Foreign direct investment (FDI) has grown markedly in the world economy since the 1970s. However, the underlying growth has occasionally been punctuated by relatively short-lived reductions in FDI, giving rise to the ‘wave-like’ pattern exhibited in Figure 1. In explaining the motives driving FDI, economists have typically distinguished between horizontal and vertical strategies, the former being driven by market-seeking considerations, and the latter by the desire to access inputs such as cheap labour. Understanding the different motives of FDI is important for addressing policy issues, identifying how FDI and international trade are potentially related, as well as assessing the spillovers that might be associated with FDI. Yet, attempts to uncover the distribution of the different FDI strategies have long been hindered by access to suitable data.
Figure 1. Wave-like pattern of FDI
Identifying alternative strategies from firm-level data
In recent years, there has been some progress in directly observing alternative forms of FDI from firm-specific data of multinational enterprises and, in particular, in classifying the links between the parent and affiliate firms as horizontal (where the parent and affiliate are based in the same industry) or vertical (where the parent and affiliate operate in different, but according to the input-output structure vertically-linked, industries). In particular, Alfaro and Charlton (2009) made several important observations to understand the relative importance of different FDI strategies:
- Horizontal FDI appears to be less significant than commonly assumed by international economists;
- Vertical FDI is more common and accounts for a significant proportion of FDI even between developed countries; and
- Much of vertical FDI occurs within broad (2-digit SIC) industry categories.
In a new paper (Herger and McCorriston forthcoming), we follow a similar procedure to Alfaro and Charlton to establishing the industrial links between firms involved in FDI, but in our case we focus on cross-border acquisitions. Cross-border acquisitions are an important dimension of FDI and, in any one year, can account for the majority of FDI. We use data on all cross-border acquisitions that have occurred in the world economy from 1990 to 2011. Moreover, since we can identify the highly disaggregated activities in which the acquirer and target firms are involved in, and cross-matching these firms with an input-output structure based on the industries they operate in, we are able to categorise alternative forms of cross-border acquisitions in the world economy.
New insights: The importance of conglomerate cross-border acquisitions
In part, our results confirm the main findings of Alfaro and Charlton (2009) – vertical FDI is more common than previously thought (even between developed countries), and horizontal FDI is relatively less important. However, we also establish important new insights. Above all, conglomerate strategies account for a large proportion of cross-border acquisitions. This form of FDI is where the motivation for the acquisition is not related to either a horizontal or vertical motive. The characteristic feature of these acquisitions is diversification across industrially unrelated activities. As Table 1 shows, depending on the benchmark defining vertical relations between the acquiring and target firm, up to 44% of all cross-border acquisitions around the world have been conglomerate. Even allowing for a generous benchmark, which increases the proportion of vertical deals, at a minimum our findings suggest that conglomerate strategies account for a substantial part of cross-border acquisitions.1
Table 1. Proportion of cross-border acquisitions strategies across different benchmarks defining vertical integration
To better understand this new insight, Figure 2 shows the industrial composition of conglomerate acquisitions. While diversification can in principle occur in any industry, it is noteworthy that a large part of conglomerate activity involves the financial sector (and in particular holding and investment firms).
Figure 2. Industrial composition of conglomerate cross-border acquisitions (58,816 deals)
Source: SDC Platinum.
Note: 5% cutoff to define vertical relatedness.
A second new insight of our paper relates to how the different strategies for cross-border acquisitions have changed over time. In contrast to Alfaro and Charlton (2009), where the data was based on a single year, we can track the changes in cross-border acquisitions since 1990 and gauge how the distribution of the different FDI strategies has evolved. While all forms of cross-border acquisitions exhibit the typical wave-like pattern, the most volatile growth has occurred with conglomerate acquisitions (Figure 3). This is particularly notable around the two financial boom-and-bust-periods, with the decline from the peaks of the number of cross-border acquisitions being primarily associated with conglomerate acquisitions.
Figure 3. Cross-border acquisitions over time and their composition, 1990-2011
Financial markets and conglomerate cross-border acquisitions
Concurring with these observations, our results indicate that conglomerate cross-border acquisitions are indeed heavily driven by financial factors. Specifically, we draw on the insights of Baker et al. (2008) who – based on evidence from the US – have shown that parts of FDI are driven by misvaluation in financial markets. We find this effect also in our data. But more directly, the effect of misvaluation relates primarily to conglomerate acquisitions, whilst it does not arise with either horizontal or vertical cross-border acquisitions. This insight is robust to a wide range of specifications relating, for example, to changing the benchmarks to classify cross-border acquisition deals.
Why do conglomerate cross-border acquisitions matter?
A vast body of research on FDI has focused on alternative strategies, how the corresponding motives would underpin the distribution of FDI flows, and the potential benefits that would be associated with FDI. While it may not be surprising to casual observers that multinational enterprises often relocate parts of their production process by acquiring firms abroad, what comes out from our analysis is that conglomerate FDI without an apparent industrial motive is also important. This, in turn, poses a number of new questions. Beyond being driven by financial misvaluation across countries, could there be other factors driving conglomerate FDI (e.g. risk diversification, managerial synergies and so on)? And what are the potential benefits to the host country (if any)? Finally, are the benefits of conglomerate FDI comparable to those of horizontal and vertical FDI or do they differ due to the specific role of relatively short-lived financial factors?
Alfaro, L and A Charlton (2009), “Intra-industry foreign direct investment” American Economic Review 99: 2096-2119.
Baker, M, C F Foley and J Wurgler (2009), “Multinationals as arbitrageurs”, Review of Financial Studies 22: 337-369.
Herger, N and S McCorriston (forthcoming), “Horizontal, vertical and conglomerate FDI: evidence from cross-border acquisitions”, IMF Economic Review.
1 The residual category relates to the case where we cannot specifically allocate the firm level deal into one of the other categories.