May the index be with you – measuring the globalization of FDI (OeNB Bulletin Q2/25)

Christian Alexander Belabed, Julian Mayrhuber ¹

JEL classification: F02, F21, F36

Keywords: (de)globalization, FDI, geopolitical fragmentation

The discussion of deglobalization and the potential fragmentation of the global economy has gained more and more traction in academic and policy circles alike. ² Its topical relevance for monetary policy and central banks is extensively covered in ECB (2021, 2024). While the jury is still out on whether deglobalization is happening or not ³ , there is no doubt that important indicators such as cross-border investment have lost momentum since the financial crisis and have not recovered through various phases of globalization since then (Abeliansky et al., 2024a).

In the most recent literature, Aiyar et al. (2023) discuss the potential effects of geoeconomic fragmentation on capital, trade, migration, the global financial system, and a way to preserve some of the economic gains of globalization and a multilateral global economic system. In a similar vein, Gopinath et al. (2024) find that geopolitical de-alignment has had a negative effect on cross-border trade and investment since the Russian invasion of Ukraine. By comparing current developments to the Cold War era, they project that fragmentation could worsen significantly if geopolitical tensions persist. Aiyar and Ohnsorge (2024) discuss the risks and benefits of countries becoming “connector countries” – countries which expand their trade and investment links with opposing blocs to shore up gains from serving both ends. According to their analysis, Mexico emerges as a vertical connector. It shores up FDI from China to produce goods for export to the USA. Vietnam, in contrast, has a very diversified set of export partner countries. It serves as a horizontal connector. This suggests that different strategies can lead to very different outcomes regarding FDI relationships and a country’s exposure to geopolitical and geoeconomic shifts. Catalán et al. (2024) present a gravity model to estimate the effects of increased geopolitical distance on the allocation of investment funds’ assets. They find that geopolitically more distant recipient countries receive less funds than geopolitically close funds. Regarding FDI, however, Tan (2024) argues that FDI fragmentation, based on geopolitical de-alignment, is not a widespread phenomenon. According to the author, FDI fragmentation is restricted to a few industries of strategic value that mainly affect US FDI outflows. In addition, Tan argues that this is not a clear indication of a widespread disruption of investment links between non-allied countries or blocs. Abeliansky et al. (2024b) conclude that FDI and portfolio investment (PI) is affected by increasing geopolitical distance between countries. The effect is also more pronounced in recent time periods and for emerging markets.

Since the picture of the (de)globalization of FDI remains a bit blurry, we approach the issue by developing an FDI globalization index. First, let us discuss existing contributions. The most interesting index of globalization is the one proposed by Gygli et al. (2019), the KOF Globalization Index. Gygli et al. (2019) collect a large set of 43 indicators of three overall dimensions of globalization: 1) economic globalization, including trade in goods and services, FDI and PI or trade regulation and investment restrictions among others; 2) social globalization, comprising tourism, migration, patents, the number of international airports or press freedom; and 3) political globalization, including the number of embassies in a country, international treaties or treaty partner diversity. By integrating these components, the KOF Index produces a composite score ranging from 0 to 100, with higher values denoting greater global integration.

While the inclusion of multiple variables enhances the comprehensiveness of the index, it also increases the risk of data discontinuities, particularly for countries with limited reporting capacity. Missing or inconsistent data for any component can undermine the reliability of the index over time, especially for longitudinal studies. Additionally, the weighting scheme used to aggregate these indicators, though methodologically rigorous, may not fully reflect the relative importance of different components across diverse economic contexts.

Furthermore, there are no bilateral indicators, in particular regarding de facto cross-border investment. The only bilateral indicator used in the category financial globalization is the number of bilateral international investment agreements and bilateral treaties with investment provisions, i.e. de jure variables. Finally, the authors use the Herfindahl-Hirschmann Index (HHI), a common measure of concentration, only for trade globalization. It could be used for cross-border investment variables as well. These caveats suggest that while the KOF Index is a powerful tool for analyzing financial globalization, its robustness and applicability depend on the availability and quality of data as well as on its capacity to adapt to the evolving nature of global financial integration. Another example of a globalization index is Vujakovic (2009), who includes much fewer indicators and does not distinguish between de facto and de jure categories. To the best of our knowledge, this index has never been updated again, hence it is impossible to assess ongoing changes. Other, much less prominent examples of globalization indices are discussed in Martens et al. (2015).

Given these issues with existing indices of (financial) globalization, we propose a new index of FDI globalization based on a smaller set of indicators for inward and outward FDI. We build on previous work by Fitter et al. (2024), who developed an index of globalization for migration building on the work by Czaika and de Haas (2015). Our overall index of FDI globalization captures the dynamics of three dimensions of FDI globalization (intensity, spread, distance), out of which two dimensions (spread and distance) are bilateral indicators. Based on these three dimensions, we build overall global indices of in- and outward FDI, which are then aggregated into an overall FDI globalization. ⁴ Using a limited set of dimensions has the comparative advantage of a lower risk of data-related issues such as the discontinuation of time series. To the best of our knowledge, our index is the first to focus exclusively on FDI, which can be considered as one of the most exposed types of cross-border investment in a changing landscape of globalization. Our index will, of course, be compared to popular alternatives to assess its relative usefulness despite a much smaller set of variables.

Our findings suggest that the globalization of FDI has undergone significant fluctuations over the past two decades, with notable declines during major crises such as the 2008 financial crisis or the current ongoing geopolitical fragmentation. These shifts underscore the vulnerability of cross-border investment to systemic shocks. Rising geopolitical tensions and fragmentation have reduced the diversification of FDI partnerships, as measured by the spread dimension. Countries with fewer and a rather concentrated group of FDI partners are increasingly exposed to risks associated with economic dependencies and disruptions in global value chains. The proposed index offers enhanced sensitivity to short-term disruptions compared to traditional measures, providing a more dynamic and usable tool for analyzing FDI globalization and its implications in a rapidly changing economic landscape. Our ultimate objective is to construct a comprehensive, yet straightforward index that captures the essence of (de)globalization by encompassing cross-border investment, trade and migration.

The remainder of the paper is structured as follows: Section 1 recapitulates the evolution of global FDI flows. Section 2 discusses the underlying data and the construction of the index. Section 3 presents empirical findings for each dimension and the overall index of FDI globalization. Section 4 concludes.

1 Phases of (de)globalization and capital flow dynamics

Before we introduce our FDI globalization index, a quick recapitulation of FDI trends in the last almost 30 years seems useful. Chart 1 illustrates global FDI flows, measured as a percentage of world GDP, between 1995 and 2023. The data are divided into four distinct globalization phases, highlighting shifts in FDI dynamics over time (see Abeliansky (2024a) for a detailed description of the four phases). During phase 1 (“goldilocks phase of globalization” from 1995 to 2007), FDI flows, which became a cornerstone of global economic integration grew significantly. This phase culminated in a peak in total capital flows at more than 3% of world GDP before the 2008 global financial crisis. Phase 2 (“hangover” from 2008 to 2013) marks the impact of the financial crisis and is characterized by a significant contraction in FDI flows. The remainder of phase 2 saw a stagnation of FDI flows in percent of GDP, highlighting a phase of reduced appetite for cross-border investment, most likely owing to balance sheet repair pressures and overall subdued global demand. Phase 3 (“return of imperialism” from 2014 to 2018, which included Russia’s annexation of Crimea and the Brexit referendum) saw moderate FDI growth, even though overall levels did not return to the pre-crisis peak. This phase aligns with a more fragmented globalization trend, where FDI flows were less exuberant and marked by regional dynamics rather than broad-based global expansion as during phase 1. In phase 4 (“multiple crises” from 2019 to 2023), the chart reflects a further decline of global FDI flows, exacerbated by events such as the COVID-19 pandemic, war in Ukraine and increasing geopolitical tensions. This phase underscores the challenges of globalization in the face of rising protectionism, economic uncertainty and supply chain disruptions. Overall, the chart encapsulates the evolution of global capital flows through varying phases of globalization, demonstrating the interplay between economic integration, systemic shocks and shifting geopolitical landscapes. Against this background, we construct an index of FDI globalization taking into account other dimensions than FDI flows in percent of GDP. We also aim for the index to be dynamic enough to signal structural shifts as early as possible, which distinguishes it from other indices, as we explain below.

Chart 1

2 Data and index construction

2.1 Data sources

We take data on FDI stocks and flows from UNCTAD’s FDI/MNE database. This database offers both bilateral and aggregated FDI flows and stocks (inward and outward) per country. Data for GDP are taken from the World Bank and data for geographical distance from the CEPII Gravity database. We use data from 2001 to 2022 to ensure a broad data coverage across countries, since bilateral data for 2023 are missing for many countries. The final sample includes 213 countries, 41 of which are classified as advanced economies (AEs) and 155 as emerging markets and developing economies (EMDEs), 17 countries, mostly small island states, are not classified.

Measuring the (de)globalization of FDI, or of any other variable, presents significant challenges due to the inherent complexity of globalization dynamics. By reducing the number of indicators used in our analysis, we mitigate the risks associated with data reliability, such as untimely publication, structural breaks or the discontinuation of time series, which often hinder the robustness of more intricate indices. A streamlined approach also enables the potential integration of our FDI index with existing indices on migration and the planned development of a trade index.

2.2 Intensity

To capture the first dimension, intensity, we start by calculating country-level FDI intensity. We use FDI flows since they have more variation over time compared to stocks, which move quite slowly and would not as quickly indicate shifts in the globalization of FDI. Outward FDI intensity, $OI_{i,t}$, is described by

$$OI_{i,t} = \frac{{fo}_{i,t}.}{Y_{i,t}}$$

(1)

where ${fo}_{i,t}$ are outward FDI flows from country $i\ $divided by the country’s nominal GDP $Y_{i,t}$, with time index $t$. Inward FDI intensity on the country-level is described by

$$II_{i,t} = \frac{{fi}_{i,t}.}{Y_{i,t}}$$

(2)

where ${II}_{i,t}$ is the intensity of inward FDI ${fi}_{i,t}$ of country $i$ divided by the country’s nominal GDP $Y_{i,t}$, with time index $t$.

A higher intensity of inward and outward FDI is interpreted as a sign of more globalized FDI flows. Both variables are interpreted such that an increase of inward or outward FDI flow intensity represents a more globalized FDI landscape.

The global-level intensity of FDI in- and outflows is given by the following two equations. They are a simple, unweighted average across all countries. First, the equation for global FDI outflow intensity:

$${OI}_{g,t} = \ \frac{1}{n}\ \sum_{i = 1}^{n}{OI}_{i,t}$$

(3)

Global FDI inflow intensity is given by

$${II}_{g,t} = \ \frac{1}{n}\ \sum_{i = 1}^{n}{II}_{i,t}$$

(4)

Global FDI intensity, the first dimension is the geometric average of global inflow and outflow intensity, given by

$$I_{G,t} = \ \left\lbrack {OI}_{g,t}{II}_{g,t} \right\rbrack^{\frac{1}{2}}$$

(5)

2.3 Spread

The second dimension we examine is the spread of FDI, which refers to the number of partner countries and the distribution of bilateral shares over which cross-border investments are dispersed from a country’s perspective. To calculate the spread of FDI, we use a widely used measure of concentration, the HHI. Due to the nature of the HHI, we cannot rely on FDI flows and use stocks instead. While FDI stocks can often be negative due to disinvestment and reverse investment, such negative stocks occur much less than negative FDI flows. By using stocks and excluding the few negative values, we ensure a more stable and interpretable dataset for assessing the concentration of FDI relationships. Country-level outward FDI spread is then given by

$$OS_{i,t} = 1 - \sum_{j = 1}^{N}\left( \frac{{so}_{ij,t}}{{so}_{i,t}} \right)^{2}$$

(6)

$OS_{i,t}$ is the spread index of outward FDI of origin country $i$ in time $t$, ${so}_{ij,t}$ is the outward FDI stock of country $i$ in destination country $j$, ${so}_{i,t}$ is the total outward FDI stock of country $i$. An increasing concentration of destination countries will lead to a higher HHI and – as a result of its subtraction from unity – result in a lower country-specific outward FDI spread index and signal increasing deglobalization. A higher spread, i.e. a lower HHI of the outward stock of FDI would result in a higher spread index and signal a more globalized environment for outward FDI. Equivalently, for inward FDI we define

$$IS_{i,t}\, = 1 - \,\sum_{j = 1}^{N}{\,\left( \frac{{si}_{ij,t}}{{si}_{i,t}} \right)^{2}}$$

(7)

$IS_{i,t}$ is the spread index of inward FDI of receiving country $i$ in time $t$, ${si}_{ij,t}$ is the inward FDI of country $i$ from country $j$, ${si}_{i,t}$ is the total inward FDI of country $i$. An increasing concentration of source countries will lead to a higher HHI and – because of its subtraction from unity – result in a lower country-specific inward FDI spread index and signal increasing deglobalization.

Global outward FDI spread is, again, a simple unweighted average given by

$${OS}_{g,t} = \ \frac{1}{n}\ \sum_{i = 1}^{n}{OS}_{i,t}$$

(8)

Global inward FDI spread is given by

$${IS}_{g,t} = \ \frac{1}{n}\ \sum_{i = 1}^{n}{IS}_{i,t}$$

(9)

The global spread of FDI stocks is then, finally, given by

$$S_{G,t} = \ \left\lbrack {OS}_{g,t}{IS}_{g,t} \right\rbrack^{\frac{1}{2}}$$

(10)

A higher spread index signals increasing globalization since countries conduct cross-border investment with a more diverse set of partner countries.

2.4 Distance

The third and last input to our global index is the average geographical distance travelled by one US dollar of FDI. To capture this dimension, we, first, calculate country-level distances of inward and outward FDI, which we then multiply by the countries’ shares in inward and outward FDI. We use stock data again since negative flows would result in negative distances. We start with the distance index for outward FDI given by

$$OD_{i,t} = \sum_{j = 1}^{N}d_{ij,t}\frac{{so}_{ij,t}}{{so}_{i,t}}$$

(11)

where $d_{ij,t}$ is the geographical linear distance between two countries $i$ and $j$ measured in kilometers between the two countries largest cities, $\frac{{so}_{ij,t}}{{so}_{i,t}}\ $is the share of outward FDI of country $i$ in destination country $j$ in the total outward FDI stock of country $i$. The average country-specific distance for inward FDI is given by

$${ID}_{i,t}\, = \sum_{j = 1}^{N}\, d_{ij,t}\frac{{si}_{ij,t}}{{si}_{j,t}}$$

(12)

Where $d_{ij,t}$ is, again, the geographical distance between two countries $i$ and $j$, $\frac{{si}_{ij,t}}{{si}_{i,t}}\ $is the share of inward FDI of country $i$ from source country $j$ in total inward FDI of country $i$. The average distance for global inward and outward FDI is given by the next two equations:

$${OD}_{g,t} = \ \frac{1}{n}\ \sum_{i = 1}^{n}{OD}_{i,t}$$

(13)

and

$${ID}_{g,t} = \ \frac{1}{n}\sum_{i = 1}^{n}{ID}_{i,t}$$

(14)

The global distance is calculated as before and given by

$$D_{G,t} = \ \left\lbrack {OD}_{g,t}{ID}_{g,t} \right\rbrack^{\frac{1}{2}}$$

(15)

FDI positions become more globalized if the average geographical distance becomes larger. In the next section, we combine all previous dimensions into one overall index of FDI globalization.

2.5 Overall global index of FDI globalization

To construct the overall index of FDI globalization, we combine the intensity, spread and distance indices, creating a composite measure that reflects the various dimensions of globalization.

We build a globalization index for both outward and inward FDI. The globalization index for outward FDI is given by

$${OG}_{g,t} = {\lbrack{OI}_{g,t}\ {OS}_{g,t}\ {OD}_{g,t}\rbrack}^{\frac{1}{3}}$$

(16)

and the globalization index for inward FDI by

$${IG}_{g,t} = {\lbrack{II}_{g,t}\ {IS}_{g,t}\ {ID}_{g,t}\rbrack}^{\frac{1}{3}}$$

(17)

The overall FDI globalization index is then constructed as the geometric average of the outward and inward FDI globalization index:

$$G_{g,t} = {\lbrack{OG}_{g,t}\ {IG}_{g,t}\rbrack}^{\frac{1}{2}}$$

(18)

This final FDI globalization index $G_{g,t}$ incorporates the intensity, spread and distance subindices, capturing both inward and outward FDI.

3 Empirical findings

3.1 Intensity of global FDI

Chart 2 depicts the intensity indices ${\ (OI}_{g,t},\ {II}_{g,t})$, constructed as the average across countries for the ratio of FDI flows to GDP, over the period from 2001 to 2022. It differentiates between indices for inward and outward FDI flows, with their geometric average represented as the overall intensity index $I_{g,t}$. The overall discrepancy between inward and outward FDI intensity indices can be attributed to the fact that in most countries, inward FDI flows tend to exceed outward flows. In 2022, 81% of all countries reported higher inflows than outflows, despite globally aggregated inflows being 13% lower than outflows that year. This global mismatch between recorded inflows and outflows stems from data misreporting. Over the sample period from 2001 to 2022, globally recorded inflows have been, on average, 5% larger than outflows, contributing to generally higher indices for inward FDI flows. However, as seen in 2022, when globally reported outward FDI exceeded inward FDI, the key reason for higher inward intensity indices is that the majority of countries report more inward FDI, whereas outward FDI is concentrated in fewer countries, often advanced economies such as Japan or Germany.

Despite this overall difference in levels, inward and outward intensity indices generally follow the same trend.

Chart 2

From 2001 to 2007 (phase 1), we see a general upward trend, reflecting a period of robust global economic growth and increasing cross-border investments. Inward and outward FDI flows followed a similar trajectory during this time. A sharp peak is evident in 2007, marking the height of globalization in terms of FDI intensity, with all indices reaching their highest values in this phase. However, this was followed by a steep decline in 2008 and 2009 due to the global financial crisis, which significantly disrupted cross-border investment activities. The recovery in FDI flows after 2009 (phase 2) was moderate and uneven, as the intensity index failed to return to its pre-crisis peak. From 2010 to 2014, FDI flows exhibited relative stability, with inward and outward flows maintaining similar patterns. However, another surge in intensity occurred in 2015, suggesting a brief period of renewed global economic integration. This singular spike in 2015, however, was almost entirely driven by the Cayman Islands (see chart B1 in box 1 for more information). We decided to keep this observation in the chart since the emergence and existence of offshore financial centers like the Cayman Islands are a feature of a globalized FDI environment. The spike in 2015 was followed by a decline in 2016–17, reflecting increased economic uncertainty and rising protectionist tendencies worldwide (phase 3). The COVID-19 pandemic in 2020 (phase 4) led to a further contraction in FDI flows, evident in a sharp dip in all indices. While some recovery is visible in 2021, the intensity of both inward and outward FDI flows remained below historical peaks, reflecting the ongoing challenges in global investment dynamics. The chart also highlights the volatile nature of FDI intensity over time and underscores its sensitivity to structural shifts in globalization trends, which renders it a useful proxy for the state of globalization. Overall, FDI intensity broadly follows our four distinct phases of globalization: increasing during phase 1, stagnating during phase 2 and decreasing during phases 3 and 4, notwithstanding some increases in between.

3.2 Spread of global FDI

Chart 3

Chart 3 illustrates the spread indices for FDI stocks ${\ (S_{g,t},OS}_{g,t},\ {IS}_{g,t})$ from 2001 to 2022, measuring the degree of globalization in terms of the concentration of inward and outward FDI across partner countries. A higher spread index indicates a more diversified network of FDI partnerships, reflecting greater globalization. From 2001 to 2007 (phase 1), the spread indices for both inward and outward FDI remained relatively stable, with minor fluctuations. This period represents a consistent level of globalization in terms of FDI diversification. However, a sharp decline of the spread of outward FDI stocks is observed in 2008 and 2009 (phase 2), coinciding with the global financial crisis. The reduction of the spread index during this time suggests a contraction in the number of active FDI partnerships, likely driven by heightened economic uncertainty and a general reduction in cross-border investment activities. Post-crisis recovery is evident from 2010 onward, with both inward and outward spread indices gradually increasing. This trend reflects a resumption of global economic integration and a rebuilding of FDI networks. The outward spread index, however, shows greater volatility compared to its inward counterpart, with notable dips in the early 2010s followed by a marked upward trend after 2015. This divergence suggests that outward FDI diversification experienced more pronounced structural changes during the decade. The period from 2015 to 2022 (phases 3 and 4) highlights a significant divergence between inward and outward spread indices. While the inward index stabilizes at a relatively high level, the outward index shows a strong upward trajectory, particularly after 2015. This indicates an expansion of outward FDI partnerships, signaling countries’ increased engagement in cross-border investments. The overall spread index, representing the geometric average of inward and outward indices, reflects these dynamics, capturing the gradual, yet uneven process of globalization in FDI stocks. The chart underscores the importance of partner diversification as a key metric of globalization. It reveals how external shocks, such as the financial crisis, can disrupt established FDI networks, while subsequent recovery and policy shifts drive renewed expansion. The spread index thus provides valuable insights into the evolving patterns of FDI and their implications for global economic integration. Conversely, a higher concentration of FDI stocks, i.e. less diversification, exposes countries to risks related to global supply chain dependency – one of the main risks of deglobalization.

3.3 Distances traveled by FDI

Chart 4

Chart 4 presents the third and last dimension of our globalization index: the distance in kilometers for FDI stocks over the period from 2001 to 2022. The chart is divided into three series: the overall distance index ${(D}_{g,t})$, the distance for inward FDI ${(ID}_{g,t})$ and the distance for outward FDI ${(OD}_{g,t})$. The distance index is a useful metric for assessing the geographical scope of FDI and can serve as a proxy for the degree of globalization. A higher distance value suggests investments are reaching more geographically distant destinations, indicating increased integration across global markets.

The disparity between inward and outward indices is largely driven by countries in Africa, Asia and Latin America and the Caribbean (LAC), which on average receive inward FDI from countries much farther away than the destinations of their outward FDI (see table A1). Inward positions in these regions often originate from the USA or China, while outward positions are typically directed to geographically closer destinations. Additionally, outward FDI positions in many of these countries only play a minor role, 92% of the countries in Africa, Asia and LAC had higher inward than outward FDI positions in 2022, often with substantial differences. If the relatively fewer outward positions were directed to more distant locations, the outward index would still be higher. However, when combining the results of the distance index with the spread index, it becomes evident that outward FDI, particularly from Asia and LAC, is more concentrated and tends to target closer destinations compared to inward FDI.

With these reasons for the differences between outward and inward indices in mind, we see that the global trends reveal a decline in the average distance for both inward and outward FDI from 2001 to approximately 2006. The distance for overall FDI follows a similar downward trend during this period, suggesting a counterintuitive contraction of the distance of investments. This pattern, however, could reflect a concentration of FDI flows among regions that are closer to each other, potentially driven by regional integration processes, such as the rise of intraregional trade agreements or phases of deeper economic and political integration as pursued by the Association of South-East Asian Nations (ASEAN) or Mercosur. Between 2006 and 2009, the decline stopped, with the indices showing minimal fluctuation. The global financial crisis of 2008 does not seem to have caused a significant disruption to average distances, though overall investment volumes were affected during this time. The overall distance experienced a gradual rise from 2010 to 2022, which could be attributed to improvements in global connectivity, advancements in information and communication technologies and growing confidence in legal frameworks supporting cross-border investments. Overall, however, given the relative stability of the distance index, we conclude that the distance dimension does not significantly influence the composite results of the globalization index.

3.4 Overall index of FDI globalization

Chart 5 shows the overall FDI globalization index ${(G}_{g,t})$, the inward FDI globalization index ${(IG}_{g,t})$, and the outward FDI globalization index ${(OG}_{g,t})$, each reflecting a specific aspect of FDI globalization dynamics. From 2001 to 2007 (phase 1), all indices displayed an upward trend, indicating a period of increasing globalization. The inward FDI globalization index rose more sharply than the outward globalization index, suggesting that recipient countries experienced significant integration into global FDI networks during this time. This trend aligned with global economic growth and the proliferation of trade agreements facilitating cross-border investments. Between 2008 and 2009 (start of phase 2), the effects of the global financial crisis became apparent, as all indices experienced a slight stagnation or decline. This downturn was particularly pronounced in the outward FDI globalization index, likely reflecting reduced investor confidence and constrained financial capabilities of multinational enterprises.

Chart 5

The recovery began in 2010, marked by a steady rise in all indices until approximately 2015 (phase 3), when the globalization index and its components reached their peak values. As we discussed before, the 2015 peak was driven by the Cayman Islands, which experienced a significant increase of its already elevated FDI intensity. Once we exclude the Cayman Islands, the 2015 spike disappears (see chart B1 in box 1). However, offshore financial centers such as the Cayman Islands are an essential part of every story of globalization, so we decided to leave the “Cayman spike” in the chart. Post-2015, a period of volatility set in, characterized by sharp fluctuations in the globalization index and its inward and outward components. The peak in 2015 was followed by a noticeable decline in all indices, potentially driven by geopolitical uncertainties, rising protectionism and shifts in global economic priorities. The dip in 2020 (phase 4) is particularly significant, coinciding with the COVID-19 pandemic, which disrupted global supply chains, curtailed international investment flows and heightened economic nationalism.

The globalization index and its components remained below their peak levels until 2022, indicating a partial recovery but also highlighting persistent challenges to FDI globalization. The divergence between inward and outward indices in recent years suggests asymmetrical patterns in FDI integration, where outward flows may still face barriers related to risk aversion and geopolitical tensions, while inward flows appear to stabilize, reflecting a consistent demand for FDI in recipient countries. Finally, the overall index suggests that in 2022, the level of FDI globalization was back at its level seen about 20 years ago during the “goldilocks phase” of globalization.

3.5 Globalization indices on the regional level

In addition to analyzing global trends in globalization indices over time, it is also valuable to examine these indices at a regional level. Table A1 presents the intensity, spread, distance and globalization indices for inward and outward FDI as well as the corresponding combined overall indices. We noted above that the intensity index is higher for inward FDI, as most countries (81% in 2022) report higher inward than outward FDI flows. Table A1 highlights that this pattern is particularly pronounced in Africa, non-euro area Europe, LAC and Oceania, where inflows significantly exceed outflows. This reflects the tendency of EMDEs to have relatively small outward FDI and larger inward flows.

Table A1 also reveals that, as expected, the intensity index is very high for OFCs, but also remarkably high for the euro area. This is driven by a few euro area countries with substantial FDI records, including Cyprus, Luxembourg, Ireland, Malta and the Netherlands.

The differences in the intensity index for inward and outward FDI in OFCs seem surprising, as OFCs primarily act as conduits for investments, meaning inward positions should generally match outward positions. While this might be less applicable to countries like Singapore, with its growing economy and significant industrial sector, in other OFCs, such as the Cayman Islands, the notably larger inward positions are unexpected and may indicate a potential misrecording of FDI positions.

For the spread index, we also generally observe higher values for inward FDI than for outward FDI. The largest differences between inward and outward spread indices are observed in LAC, which can be attributed to the fact that most LAC countries’ outward FDI is directed toward the USA or a few destination countries from the region, while inward FDI positions are also influenced by Asian countries like China or several euro area countries. As described above, this also helps to explain why the distance index shows higher values for inward than for outward FDI. Countries in regions like LAC tend to invest in geographically closer destinations, whereas their inflows typically originate from more distant sources.

The euro area and non-euro area Europe present interesting examples, with relatively short distances observed for both inward and outward FDI, as expected. This pattern reflects the high degree of economic integration in Europe, particularly in FDI relations. However, this highlights a potential limitation of the distance index: stronger integration within Europe results in relatively shorter FDI distances, which is interpreted by the index as “deglobalization.” Nevertheless, the distance index is only one component of the overall globalization index. Despite its potential drawbacks, it provides valuable insights and remains an important element of the broader globalization index.

As a result of generally higher intensity, spread and distance indices, the globalization index also tends to be higher for inward than for outward FDI, with the only exceptions being the euro area and North America. Regarding overall globalization trends, many regions, such as Africa, non-euro area Europe, LAC, North America and OFCs, show a decline in the globalization index from the 2014–2018 period to the 2019–2022 period, indicating increasing fragmentation of global FDI networks. The regional-level analysis highlights the importance of examining the different components of the index. As a composite index, the globalization index enables a straightforward analysis of globalization trends across regions and dimensions.

3.6 A comparison of our index with the KOF globalization index

Finally, we compare our index of FDI globalization to one widely used alternative, the globalization index of the at ETH Zurich’s Konjunkturforschungsstelle (KOF), which was developed by Gygli et al. (2019). In chart 6, our FDI globalization index, represented by the blue line, is juxtaposed with the KOF globalization index (red line), the KOF de facto globalization index (green line), and the KOF de facto financial globalization index (yellow line). We do not consider the de jure components of the KOF index since we do not use any de jure categories and the comparison would be misleading. The chart spans the period from 2001 to 2022, highlighting trends and variations in globalization metrics across these indices.

Chart 6

One prominent feature of our new FDI globalization index is its marked dynamism, as evidenced by the pronounced peaks and troughs over the observation period. Notable fluctuations occurred around 2008 (end of phase 1) and 2020 (start of phase 4), corresponding to the onset of the global financial crisis and the COVID-19 pandemic, respectively. These sharp movements indicate that the index is sensitive to underlying changes in the global economic structure. In contrast, the KOF indices display a relatively stable and gradual progression, with smaller deviations over time. For instance, the overall KOF globalization indices (red and green lines) maintain a consistent upward trajectory with only minor adjustments, suggesting that these measures are less reactive to short-term disruptions or structural shifts in the global economy. The KOF financial globalization index (yellow line) captures the decline after the financial crisis while otherwise displaying a similar stability and upward trajectory as the overall indices. The lack of significant variation in all KOF indices underscores their more inertial nature, potentially reflecting longer-term trends or significant smoothing due to some of the 43 variables entering the KOF index rather than immediate economic changes. The smoother progression of the KOF indices may render them less suitable for capturing quick shifts, especially in volatile periods, compared to the FDI globalization index. Finally, our FDI globalization index fits well into the narrative of changing phases of globalization explained in detail in Abeliansky et al. (2024a): A first increase of FDI globalization during the “goldilocks phase” was followed by years of stagnation during the “hangover phase” before FDI globalization started a decline during the “return of imperialism phase” from 2014 to 2018 and the “multiple crises” phase starting in 2019. ⁵

In summary, our FDI globalization index offers a highly dynamic and responsive metric, capable of reflecting rapid changes in the global economic environment. Its sensitivity to critical global events distinguishes it from the steadier, less volatile profiles of the KOF indices. This distinction highlights the utility of the FDI-focused index as a more nuanced and real-time indicator of globalization, particularly in the context of FDI and its role in shaping the global economic landscape.

4 Conclusion

This paper introduces a novel index of FDI globalization, designed to address shortcomings of existing measures, including a limited focus on bilateral data and the inability to quickly capture shifts in global FDI patterns. By employing three simple dimensions – intensity, spread and distance – our index provides a streamlined, yet comprehensive tool for analyzing the evolution of FDI in an era characterized by mounting geopolitical tensions and deglobalization pressures. Unlike traditional indices, such as the KOF Globalization Index, our approach includes bilateral indicators and ensures methodological simplicity to enhance reliability and applicability, particularly for policymakers and researchers monitoring rapid changes in the global economic landscape.

Our index of FDI globalization suggests that the world has been experiencing a prolonged time of FDI de-globalization since 2016, a few exceptions notwithstanding. This decline was preceded by a stagnating index after the onset of the financial crisis of 2008. In 2022, the last year for which we have data for most countries, the index reached the lowest level since 2003.

The regional analysis of the FDI globalization index reveals three key findings. First, the index is generally higher for inward than for outward FDI, particularly in Africa, non-euro area Europe, LAC and Oceania, where inflows significantly exceed outflows. Second, offshore financial centers (OFCs) and the euro area exhibit particularly high intensity indices, with the latter being driven by a few countries with substantial FDI activity. Third, the globalization index indicates an overall decline from the 2014–2018 period to the 2019–2022 period in Africa, LAC, non-euro area Europe, North America and OFCs, pointing to increasing fragmentation in global FDI networks. These results underscore the importance of analyzing the index's components separately to capture the nuances of regional FDI patterns. Taken together, the developments on the global and regional level point, at the very least, to an ongoing trend toward a less globalized environment for FDI.

Our findings offer valuable policy implications. As deglobalization trends and geopolitical fragmentation are increasingly reshaping global investment dynamics, our FDI globalization index helps policymakers identify vulnerabilities and opportunities in cross-border investment networks. For instance, understanding the spread and intensity of FDI can inform strategies to mitigate risks associated with economic dependencies or disruptions in global value chains. Additionally, the distance dimension sheds light on the role of regionalization as a counterbalance to declining globalization, emphasizing the importance of fostering resilient regional partnerships.

Looking ahead, this index can serve as an integral part of a broader integration of globalization dimensions. Our future research will focus on developing an analogous index for trade, enabling the construction of a comprehensive globalization measure that covers trade, migration and FDI. Such an integrated approach could offer a holistic perspective on globalization trends, further enhancing its utility for academics and policymakers alike.

5 References

Abeliansky, A., C. A. Belabed and J. Mayrhuber. 2024a. (De)globalization monitor: capital flows and cross-border investment . OeNB-Report 2024/8.

Abeliansky, A., C. A. Belabed and J. Mayrhuber. 2024b. Enjoy the silence? (De)globalization and cross-border investment – a gravity approach. OeNB Bulletin Q2/2024.

Aiyar, S. and F. Ohnsorge. 2024. Geoeconomic fragmentation and “connector” countries. CEPR Discussion Paper No. 19352. CEPR Press.

Aiyar, S., J. Chen, C. H. Ebeke, R. Garcia-Saltos, T. Gudmundsson, A. Ilyina, A. Kangur, T. Kunaratskul, S. L. Rodriguez, M. Ruta, T. Schulze, G. Soderberg, J. P. Trevino. 2023. Geoeconomic Fragmentation and the Future of Multilateralism. Staff Discussion Note SDN/2023/001. International Monetary Fund.

Catalán, M., S. Fendoglu and T. Tsuruga. 2024. A Gravity Model of Geopolitics and Financial Fragmentation. IMF Working Paper 24/196.

Czaika, M. and H. de Haas. 2015. The Globalization of Migration: Has the World Become More Migratory? In: International Migration Review 48(2). 283–323.

ECB. 2021. The implications of globalization for the ECB monetary policy strategy. ECB Occasional Paper 263.

ECB. 2024. Navigating a fragmented global trading system: insights for central banks. ECB Occasional Paper 365.

Eurostat. 2024 . Balance of Payments Vademecum. Appendix 7. Available under Publications - Balance of payments - Eurostat

Fitter, J., A. K. Raggl and P. Ramskogler. 2024. The (de)globalization of migration: has the polycrisis period changed the patterns of global migration? OeNB Bulletin Q3/24.

Gopinath, G., P.-O. Gourinchas, A. F. Presbitero and P. Topalova. 2024. Changing global linkages: a new cold war? IMF Working Paper 24/76.

Gygli, S., F. Haelg, N. Potrafke and J.-E. Sturm. 2019. The KOF Globalisation Index – revisited. In: The Review of International Organizations 14 (2019). 543–574.

IMF. 2023a. Geoeconomic fragmentation and foreign direct investment. In: IMF World Economic Outlook (chapter 4). April.

IMF. 2023b. Geopolitics and financial fragmentation: Implications for macro-financial policy. In: IMF Global Financial Stability Report (chapter 3). April.

IMF. 2024. Global prospects and policies. In: IMF World Economic Outlook (chapter 1). October.

Martens, P., M. Caselli, P. De Lombaerde and J. A. Scholte. 2015. New directions in globalization indices. In: Globalizations 12(2). 217–228.

OECD. 2024. Economic Outlook – resilience in uncertain times. Volume 2024. Issue 2.

Tan., J. 2024. How widespread is FDI fragmentation? IMF Working Paper 24/179.

UNCTAD. 2023. World Investment Report 2023 – Investing in sustainable energy for all.

UNCTAD. 2024. World investment report – Investment facilitation and digital government.

Vujakovic, P. 2009. How to measure globalization? A new globalization index (NGI). WIFO Working Papers 343/2009.

6 Appendix A

7 Appendix B

Table A1