the activities of multinational enterprises viewed from a post-Keynesian perspective.