Lies, damned lies and statistics: How to make sense of FDI data

Foreign direct investment is a ubiquitous term used worldwide but what does it actually mean? The standard answer regurgitates the official OECD/IMF definition, which is when the direct investor owns at least 10% of the voting power of an enterprise in another country with the crossborder FDI financial ‘flow’ calculated based on equity investment plus reinvested earnings plus inter-company loans between parent firms and foreign affiliates.

The official OECD/IMF definition of FDI is on the one hand very broad, covering all types of direct investment when there is more than 10% voting power, generally achieved by owning more than 10% of the equity. The US is the only country in the world that publishes a breakdown of FDI flows by M&A versus greenfield FDI and new investment versus expansions; for all other countries the official FDI flows data does not break down the type of direct investments taking place. For example, if FDI flows are going up or down in a specific country, we do not know if this is due to swings in M&A activity or fundamental changes in location competitiveness and market demand impacting greenfield investment decisions. Looking at the headline FDI flows and stocks of a country may say very little about its attractiveness for the types of FDI that have a direct impact on economic development. 

On the other hand, the OECD/IMF official definition of FDI can be misleading – why are the Cayman Islands, Luxembourg, Hong Kong and Netherlands among the world’s leading sources of global FDI? Why did the US have huge negative FDI outflows in 2018 and Ireland have negative inward FDI in 2018 while FDI flows to the UK grew strongly despite Brexit? We have seen many governments base FDI strategies on an analysis of global FDI flows without understanding the methodology behind the data and the caveats needed. Because Luxembourg is a top 10 global source of FDI while the US has negative FDI flows does not mean that IPAs should target Luxembourg for FDI instead of the US. This example is very obvious but it becomes less obvious when we look at countries such as the Netherlands (a hotspot for special purpose enterprises – SPEs) or Hong Kong (where due to ‘round tripping’ from China a large chunk of FDI outflows is not really FDI). How important are they really as a source of FDI? 

The FDI flows data also misses out a significant amount of capital investment being made by foreign companies as it does not include capital raised locally – which in some cases can be up to 60% of total project investment value. Official FDI flows data underestimates the contribution of foreign investors to domestic capital formation. For countries seeking to attract FDI to increase foreign exchange earnings, capital being raised locally is of particular concern as it reduces the potential impact of FDI on foreign exchange.

FDI flows data is showing the financial flow. While this is very important as a barometer of capital investment and foreign exchange, for many countries, regions and cities it is less important than job creation and the other non-financial benefits that FDI can bring. 

The IPA perspective

Alternative data sources are therefore needed to understand FDI in more detail. 

IPAs acutely recognise this and in fact most national IPAs in developed economies and most sub-national IPAs across the world do not have targets for attracting FDI flows; their targets are primarily around the number of greenfield FDI projects (companies) they attract and around the job creation and increasingly quality of these investments. These metrics are very different conceptually and methodologically to the official definition of FDI. Most IPAs are focused on greenfield FDI and mostly on majority-owned investments, which is a far narrower definition of FDI than the OECD/IMF 10% rule. Increasingly, IPAs are also dabbling in supporting M&A and ‘new forms of investment’ (NFI), such as strategic alliances and partnerships, but this is mostly done on a case-by-case basis; very few IPAs have targets for attracting and facilitating M&A and NFI. Indeed, Dubai FDI is the only IPA in the world to systematically track and publish data on greenfield, M&A and NFI (see www.dubaifdimonitor.com).

IPAs also understand that FDI is a relationship not a transaction; it can take several years of close contact and support of a foreign investor from when they first express an interest in their location to when they make their full location assessment and announce they are going to invest. It can then often take a year or more for the project to actually be implemented due to the multitude of transactions needed from getting the relevant permits and licences, to finding and leasing, acquiring or building sites and property, to signing agreements with utility providers and suppliers, to hiring a workforce. The actual inward FDI flow emanating from an investment project may only take place years after the IPA has secured the investment project. The inward FDI flow may also be staggered based on the phasing of the investment, and capital can be raised locally so that the full capital investment impact of the investment may not be captured in the FDI flows data. From the point of view of the IPA it is a huge success when a company announces its decision to invest in their location and then a big party when the investor actually opens their investment. IPAs therefore typically record an investment success at the point a company announces its investment and they record the capital investment (and job creation) as the total planned by the investor, which may not be the same as what is eventually registered as an FDI inflow.

Plugging the gap in knowledge

The flagship annual FDI report of fDi Intelligence (The fDi Report 2019) is based on the fDi Markets database, which was launched in 2003 and has more than 16 years of time series data.

fDi Markets was developed to help overcome some of the limitations of official FDI data. It focuses on greenfield FDI only, which is the type of FDI universally accepted as having the most direct benefits for economic development. fDi Markets has tracked 240,000 greenfield FDI projects creating an estimated $13trn in capital investment and generating more than 37 million direct jobs.

The methodology of fDi Markets follows a very similar approach to that of IPAs – tracking mainly wholly owned foreign subsidiaries, only investments that create jobs and capital investment, tracking the total amount of capital investment and jobs planned by the project, and tracking investments at both the announcement and opening phases. fDi Markets also addresses a key weakness of much official FDI flows data by recording the real ultimate parent company for the investment (not an SPE or round tripping) as well as providing a far deeper sector analysis and sub-national analysis, where FDI flows data is not published for many countries. 

When reviewing The fDi Report, the reader should be fully aware of the different definitions and methodologies for measuring FDI as used by fDi Markets versus official FDI flows data. Certainly, it is very interesting to compare and contrast official FDI flows data with fDi Markets data but the caveats and methodological differences mentioned above should be taken into consideration. 

When reviewing the data and analysis in The fDi Report the reader should consider the datasets more closely aligned to the FDI accounting methods used by IPAs rather than the OECD/IMF official method.

The reader should also be aware of intrinsic limitations of any data tracking service; it is not possible to track every FDI project – only data that is publicly available and verifiable as well as data which IPAs are willing to share can be compiled and published. As companies often do not release the amount of capital investment or jobs they are creating, fDi Markets has built-in algorithms that estimate project size where there are gaps. As every FDI project is unique the estimates are exactly that – an estimate of what the project size is likely to be based on other projects in the same sector and country. 

While fDi Markets provides a unique time series dataset and is invaluable for analysing greenfield FDI trends, as with official FDI data it has its own intrinsic limitations in data capture and should be seen as the industry-leading barometer of what is happening in greenfield FDI rather than as a definitive measure of the size of the FDI market.

Dr Henry Loewendahl is CEO of consultancy Wavteq and senior vice president representing fDi Intelligence.

To download a copy of The fDi Report 2019, click here.