India has become increasingly important both as a source and as a host of foreign direct investment (FDI) over the last decade. But trying to make sense of the numbers regarding FDI inflows and outflows can be quite challenging.
The data on bilateral FDI outflows is sketchy; finance ministry reports the value of aggregate FDI outflows from India and the value of approvals of FDI outflows at a bilateral level, but a consistent time series of the actual value of outflows with a country-wise breakdown isn?t available in public domain. While data on actual FDI inflows is reported by the Department of Industrial Policy and Promotion at a disaggregated country level, there are serious concerns about the usefulness of bilateral FDI inflows data available in the public domain.
For example, the data on FDI inflows into India almost always shows Mauritius to be the largest source of foreign investment flows into the country. But Mauritius is widely regarded as an offshore financial centre (OFC) used by foreign investors as an intermediary to reach India, predominantly to capitalise on tax rebates. Conversely, as Indian companies become globalised, many have chosen to either use their overseas locally-incorporated subsidiaries to invest overseas or have established holding companies and/or special purpose vehicles in OFCs or other regional financial centres like Singapore or Netherlands to raise funds and invest in third countries.
Apart from such transhipping, some of the inflows from Mauritius in particular, but also from other OFCs, could also be round-tripping back to India to escape capital gains tax or for other reasons, not unlike the investments dynamics between China and Hong Kong.
Bilateral FDI data?which only captures the actual flow of funds rather than ultimate ownership?may present a rather distorted picture of the extent of linkages between India and the rest of the world. Usefulness of such data for research and policy purposes needs to be examined as it gives a misleading picture. There are serious lacunae in the way direct cross-border investment flows are reported, hence the need for alternative datasets.
In order to understand de facto linkages between India and the world, one needs to examine the data on actual ownership of the foreign investment flows coming into the country. While data on individual firms that have invested in India may be available via firm-level surveys, for a more complete picture of FDI inflows into the entire economy one needs to examine an aggregation of all such firms investing in India from different parts of the world. This would be a prohibitively costly exercise. A more feasible alternative would be to examine the data on mergers and acquisitions (M&As) made by global firms in India and Indian firms globally. Such data, tracking actual ownership of purchases and sales, is maintained by many commercial entities, unlike the data on FDI flows that is compiled by national sources.
Two basic conclusions arise. One, many acquisitions by the US and the UK have been channelled via Mauritius. Two, Indian companies have largely been using Singapore, Netherlands and OFCs as intermediaries to purchase assets overseas, primarily in the developed world and resource-rich countries. While OFCs are used as tax havens, both Singapore and Netherlands are attractive hosts for holding companies from India and elsewhere for (a) their low and simple tax rates, (b) the large number of double tax treaties between the two countries and the world, (c) use of English, (d) human capital and (e) excellent logistics plus air and sea connections. All this explains their attraction to Indian businesses eager to internationalise their operations.
Figures 2 and 3 capture data on FDI inflows (reported by the Indian government) and M&A purchases (reported by commercial entities) that have taken place in India (by source of origin) in 2000-07. A comparison reveals the above discussed inconsistencies.
It is interesting to see that most of the OFCs like Mauritius (mainly) but also Cyprus, Cayman Island and Bermuda, which comprise nearly 50% share of the total FDI inflows (as reported by government sources) do not even figure in the data on inbound M&As to India.
Focusing on the FDI data, only 18% of inflows to India have been by the US and the UK combined, while about 15% are by the non-UK European countries (mainly Netherlands, France and Germany) and about 10% by East Asia (mainly Singapore and Japan). In contrast, the M&A data on foreign acquisitions in India tell quite a different story. The US is the single largest acquirer of Indian companies (35%), followed by the UK (16%) and the rest of Europe, including Netherlands (27%) and East Asia (18%) (distributed between Japan, Singapore, Malaysia and Hong Kong). So almost all of the inbound acquisitions to India have been by the US, Europe and Asia. This appears to offer a far more informative geographical breakdown of sources of direct investment equity flows to India compared to the FDI data.
As noted, similar bilateral data on India?s actual FDI outflows are not publicly available on a systematic time series basis. While approvals may not provide a fully realistic picture as not all approvals are realised, available data, at least for aggregate actual outflows, suggest that there is a reasonable degree of correlation between approved and the actual outward FDI flows from India.
Accordingly, the outward FDI approvals data ought to offer some useful insight when compared to data on India?s M&A purchases overseas. It is well known that Indian businesses have been very active in overseas investments in the last few years, particularly since 2006. Anecdotal evidence and examples point to the fact that many of these investments have been in developed countries like the US, the UK and rest of Europe. Notable instances would be Tata Steel?s purchase of Corus and Tata Motors? purchase of Jaguar and Land Rover in the UK, and Hindalco?s acquisition of the Canadian aluminium giant Novelis.
Developed countries like the UK and the US have surprisingly small shares of India?s approved outward FDI (6% each) for recent periods for which detailed data are available (2002-08) compared to Singapore (22%), the Netherlands (15%) and Mauritius and other OFCs in total (25%). So, over 50% of India?s approved FDI is destined to the financial centres (regional and offshore).
Examination of M&A purchases for more or less the same period (2000-07), however, reveals quite a different picture. Canada emerges as the top host country for India?s outbound acquisitions with a 34% share, followed by the US with a 24% share. While Indian companies have undertaken a number of varied purchases in the US, acquisitions in Canada have been concentrated in resources, including Novelis mentioned above. Apart from these, around 16% of India?s acquisitions have been aimed at resources rich countries (Russia, Egypt, Australia and South Africa) and the rest to the UK and Europe (17%). The Tata Motors? acquisitions of Jaguar and Land Rover Brands from the UK do not show up in our data as they were done in early 2008. It is likely that an extension of the data to 2008 would see a jump up in the UK as a source of Indian outbound M&As, as would Europe in general.
One clearly has to be cautious in comparing the two sets of data (FDI versus M&A), as the M&A data excludes greenfield investments. While M&As are growing as the preferred mode of foreign entry, M&A data are not from national sources, being sourced from commercial entities, and there are questions about consistency in terms of company coverage, definitions and such. In addition, tracking transactions based on ownership is always tricky, particularly given the increasing complexity of global businesses. All this being said, the important point is that India?s FDI data at a bilateral level may offer quite a misleading indication of the extent of real linkages and should be interpreted with extreme caution, a point that researchers and analysts have failed to adequately appreciate.
Ramkishen S Rajan is an associate professor at George Mason University, Virginia, US. His coauthor Sasidaran Gopalan is a research associate at the Institute of South Asian Studies (ISAS), National University of Singapore
