Globalization: Innovations in macroeconomic statistics for an interconnected world

Abstract

Globalization has fundamentally altered the landscape of production and income allocation, placing greater demands on macroeconomic statistical frameworks. The rise of multinational enterprises (MNEs), with complex cross-border operations and financial strategies, including transfer pricing and the use of special purpose entities (SPEs), has blurred the connection between where goods are produced and where income accrues. This disconnect complicates the interpretation of key indicators and can lead to misdiagnosis of domestic economic conditions and risks. To address these challenges, user needs have shifted toward more nuanced measures that reveal who truly benefits from economic activity. These evolving needs have driven significant updates to international statistical standards, including the 2025 SNA and BPM7, which introduce new frameworks and supplementary measures designed to better capture the realities of a globalized economy. This article examines the profound impact of MNEs on macroeconomic statistics, highlights the limitations of existing frameworks, and assesses recent advances in international guidance. Drawing on country experiences, it demonstrates the importance of adapting measurement approaches to ensure that economic data remain meaningful and informative in an interconnected world.

1 Introduction

Globalization has transformed the landscape of production and income allocation, challenging the foundations of traditional macroeconomic statistics. As multinational enterprises (MNEs) expand their operations across borders, the conventional residence-based accounting frameworks—such as the System of National Accounts (SNA) and the Balance of Payments and International Investment Position Manual (BPM)—struggle to accurately capture the complexities of modern economic activity. This shift has prompted statisticians and policymakers to reconsider how economic output and income are measured in national accounts and interpreted by users.

MNEs play a central role in this transformation. They distribute their design, manufacturing, and marketing functions across multiple economies. Their sophisticated financial and tax strategies, including transfer pricing and the use of special purpose entities (SPEs), often result in profits being shifted away from the location of actual production. These practices blur the link between where goods are produced and where income accrues, complicating the interpretation of key aggregates such as gross domestic product (GDP). The measurement challenges are further compounded by the fragmented global production arrangements and the relocation of intellectual property products (IPPs) to low-tax jurisdictions. Statisticians therefore face the difficult task of attributing output, income, and assets to the units that control production and bear risk, rather than simply recording activity based on physical location.

In response to these challenges, user needs have evolved toward measures that complement GDP, rather than relying solely on this aggregate. There is a growing demand to give greater prominence to long-standing indicators—such as GNI, net national income (NNI) (Net national income accounts for the depreciation charges of both domestically controlled and foreign controlled units.), gross national disposable income (GNDI), and net national disposable income (NNDI)—that reveal who benefits from economic activity. Users also seek greater transparency regarding MNE and SPE activities, as well as more granular information on cross-border financial flows and interconnectedness. These needs have driven significant updates to international statistical standards, including the System of National Accounts, 2025 (2025 SNA) and Integrated Balance of Payments and International Investment Position Manual, seventh edition (BPM7), which introduced new frameworks and supplementary measures to better reflect the realities of globalized economies.

This article explores the profound impact of MNEs on macroeconomic statistics, the limitations of earlier statistical frameworks, and the ongoing efforts to improve statistical visibility and relevance. Drawing on recent developments in international standards and the experiences of selected economies, it highlights the importance of adapting measurement approaches to ensure that economic data remain meaningful and informative in an increasingly interconnected world.

2 Activities of MNEs and measurement challenges

MNEs—at the core of globalization—have transformed global production and trade by distributing design, manufacturing, and marketing functions across multiple economies. This fragmentation weakens the link between where production occurs and the residence of the institutional units that control it, and challenges the residence-based principles that underpin the SNA and BPM frameworks. ¹

To maximize shareholder returns, MNEs use sophisticated financial and tax strategies. They expand into tax-friendly jurisdictions to minimize corporate income taxes. Furthermore, through intra-group transfer pricing, relocation of IPPs, corporate inversions that change legal domicile without changing economic activity, and internal financing structures that exploit regulatory differences, profits may be recorded in locations with minimal production activities. The OECD/G20 Base Erosion and Profit Shifting (BEPS) framework documents how these strategies erode national tax bases and disconnect reported profits from real economic activity. ²

MNEs organize their production, asset acquisition and internal financial operations through subsidiaries, branches, SPEs, and other affiliates. SPEs are legal vehicles—often with minimal staff or physical presence—created to hold assets, manage financing, reduce regulatory or tax burdens, and channel royalties and dividends within MNE groups (The 2025 SNA and BPM7 provide more detailed guidance on SPEs.). Although SPEs have limited or no domestic economic activity, they can materially distort macroeconomic indicators if they are not separately identified.^1,3

Together, these challenges leave macroeconomic statisticians with the task of reconciling economic substance with statistical guidance. In practice, the challenges for compilers include: (i) attributing output, income, and assets to the unit that controls production and bears risk, even when physical manufacturing is carried out by another unit in a different economy; (ii) correctly classifying global production arrangements (such as processing, factoryless goods production (FGPs), and merchanting) since misclassification can distort GDP and trade statistics; (iii) determining the economic ownership of IPPs (and the risks) in order to accurately assign value added and primary income; (iv) ensuring SNA–BPM consistency so that manufacturing services, merchanting margins, and IPP flows align across national and external sector accounts; and (v) identifying/profiling MNEs and SPEs to prevent duplication and misinterpretation.^1,3

In economies with significant MNE operations, balance sheet indicators (such as total corporate debt to GDP) can misstate domestic risks unless foreign-controlled corporations’ borrowings are separately identified. Disaggregation of the corporate sector balance sheets based on ownership and control allows compilers to distinguish liabilities incurred by domestically controlled non-financial corporations (NFCs)—where the risks are more likely to reside within the domestic financial sector—from those arising from cross-border intra-group financing. Relying solely on consolidated or aggregate data—in the case of debt, for instance—can lead to misdiagnosis of domestic vulnerability.

These challenges underscore the updates introduced in the 2025 SNA to better reflect the effects of globalization. As MNEs expanded their activities and started reallocating profits across jurisdictions, the residence-based recording principles of earlier SNA and BPM frameworks became increasingly challenging in practice. The resulting divergence between GDP and GNI serves as an empirical indicator of these challenges, strengthening the case for improved linkage between production and economic ownership in the statistical frameworks. ⁴

Figure 1 illustrates the widening divergence between GDP and GNI in several highly globalized economies. In economies with significant MNE presence—such as Ireland, Luxembourg, and Singapore—this divergence has widened since the early 2000s, reflecting the growing separation between the location of production and income accrual.^3,4 GNI measures the income earned by the residents of an economy, regardless of where it is generated, and therefore less affected than GDP by globalization (In general, earnings of MNEs reflect income from foreign affiliates (including reinvested earnings on direct investment) while income of domestic affiliates of foreign MNEs are subtracted.). While the difference between GDP and GNI may be modest for large, open economies, it can be very significant for smaller, highly globalized ones.

Figure 1.

GDP-to-GNI ratios in selected economies, 2000–2023. Notes: Ratios > 1 indicate that GDP exceeds income accruing to residents (gross national income, GNI), consistent with net primary income outflows. For Ireland, Luxembourg, and Singapore, the narrowing of the GDP–GNI gap in 2023 reflects a combination of lower GDP growth and improved GNI, largely driven by reduced reinvested earnings and higher income earned abroad by domestically controlled multinational enterprises. Earlier sharp movements—such as those observed in Ireland in 2015–2016—largely reflected balance sheet relocation and corporate restructuring rather than underlying domestic activity. Source: IMF staff estimates, based on data from OECD Data Explorer.

3 Evolving user needs for complementary aggregates

In MNE-intensive economies, reliance on GDP alone is increasingly inadequate for economic analysis and policymaking. Users therefore demand complementary measures that capture productive capacity, the proportion of economic value retained by residents, the extent of foreign-control, and exposure to cross-border risks.^1,3,5 This entails: (i) greater emphasis on income and net measures such as GNI, NNI, GNDI, and NNDI; (ii) more visible identification of MNE and SPE activities within the SNA and BPM frameworks to show growth arising domestically from that driven by foreign-controlled activity; (iii) indicators on interconnectedness, including extensions that go beyond the current conceptual frameworks such as trade in value added (TiVA); (iv) more information related to cross-border flows.

Ireland's experience demonstrates how large divergence between GDP growth and income accruing to residents can distort users’ interpretation of headline aggregates. In 2015, Ireland's Central Statistics Office (CSO) reported a 36 percent increase in nominal GDP (25 percent increase in real GDP) and a 25 percent increase in GNI despite little change in employment.^6,7 The shift was largely driven by MNE relocation of part of their corporate balance sheets—primarily IPP—to Irish enterprises, and recording the associated output for using the IPPs in the production of goods. Much of this production occurred using contract manufacturers (including those domiciled abroad). ⁷ The change reflected a transfer of ownership and risk, rather than just physical manufacturing activity. Furthermore, since the ultimate owners of the Irish enterprises were foreign, much of the recorded value added accrued to non-resident owners as reinvested earnings (Note, reinvested earnings are calculated only for direct investment (and not for portfolio investment).). Hosting high-value functions such as research and development, design, or marketing and sales meant that goods output was being attributed to the economy even when there was not much associated physical manufacturing occurring within their borders.

The contrast with Canada further underscores why prominence of less globalization-sensitive aggregates matter for users. In Canada—a large, diversified economy with predominantly domestic ownership—GDP and GNI move closely together (generally within 2 percent). However, NNI lies consistently below both, reflecting high capital consumption from investment-intensive industries. In this case, depreciation—not profit-shifting—explains the difference between GDP and NNI.

Figure 2 compares GDP, GNI, and NNI for Ireland and Canada between 1995 and 2023. The two panels highlight how these aggregates can convey very different economic narratives depending on the country's exposure to MNE activities and the perspective from which ownership and income are viewed. To improve transparency on domestic conditions, the Ireland CSO introduced a modified gross national income (GNI*). This aggregate excludes the retained earnings of redomiciled headquarters and the depreciation of certain domestic capital assets of foreign-controlled enterprises (This is depreciation related to both the cross-border additions to the stock of IPP assets and the stocks of aircraft involved in international leasing for Ireland.). The CSO further recognized that NNI provided an even more representative measure of income available to the residents in economies with large outflows of property income and depreciation charges.

Figure 2.

GDP, GNI, and NNI in Ireland and Canada. Panel A: Ireland, billions of euros. Panel B: Canada, billions of dollars. Notes: In Ireland, the divergence between GDP, GNI, and NNI reflects the impact of multinational enterprise activity on recorded output relative to income retained by residents. For Canada, the GDP, GNI, and NNI move closely together, reflecting greater alignment between production, income, and ownership. Source: IMF staff estimates, based on data retrieved from the Central Statistics Office (Ireland) and Statistics Canada.

These cases demonstrate the growing demand for macroeconomic indicators that reconcile global production, ownership, and income. Section 4 details the improvements to the statistical frameworks to address these user demands.

4 How do the new 2025 SNA/BPM7 meet user needs?

To mitigate the effects of the challenges outlined above, the 2025 SNA and BPM7 address globalization issues by introducing several conceptual refinements and supplementary indicators. These changes are aimed at providing clarity on production and income attribution in economies affected by MNE activity. These enhancements do not affect the core framework but strengthen their application in the context of increasing international fragmentation and cross-border asset mobility. The enhancements include a collaboratively authored chapter that consolidates previously dispersed content to provide a more integrated and comprehensive approach to globalization-related topics. This chapter—common to both the SNA 2025 and BPM7—places special emphasis on supplementary statistics as well as additional measures and data that enrich traditional macroeconomic analysis. As a result, users gain a deeper and more nuanced understanding of global economic dynamics.

The key updates in 2025 SNA and BPM7 are discussed in detail in the sections below. These updates include enhanced guidance on global manufacturing and distribution arrangements, providing more accurate and comprehensive treatment of activities such as FGPs and ownership of IPPs. The standards also strengthen the recording of MNE activities with the aim of better capturing the economic activities that operate across multiple economies. In addition, the joint 2025 SNA and BPM7 Globalization chapter introduced standard breakdowns for the corporate sectors, distinguishing between foreign-controlled corporations, public corporations, and national private corporations. This granularity clarifies control and ownership structures, especially for national private and public corporations that are part of domestic MNE groups.

For economies where SPEs are significant, the 2025 SNA and BPM7 recommend separately identifying these entities as supplementary information, providing more visibility into their role in global production and financial networks. Finally, the standards encourage the development of extended supply and use tables (eSUTs) and estimates on TiVA as supplementary information, offering a more detailed and accurate representation of global value chains (GVCs) and the distribution of value added at various stages of production.

4.1 Global manufacturing and distribution arrangements

Clarifying global manufacturing and distribution arrangements is central to implementing the principle of economic ownership and control. In response to policy demand to better understand these types of arrangements, the 2025 SNA and BPM7 include descriptions, examples, and a decision tree to identify the appropriate arrangement and its recording.

Distribution arrangements consist of re-exports and merchanting. In both cases, a unit in one economy purchases goods from a non-resident with the intention of selling them—with no substantial transformation from the state in which they were previously purchased—to another non-resident. In merchanting, the goods never physically enter the merchant's economy and are recorded as net exports of goods in the economy of the merchant. Only the trade margin, the difference between the sale and purchase prices, constitutes output of the merchant. However, this margin is recorded in the goods account.

Global manufacturing arrangements include processing arrangements and FGPs. In both cases, a principal outsources the manufacture of goods. In a processing arrangement the principal provides the material inputs while the contractor provides manufacturing services without owning the goods during production. A factoryless goods producer provides the intellectual property—the design or blueprint—into the manufacturing process and completely outsources the production to the contractor, including the acquisition of all or most of the material inputs. FGP was introduced to the macroeconomic standards of the 2025 SNA and BPM7 recognizing the key role of IPPs in production and the model where companies manufacture goods without owning or operating physical factories.

Figure 3 shows an illustrative example of an FGP arrangement. ⁸ In this scenario, the principal located in economy A has the know-how and contracts a contractor in economy B to make a high-end jacket. The contractor in economy B sources materials and produces the jacket according to the design provided by the principal, without paying for the design (IPP). The principal buys the finished jacket at a price that includes the material inputs and work, then sells it to a final buyer in economy C, with the sales price incorporating additional value attributable to the product design and the branding (IPP). The jacket ships directly from economy B to economy C, bypassing economy A. Under 2025 SNA and the BPM7, economy A's output is classified as manufacturing.

Figure 3.

Factoryless goods production (FGP). Source: IMF staff.

Other scenarios of FGP arrangements can have two or more of the entities involved located within the same economy. A situation where finished goods are sold in the same economy as the contractor is common in larger economies. In such cases, there are no entries in the customs data upon completion of the manufacturing process and subsequent sales of the goods. However, the balance of payments of the economy of the principal should record imports from the economy of the contractor, followed by exports to the economy of the contractor. And the economy of the contractor should record exports to the principal followed by imports from the principal.

Significant discrepancies can arise between balance of payments figures and customs data due to global manufacturing. In processing arrangements, customs data typically show a higher trade balance for the contractor's economy compared to its balance of payments goods account, while the principal's economy exhibits the opposite pattern. This discrepancy also applies to FGP arrangements for the principal's economy and may occur in the economy of the contractor, especially when the contractor's economy is also the final buyer's economy. This pattern is observed for China, see Figure 4. The example of China is further elaborated in the Compilation Guidance Note on Factoryless Goods Production. ⁸

Figure 4.

Goods trade balance, customs vs balance of payments. Notes: The bars show the difference between BOP- and Customs-based trade values expressed as a percentage of GDP (right-hand axis). In the most recent period, the lower discrepancy mainly reflects changes in the gap between customs and BOP data. Source: China General Administration Customs, SAFE; Haver Analytics; and IMF staff estimates.

4.2 Improving statistical visibility for the role of MNEs

Recognizing the prominence of global production and its effects, the joint BPM7/2025 SNA Globalization chapter defines the framework for measuring and recording activities of MNEs and SPEs. However, another challenge is how to make those activities visible in the macroeconomic statistics without distorting the interpretation of prominent macroeconomic aggregates. The 2025 SNA and BPM7 recommend complementary approaches to restore analytical meaning. These include: (i) emphasizing existing indicators and supplementary statistics in the SNA/BPM framework; and (ii) separate identification of foreign-controlled and MNE-related units within the institutional sector accounts.

4.3 Emphasizing existing SNA aggregates

While GDP remains the principal measure of economic activity, the last decade has illustrated that it does not always readily convey the full picture when there is substantial MNE activity. In highly globalized economies, a large share of value added is generated by foreign-controlled corporations, and much of their operating surplus accrues to non-resident investors as dividends, interest, or reinvested earnings. For this reason, the SNA 2025 encourages greater emphasis on existing income-based indicators—including net domestic product (NDP), GNI and NNI, GNDI, and NNDI, and household (adjusted) disposable income—as complements to GDP. These aggregates are internationally established and widely recognized indicators that provide a consistent and comparable framework for analyzing economic activity across countries. Furthermore, they are generally less sensitive to globalization-related distortions, making them more suitable for understanding the underlying dynamics of national economies.

The SNA's core indicators are supported by a robust conceptual framework, which has been refined over decades by international organizations and national statistical offices (NSOs). This ensures that the data produced are comparable across countries, maintain objectivity and statistical feasibility, and allows policymakers and analysts to benchmark performance and make informed decisions. In contrast, modified measures—such as Ireland's GNI*—while useful for addressing specific national policy needs, lack international comparability and can complicate the interpretation of economic statistics on a global scale. In the last update of the standards, the international community reinforced that relying on existing measures allows NSOs to leverage established methodologies and maintain continuity in reporting, while also facilitating communication and education for users. Ultimately, emphasizing existing SNA indicators ensures that economic analysis remains grounded in a coherent, integrated, and internationally accepted system, supporting both transparency and comparability in global economic statistics.

For economies with significant foreign trade, adjustments for the terms of trade are necessary to understand the purchasing power of the income available to residents. In this respect, countries with significant international trade in volatile commodities may highlight indicators such as real GDI, which can show a significant divergence from GDP measured in volume terms. This divergence is described as the “trading gain” or loss, which essentially represents the difference between the changes in GDP in volume terms and real gross domestic income (GDI). This measure is particularly sensitive to the terms of trade, which are defined as the ratio of the price of exports to the price of imports. An improvement in the terms of trade—such as rising export prices and falling import prices—enhances a country's ability to purchase goods and services for consumption and investment.

4.4 Supplementary FDI statistics

In addition to clarifying production and income attribution, the BPM7 emphasizes the role of supplementary foreign direct investment (FDI) statistics in improving the interpretation of cross-border relationships. ⁵ In particular, the BPM7 highlights the analytical value of presenting FDI by ultimate investing economy, which allows users to look through complex ownership chains and separately identify the origin of the investment and associated income flows. This complements the standard presentation of FDI by immediate counterpart economy, which may obscure underlying economic relationships when MNE structures involve immediate holding entities or SPEs. Empirical evidence shows that this perspective can substantially reshape bilateral FDI patterns by reallocating investment away from financial conduit economies toward the locations of the controlling parent enterprises, thereby improving the interpretation of cross-border investment positions, income flows, and financial exposures.^9–11

4.5 Institutional sector accounts

The 2025 SNA reinforces the use of Institutional Sector Accounts (ISAs) as a key framework for analyzing and presenting the impact of globalization on production, income and wealth. Within this framework, foreign affiliates of MNEs are treated as resident units in their respective economies. This treatment is designed to allocate production to the economy in which it occurs, which is fundamental for estimating the economy's GDP and other key indicators. ¹ However, in economies where foreign-controlled corporations have a significant presence—and take advantage of global production arrangements—this residence-based allocation can obscure the distinction between domestic and non-resident economic activity. This, in turn, can affect the interpretation of other key market indicators within and across countries.

To improve transparency, the SNA 2025 encourages economies to present separate identification of corporations according to ownership and control. ¹ Economies are encouraged to present separately: (i) foreign-controlled corporations;(ii) domestically-controlled corporations that are part of MNE groups; and (iii) (SPEs) as “of which” categories. Since this disaggregation is resource intensive, the SNA limits this breakdown to the corporate sectors most affected by MNE activity—nonfinancial (S.11) and financial (S.12) sectors.^1,12 This provides transparency without imposing unmanageable reporting burdens. Figure 5 below provides an excerpt of the SNA's decision tree that guides compilers in classifying resident institutional units by sector, ownership, and control.

Figure 5.

Excerpt of 2025 SNA decision tree to classify institutional units. Notes: The figure is adapted from the 2025 SNA. The 2025 SNA provides a structured approach to classify corporations by sector, ownership, and control. Shaded cells are used to denote SPEs as well as foreign-controlled corporations and domestically controlled corporations that are part of MNE groups. Source: IMF staff, adapted from on Figure 5.1 (Illustrative allocation of units to institutional sectors), System of National Accounts, 2025. ¹

The analytical value of ISA granularity is illustrated by several national experiences. Statistics Netherlands demonstrated how the ISAs can be sub-sectored into the foreign-controlled and domestically controlled NFCs. ¹³ They further disaggregated the sector into Dutch MNEs, small and medium-size and large enterprises. This reveals how much output and profit originate in foreign-controlled units. Figure 6 shows that foreign-controlled corporations accounted for nearly one-third of total gross value added (GVA) in the Netherlands between 2015 and 2017, underscoring how GDP reflects the activities of foreign MNEs operating domestically rather than purely domestic activity.

Figure 6.

Gross value added of non-financial corporations by control and enterprise type (Netherlands). Notes: The line indicates total gross value added (GVA) of the non-financial corporate (NFC) sector. GVA is shown in millions of euros. MNE = multinational enterprise; SME = small and medium-sized enterprise; FC NFC = foreign-controlled non-financial corporation. Source: IMF Staff estimates, based on data from. ¹³

Ireland's ISA publications similarly show how ownership breakdowns can contextualize the headline aggregates. As shown in Figure 7, disaggregating GVA by ownership shows that a large share of the recorded GVA is generated by foreign-controlled NFCs, roughly 60 percent of GDP in 2022, before moderating to around 58 percent in 2023. At the same time, significant property income outflows reduced the share of income accruing to resident households.

Figure 7.

Gross value added (GVA) and property income outflows (Ireland). Panel A: GVA of the non-financial corporation (NFC) sector (millions of euros). Panel B: Property income outflows and household income. Notes: In the left panel, gross value added (GVA) is shown in millions of euros and split between domestically controlled and foreign-owned non-financial corporations. In the right panel, net property income outflows are shown on the left axis (in millions of euros), while household income and property income outflows as a share of GVA are shown on the right axis (percent). Property income outflows include dividends, interest, other investment income, and reinvested earnings. Source: IMF staff estimates, based on data from the Central Statistics Office (Ireland).

MNE activity can shift value added from resident units to non-resident sectors as foreign-controlled affiliates may record substantial operating surpluses in the domestic economy while the corresponding property income is ultimately payable to their non-resident parents. ¹ For Ireland, this pattern is evident where foreign-controlled units own the IPP and record significant gross operating surplus domestically—through various production arrangements—even though the related income accrues to foreign investors. ⁷ These results support two pillars of the 2025 SNA: (i) separately identifying foreign-controlled corporations and special purpose entities within the institutional sector accounts, and (ii) the need to complement GDP with income-based measures—such as GNI, NNI—to capture the portion of domestic production that ultimately accrues to residents. ¹

The analysis can be extended to the sectoral balance sheets—financial assets and liabilities, nonfinancial assets, and net worth by sector/subsector. These balance-sheet effects are illustrated in Figure 8.

Figure 8.

NFC debt securities and loan liabilities (Ireland). Note: Figures are shown as a percentage of GDP. Source: IMF staff estimates, based on estimates from Central Statistics Office (Ireland).

Figure 9.

NFC market debt to GDP (Canada and Ireland). Source: IMF staff estimates, based on data from Statistics Canada and Central Statistics Office (Ireland).

Economies that have effectively disaggregated the corporate sectors can assess whether, similar to the flows, the national balance sheet is also being distorted by MNE activities. In such cases, indicators such as corporate debt as a percentage of GDP can be misleading in economies with significant MNE operations. For example, the unusually large balance sheets of Dutch NFCs—reflected in the 2019 total private-sector debt exceeding 133 percent of GDP—may not necessarily be a sign of domestic financial vulnerability, but more so a structural feature of the Dutch corporate sector. ¹³ This positioned the Netherlands among the EU economies with the highest leverage, comparable to Luxembourg and Ireland. ¹³ The size of Dutch corporate sector balance sheets may be explained by the operations of MNE that locate financing and holding entities in the Netherlands. These entities record substantial cross-border assets and liabilities, inflating total debt levels, while generating limited domestic activity or financial risk.

Separating debt held by foreign-controlled NFCs from that of domestically-controlled NFCs is therefore critical for interpreting macroeconomic indicators. Aggregate debt ratios can conceal where the majority of the liabilities are concentrated and whether this debt belongs to globally integrated corporations rather than to the domestic entity itself. Without this disaggregation, headline indicators may misstate both the scale and nature of other resident exposures to the NFC sector.

In Ireland, NFCs reported debt stocks (loans and debt securities) well in excess of GDP—averaging 122 percent of GDP between 2013 and 2023. The majority of this debt is carried by foreign-controlled corporations, much of it is owed to non-resident affiliates rather than to other residentsectors. This produces large gross liabilities for the Irish NFC sector, but it also means that a portion of the recorded debt reflects cross-border intra-group financing.

In Canada, the pattern contrasts with that of Ireland. Figure 9 shows that total NFC debt as a percentage of GDP is lower on average than Ireland's NFC's marketable debt as a percentage of GDP. However, its composition is predominantly domestically concentrated. Most borrowing is undertaken by domestically-controlled NFCs, and funding is sourced primarily from the domestic sectors, particularly chartered banks. This suggests that leverage in Canada is more directly linked to domestic financial conditions. Consequently, any deterioration in the balance sheets of Canadian NFCs could transmit more readily to other resident sectors—such as banks and households—given their strong interconnections through domestic lending channels. Data from Statistics Canada indicate that in the second quarter of 2025, non-residents accounted for only about one-tenth of total lending to NFCs, underscoring that Canada's corporate borrowing is largely financed within the domestic economy. ¹⁴

4.6 GVCs, eSUTs, and TiVA

GVCs refer to the increasingly fragmented production processes that span multiple countries. In a GVC, the creation of a final product—whether a good or service—involves a sequence of activities distributed across resident and non-resident firms. These activities include research and development, production, transportation, marketing, and after-sales services. GVCs are typically coordinated by lead firms, often large MNEs, which orchestrate both core production and supporting services. The rise of GVCs has made it difficult for conventional national accounts and balance of payments data to fully capture the complexity of cross-border production, as intermediate goods and services may cross borders several times before reaching the final consumer. While one cannot fully see the activities of GVCs in conventional national accounts and balance of payments statistics, more granular information such as in a GVC thematic account, a supplementary presentation of international trade and investment income, and more detailed statistics on imports and exports (for example, the total value of reexports and main product and/or partner breakdown) can provide users with a more complete picture of the impact of globalization on the national accounts and balance of payments statistics.

The GVC thematic account approach can be used to better identify and articulate a GVC for a specific product or group of products produced within a GVC. ¹⁵ The GVC satellite account is comprised of GVC-specific SUTs, either national or multi-country, based on an enterprise-centered approach, consisting of integrated and more detailed business statistics and information on business lines and business functions, and GVC-specific ISAs. Accordingly, it includes production, including the income generated and the labor input used in this production, income and investment (both capital and financial), and give information on balance sheets and transactions. This level of detail is not readily available in the existing accounting presentations at the level of sectors or sub-sectors that contain the activities of significant GVCs in an economy.

The TiVA initiative addresses the double counting implicit in gross international trade flows. Conventional international trade data often double count flows, failing to distinguish the actual value added by each country in the production chain. TiVA measures the value that each country adds to goods and services that are exported and ultimately consumed worldwide. This perspective is crucial for understanding the real economic impact of trade, as it reveals how much domestic value added is generated by exports and highlights the foreign content of those exports. TiVA indicators are typically derived from Inter-Country Input-Output tables, which combine national supply and use tables with international trade statistics to map the global flow of value added. Several handbooks, guides, and statistics have been published since 2010 to better address the statistical challenges in understanding the nature of global production, such as the Guide to OECD's Trade in Value Added Indicators: 2023 Edition.

Measuring trade in value added terms provides new insights into trade policy, competitiveness, and the management of global systemic risk. The COVID-19 crisis revealed how GVCs can be vulnerable to economic shocks and disruptions. TiVA data captures the full scope of GVCs, giving policymakers a comprehensive view of how their economy's foreign trade flows affect the world economy. This information helps guide decisions on international trade's impact on growth, competitiveness, global imbalances, macroeconomic shocks, labor markets, and the environment.

More granularity within the SNA framework using the eSUTs also help support GVC and TiVA analyses. Enhancing standard SUTs within the SNA framework, eSUTs offer more detailed breakdowns of industries and products according to characteristics such as ownership status (foreign-controlled or domestically-controlled), enterprise characteristics and global production arrangements. This added granularity helps capture the heterogeneity in production and trade, especially for firms deeply involved in GVCs. eSUTs also connect production accounts to income, employment, and environmental statistics, providing a more comprehensive picture of the economic and social effects globalization.

Together, GVCs, TiVA, and eSUTs offer policymakers and analysts powerful tools to better understand the complexities of global production and international trade. GVC thematic accounts reveal the interconnectedness of economies for a specific product or groups of products using a bottom-up approach. TiVA statistics also provide a more comprehensive view of international trade's economic impact but using a top-down (more macroeconomic) approach. eSUTs enable more targeted and nuanced analysis by capturing firm-level differences. These approaches do not require changes to the core concepts of the SNA or BPM frameworks but rather supplement them with additional data and analytical extensions to improve the quality and relevance of statistics for globalization analysis.

4.7 Special purpose entities (SPEs)

In macroeconomic statistics, SPEs are institutional units defined by a set of specific characteristics. They are legally registered entities that have minimal physical presence in the host economy—not exceeding five employees. These entities are directly or indirectly controlled by non-resident owners. SPEs are established to gain certain advantages such as access to capital markets, isolation of financial risks, reduction of regulatory or tax burdens, or protection of confidentiality for their owners. They generally transact almost exclusively with non-resident parties, and their financial balance sheets largely consist of cross-border claims and liabilities (fiscal activities of a government-owned (non-resident) SPE should be recorded in accounts of the government, even without actual transactions between the government and the SPE.).

SPEs can be categorized by their main functions. These categories include: captive financial units, which are created by financial or nonfinancial non-resident corporations to conduct particular financial activities; specialized financial units used to isolate risks, structure financial transactions or securitize assets for the parent company; corporate groups’ nonfinancial units established for specific nonfinancial purposes, for example, SPEs that hold IPPs may generate revenues from IPP-related services; wealth management units created by households or groups of individuals for holding or managing wealth; and other structures which conduct transactions outside the above categories.

A notable aspect of some SPEs is the concept of pass-through funds, which refers to funds passing through a direct investment enterprise resident in an economy to an affiliate in another economy, so that the funds do not stay in the economy of the first enterprise.

Although SPEs have no or little physical presence, they can have a substantial impact on traditional macroeconomic statistics. For example, they can inflate direct investment statistics due to pass-through funds or services exports if they own IPPs. Recent evidence suggests that SPEs account for around 40 percent of global FDI. ¹⁰ Figure 10 shows the share of foreign financial assets and liabilities attributable to special purpose entities, illustrating the extent to which SPEs can dominate cross-border financial positions and affect the interpretation of external balance sheets. Given the importance of SPEs, the IMF collects supplementary data on SPE activity from compilers, on a voluntary basis.

Figure 10.

Foreign financial assets and liabilities of SPEs as percent of all foreign assets. Source: IMF staff estimates, based on IMF Data.

5 Data sources and statistical infrastructure

Producing granular national statistics that distinguish production and income from foreign and domestically-controlled enterprises requires an infrastructure that links business registers (BR), regulatory information, and targeted surveys into a coherent view of the MNE. In the normal statistical compilation, each economy takes their own (partial) view of the MNE as it operates in their jurisdiction. This is often incomplete and results in under coverage of the MNEs activity. Conversely, given that the MNE extends beyond the national territory, it has the potential to result in duplicate recording of economic activity across the relevant jurisdictions.^1,5,12

Recognizing these challenges, NSOs increasingly view data sharing agreements and global enterprise group registers (GGRs) as essential tools for MNEs’ production, trade, and direct investment relationships.^1,12 However, data sharing can be restricted by confidentiality legislation and other technical barriers. Therefore, the 2025 SNA promotes confidentiality-compliant exchange of aggregated information, metadata, or bilateral reconciliation tables to reduce asymmetries while preserving statistical confidentiality.

Economies that compile high-quality MNE breakdowns—such as Ireland, the Netherlands, and Denmark have developed integrated statistical infrastructure centered around detailed business registers, large case units (LCUs), leverage special enterprise surveys, and regulatory and administrative data. These tools collectively enable compilers to identify foreign-controlled entities, attribute production to ultimate owners, and align institutional sectoring with ownership and control structures in line with the 2025 SNA.^1,12 In Europe, many economies have established LCUs, with Eurostat coordinating regular exchanges through its MNE network to promote consistency and best practices.

5.1 Business registers

The foundation of this infrastructure is a statistical BR. The BR records the legal entities, their UCP, and the chain of control linking resident units to non-resident parents. As this is difficult to operationalize, many economies have enriched their business registers to include: the country of registration and identity of the UCP; a unique group identifier that links all resident entities belonging to the same enterprise group; a binary flag for entities that meet the SPE definition.^3,6 (The 2025 SNA publishes an updated definition of SPEs: An SPE, resident in an economy, is a formally registered and/or incorporated legal entity recognized as an institutional unit, with no or little employment up to maximum of five employees, no or little physical presence, and no or little physical production in the host economy. They are directly or indirectly controlled by non-residents, and are established to obtain specific advantages provided by the host jurisdiction with an objective to (i) grant its owner(s) access to capital markets or sophisticated financial services, and/or (ii) isolate owner(s) from financial risks, and/or (iii) reduce regulatory and tax burden, and/or (iv) safeguard confidentiality of their transactions and owner(s). SPEs transact almost entirely with non-residents and a large part of their financial balance sheet typically consists of cross-border claims and liabilities.)

In this context, there have been a number of ongoing data exchange efforts at the international and regional levels such as the establishment of common business registers for MNEs. Within the European Statistical System, the EuroGroups Register (EGR) now serves as a shared infrastructure linking the business registers of EU Member States. The EGR contains information on more than 151,000 MNEs active in Europe (Multinational enterprise groups in EuroGroups Register (EGR) - experimental statistics). The EGR provides national compilers with consistent ownership and control data to support production of high-quality, comparable MNE data. Building on the Eurostat experience, the UNSD also published a Global Group Register that provides information on the structure of the 500 largest MNEs. The data include their headquarters, subsidiaries, locations, and ownership structures, where available. These databases help facilitate cross-country consistency tests and bilateral reconciliation of intra-MNE flows. ¹⁶

Countries such as Netherlands and Luxembourg, rely on the EGR—combined with a number of administrative data sources—to identify foreign-controlled subsidiaries and companies whose ultimate parents are non-resident. ¹⁶ The EGR's harmonized ownership data provide the backbone for subsectoring NFCs into domestic and foreign control in line with the 2025 SNA recommendations. Statistics Netherlands extends the national business register by linking each Dutch legal unit to: (i) its enterprise group and (ii) the UCP of that group. The identification of the group structure and the UCP cannot be reconstructed purely from Dutch administrative records, since many Dutch-resident affiliates are ultimately controlled by non-residents. That information is obtained through the EGR. The Dutch register combines these EGR links with domestic sources—including administrative data from the Chamber of Commerce, structural business statistics, and supervisory/registry data so that each resident unit can be classified as (a) foreign-controlled, (b) domestically controlled multinational, or (c) other domestic enterprise. ¹³

5.2 Large case units

LCUs are specialized teams that profile large and complex enterprise groups, link legal and statistical units, and reconcile microdata across surveys and administrative records and ensure consistency between the national accounts and external sector statistics. LCUs are especially relevant in smaller, highly globalized economies—such as Ireland, the Netherlands, and Singapore—where a limited number of MNEs dominate aggregates. They coordinate classification (MNE group membership, foreign control, SPE identification); reconcile intra-group flows to minimize asymmetries and double counting and; support bilateral cooperation and data sharing with partner countries where the same group operates.^1,3,12

The experiences described in this paper show that when countries invest in this infrastructure—linking business registers to administrative sources, establishing large case units, and coordinating data sharing—they gain a clearer view of how MNEs shape national accounts aggregates. This clarity supports more coherent global statistics. It also allows users to see how globalized production affects measures such as GDP, GNI, and macroeconomic indicators that can be derived from the sectoral balance sheets. A number of international publications provide practical steps to develop granular MNE data, building on the concepts of the 2025 SNA.^3,12

From a policy perspective, improved visibility of globalization-related output, income, and financial flows has important implications for economic policy analysis. In highly globalized economies, reliance on headline GDP alone can complicate assessments of fiscal space, debt sustainability, and external vulnerabilities, particularly when large shares of recorded output and profits accrue to non-resident owners. Complementary indicators such as GNI, NNI, and sectorally disaggregated national accounts data provide policymakers with a clearer basis for evaluating the domestic income, exposure to cross-border financial risks, and the sustainability of public finances. Enhanced breakdowns of MNE activity and SPEs further support more accurate interpretation of external positions, investment flows, and financial stability indicators. Together, the statistical enhancements introduced in the 2025 SNA and BPM7 strengthen the analytical foundation for fiscal, monetary, and macroprudential policy in an increasingly globalized economic environment.

6 Conclusion

The evolving complexity of global production and the expanding influence of MNEs have exposed significant limitations in traditional macroeconomic statistical frameworks. As highlighted throughout this article, the disconnect between the location of production and the residence of economic ownership has made it increasingly difficult to interpret headline indicators such as GDP and GNI, especially in economies with substantial MNE activity. These challenges have underscored the urgent need for more nuanced and transparent measurement approaches.

The introduction of the 2025 SNA and the BPM7 marks a pivotal advancement in addressing these issues. These updated standards incorporate a range of enhancements designed to improve the visibility and accuracy of MNE and SPE activities within national accounts and balance of payments statistics. Notably, they provide clearer guidance on global manufacturing and distribution arrangements, introduce standard breakdowns for corporate sectors by ownership and control, and recommend supplementary presentations of trade and investment income. These changes ensure that economic data better reflect the realities of modern, interconnected economies.

By encouraging the use of extended supply and use tables, trade in value added indicators, and more granular institutional sector accounts, the new standards empower statisticians and policymakers to distinguish between domestic and foreign-controlled activities. This granularity is essential for understanding the true distribution of value added, income, and financial risks within and across economies. The emphasis on existing income-based measures alongside GDP further supports more meaningful analysis and international comparability.

Ultimately, the updates to the SNA and BPM frameworks represent a collective effort by the international statistical community to keep pace with globalization's impact on economic measurement. If these improvements are adopted, countries could provide users with more relevant, transparent, and actionable data—supporting informed decision making and effective policy responses in an ever-changing global environment.

As the world economy continues to evolve, ongoing collaboration and innovation in statistical standards will remain vital. The 2025 SNA and BPM7 set a new benchmark for measuring and interpreting economic activity, ensuring that macroeconomic statistics remain robust, credible, and fit for purpose in the age of globalization. These frameworks not only improve statistical precision but also strengthen economic governance, allowing policymakers to better evaluate fiscal space, external risk transmission, and corporate sector resilience in highly globalized economies.