Foreign Direct Investment (FDI) consists of controlling investments by companies and individuals that cross international borders. Essentially, it is the investments that multinational corporations use to own the different pieces of themselves in different countries. FDI is often considered to be the backbone of the global economy, as it is created whenever a company makes a “bricks-and-mortar” cross-border investment in a factory or other facility, or acquires or mergers with a company in another country. However, FDI also consists largely of “paper” shell companies used for tax avoidance and other purposes. In fact, as of 2016, roughly two thirds of all worldwide FDI recorded by the IMF was either in or from “offshore” jurisdictions whose investments are believed to consist largely or mostly of foreign- controlled shell company structures. Taking into account the double-counting of offshore investment relationships, at least one third of the underlying capital invested in the global FDI network passes through various types of offshore structures. The implications of this are profound, as its underscores that we cannot understand the sorts of arrangements revealed by, for example, the Panama or Paradise Papers, in terms of some set of “bad actors” whose activities distort the structure of the global economy. Rather, as Bill Maurer put it, “offshore in many ways is the global economy.”
Importantly, the cast of leading jurisdictions in these structures tends to be somewhat different from the stereotype of an “offshore” jurisdiction. Rather than small tropical islands, the largest hubs in the global offshore shell company network are countries in Europe such as the Netherlands, Luxembourg, Switzerland, or Belgium. Roughly two thirds of all offshore FDI is routed worldwide is routed through the first two of these countries. Moreover, even though “small islands” account for much of the remaining activity, the most important of these are British Overseas Territories and Crown Dependencies such as the Cayman Islands, British Virgin Islands, Bermuda, or Jersey. To a substantial extent, these are located with rather than outside of the UK jurisdiction. Meanwhile, at the regional level, specialized hubs help to plug nearby countries into the global shell company network. Hong Kong and Cyprus, for example, act as the leading offshore “gateways” to China and Eastern Europe respectively, while Mauritius and Singapore play a dominant role in South and Southeast Asia. Importantly, much of the “offshore FDI” passing through such jurisdiction is in fact not foreign at all, but rather consists of “round-tripping” whereby the companies in a country shift the ownership of their domestic operations—on paper—to offshore holding companies. When American corporations use so- called corporate “inversions” to move their headquarters to low-tax Ireland, for example, this shows up as an increase in Irish FDI in the United States. In other cases, the investment that enters a country from an offshore jurisdiction does in fact represent genuinely foreign money; however, this investment almost never originates from the offshore jurisdiction itself, but has rather usually been routed through shell companies established there by companies in some third country.
Surprisingly, while official data from sources such as the IMF and OECD records how much money enters onshore economies from particular offshore jurisdictions, we have very little information on where this offshore money actual comes from before it passes through any such jurisdiction. Most developing countries, and even some developed countries—including the UK—do not record any information on the ultimate as opposed to immediate origin of the investment that enters them. What this means is that their official data on inward foreign investment is to a large extent recording a geography shaped by offshore paper constructs rather than a meaningful picture of how these countries are inserted into the global economy. In the case of the UK, our analysis suggests that the actual scale of investment from some countries in Asia and the Middle East may be more than an order of magnitude larger than the officially reported value, once offshore structures are included (see technical paper). A few developed countries, such as the US, France, Germany, and Italy, do make some attempt to record the ultimate origin of investment. However, even these do not provide any breakdown of the origin of funds that enter via a particular offshore conduit jurisdiction. Moreover, they do not account for the issue of corporate inversions, wherein the parent company itself is moved, on paper, to an offshore jurisdiction. What all of this means, is that the apparent shape of the world economy that we think we see in official data is largely a mirage created by tax avoidance and other offshore strategies. In effect, we know surprisingly little about what the world economy actually looks like.
In theory, there is in fact quite a bit of information available at both the macro (country) and micro (firm) level that can help shed light on who this offshore money actually belongs to, and where it is actually going. However, the fact that it is fragmented between many different datasets, each revealing a different piece of the puzzle, has made it difficult to take advantage of this information, with most existing studies seeking to unravel the world map of FDI only exploiting one dataset at a time.
The Atlas of Offshore FDI seeks to put these pieces of the puzzle together as much as possible, drawing on a novel methodological approach based on integrating information from multiple different macro and micro-level datasets. By “triangulating” between the known data points in these resources, and the discrepancies between them, this methodology allows for a more detailed portrait of the “true” map of world FDI to be constructed than is possible through the use of any one dataset in isolation (see technical paper for details). For the first time, this map allows for the offshore investment positions in nine of the world’s largest developed and developing countries—France, Germany, Italy, the United Kingdom, United States, and “BRICs,” Brazil, Russia, India, and China—to be traced back through the offshore fog to their actual national point of origin (even if only in the form of rough estimated ranges – see discussion below).
Some of the most important elements of the global offshore FDI network are discussed in the accompanying technical paper. However, the new dataset underpinning the Atlas of Offshore FDI is vast, and to allow the complexities of the global shell company network to be explored in in detail, the data is accessible here via an online interactive mapping tool. In the menu at the upper left, one can select one of a number of major FDI destination economies. Meanwhile, with the slider bar and toggle control in the upper right, one can customize the level of detail that is shown on the origin and conduit economies from and through which this investment originates and passes. In the middle menu, one can pick a customized list of origin and conduit economies, to display their connections to the selected destination economy. One can also leave this middle menu blank, to display all origin and conduit economies that appear at a given selected level of detail. Clicking on a country or investment position in the network diagram reveals additional information in a side bar on the right, on the location and full name of a country, and the breakdown of aggregated investment positions according to their specific offshore pathway.
The green network connections in the diagram show the value of conventional FDI, wherein companies that are actually native to one country simply invest directly in another country. Meanwhile, the orange/pink network connections show conduit investments where investment between two countries is intermediated through a third, typically offshore jurisdiction. As can be seen, a very large part of this money usually consists of round-trip investments that originate from the destination country itself, which can be seen looping from the destination country back to itself via conduit jurisdictions. The animated direction of flow in the network links shows the direction of the flow of funds from origin to destination country.
A few key points need to be emphasized about information presented here: firstly, and most importantly, because there is no officially recorded data that breaks down the true origin of funds that pass through offshore conduits, this information has had to be generated through a novel methodology (as outlined in the technical paper) that produces a range of possible estimated values, rather than a precise value, for any given investment position. Given that no information at all has hitherto been provided on these investment values, the data presented here represents a significant step forward. However, depending on the quality of the data underpinning the estimation of any particular investment, the range of uncertainty can often be very large—particularly in the case of smaller investments. As such, it is important to treat the data here as a rough indication of the general structure of the global offshore FDI network, rather than a precise blueprint of this structure. From a statistical standpoint, the upper and lower estimates given in the interactive mapper are defined so that there is a 75% chance of the actual value falling between these estimates, which means that there will be many cases where the actual value is outside of this range (more detailed information on the full range and statistical breakdown of estimates can be found in the technical paper). Ultimately, it is up to the major national and international statistical agencies to compile the data that would allow for these figures to be precisely known, rather than only roughly estimated as done here.
Secondly, the diagrams in the interactive mapper do not explicitly distinguish between “offshore” and “onshore” jurisdictions; the key distinction is rather between investment that passes directly from one country to another (green), and investment that passes through any third country conduit (orange/purple). Additional information on the “offshore” characteristics of particular conduit jurisdictions (e.g. financial secrecy) can be found in the inset map is pulled up when you click on a country, as well as in the accompanying technical and highlights papers. In many cases, the determination of “offshore” or “tax haven” status is a complex issue, which requires extended discussion.
Finally, the “conduit” investment figures given here include corporate inversions, wherein the parent company itself is moved on paper from its actual home country to an offshore jurisdiction. A round-trip investment position shown in the map that passes from the US to back to itself via Ireland, for example, would include both cases of US headquartered companies owning their US operations via Ireland, and cases where US companies have inverted to Ireland, meaning that their US operations are now nominally controlled by an “Irish” parent company. For the diagrams showing investments in the US, UK, France, Germany, and Italy (but not Brazil, Russia, India or China), you will sometimes see investments passing through two layers of conduits. Where this occurs, it shows that companies have performed an inversion to move their headquarters to the first conduit jurisdiction, but are investing in other countries via the second conduit jurisdiction rather than directly form the jurisdiction where they have inverted. For example, American companies that have used inversions to move their headquarters to Ireland, appear to mostly own their US operations via conduit structures in Luxembourg, rather than directly from the parent company in Ireland. In the interactive maps, this example would be shown as an investment originating from the US, and then passing first through Ireland, then through Luxembourg, before finally ending in the US as a destination. Note that for Brazil, Russia, India and China, due to lower data quality, conduit investment positions are only categorized by the immediate point of entry. If a Mainland Chinese company, for example, inverts to the Cayman Islands, but owns its Mainland operations via a conduit in Hong Kong, the map will show the investment looping directly from the Mainland back to itself via Hong Kong, omitting the Cayman Islands parent company.