Part CXV · The Asset That Is a Policy

Companion essay · European investment

Should Europe borrow more to innovate?

Yes, selectively: there is a strong case for additional European investment financed partly by borrowing, especially for projects that benefit several countries. But the justification is the productive capacity and public benefit created—not a target of bringing European debt closer to US debt. Lower debt can provide room to act; using that room well matters more than using all of it.

Manus AI · 12 November 2026

What your EU–US comparison gets right

Your recollection of the debt comparison fits our existing, matched 2025 fiscal baseline. It puts general-government gross debt at 82.6% of GDP for the EU27 member-country aggregate and 124.4% for the US. The European ratio is 66.4% of the US ratio—approximately two-thirds. These are the frozen October 2026 IMF inputs already used in our model, including some estimates and national accounting qualifications, not new current-quarter observations. The European aggregate includes national-government positions, but does not add a separately measured EU-institution balance sheet or perform a further cross-member consolidation. 1

Europe also has a larger population: Eurostat reports 452.0 million on 1 January 2026, while the latest US annual Census estimate available for this comparison is 341.8 million on 1 July 2025. Those reference dates differ. A larger population expands the potential market, but does not itself establish how much a government can repay; employment, productivity, age structure and taxable income matter. 2 3

“Comparable GDP” needs the PPP distinction we have just developed. In our frozen 2025 inputs, EU27 output is about $21.3 trillion at market conversion versus $30.9 trillion for the US. At purchasing-power parity, the comparison is much closer: 29.5 versus 30.9 trillion international dollars. PPP helps compare economic volume; it is not the currency available to service bonds. 1

Your external-balance observation points in the right direction, but “neutral trade” is not the best description. In Q2 2026, the EU had a seasonally adjusted current-account surplus of 1.6% of GDP, while the US had a deficit of 3.0%. The current account includes goods, services, investment income and transfers. Customs-based goods trade alone can give a different answer and must not be substituted silently. 4 5

Compare the starting debt ratios

2025 baseline · IMF October 2026

Gross general-government debt as a percentage of nominal GDP, not PPP. This is a baseline comparison, not a forecast or a measure of borrowing headroom. A lower debt ratio alone does not answer whether borrowing for innovation is justified.

US general government includes central, state and local government. EU27 is a GDP-weighted member-government aggregate: it excludes separate EU-institution debt and has no cross-member re-consolidation. It is not one borrower or a pooled tax base. The baseline mixes IMF-reported actuals and staff estimates; reported actuals remain revisable.

Italy and the Netherlands are examples already included in EU27, not extra components to add. Changing the view or order never changes the values or the fixed 0–160% axis. The table always includes all four rows.

With JavaScript, hover, focus or tap a value card or a table row’s native button to select its metric and scope detail. Buttons support Tab, Enter and Space. Without JavaScript, the two-bar chart, all four table rows, sources and native expandable limitations remain available; interactive controls are disabled.

2025 gross general-government debt / nominal GDP: EU27 member governments 82.6%; United States 124.4%2 bars on the same fixed horizontal axis from zero to 160%. 2025 baseline · IMF October 2026. Values are rounded to one decimal for display; exact source values are preserved in the input download. EU27 is a member-government aggregate, not one borrower. The table below contains all four rows, including Italy and the Netherlands.
Fixed 0–160% scale; bars begin at zero. Member examples are not additive.

Selected: EU27 member governments

Metric · 2025
Gross general-government debt / nominal GDP: 82.6%
Government scope
GDP-weighted aggregate of 27 member general governments. Excludes separate EU-institution debt; no cross-member re-consolidation. Not one borrower or a pooled tax base.
Observation status
Mixed IMF-reported actuals and staff estimates; actuals remain revisable.

Display rounded to one decimal; exact source number retained in the model-input download. Extra digits in the weighted aggregate do not imply extra measurement precision.

All four 2025 debt ratios · IMF October 2026 · always available without JavaScript
Government / aggregateGross debt / nominal GDP (%)Scope and status
82.6%GDP-weighted aggregate of 27 member general governments. Excludes separate EU-institution debt; no cross-member re-consolidation. Not one borrower or a pooled tax base.Mixed IMF-reported actuals and staff estimates; actuals remain revisable.
124.4%US general government includes central, state and local government, using IMF within-country consolidation. Not federal debt alone.IMF-reported actual; revisable.
137.1%Italy general government, as provided by IMF. Member example already included in EU27; not additional debt to add to the aggregate.IMF-reported actual; revisable.
44.7%Netherlands general government, as provided by IMF. Member example already included in EU27; not additional debt to add to the aggregate.IMF-reported actual; revisable.
How to read the aggregate and its limitations

EU27 = sum(member nominal GDP in USD × member debt_pct_gdp) / sum(member nominal GDP in USD), at average market exchange rates; not PPP. Italy and the Netherlands are illustrative members, not a representative sample or a ranking of all EU members.

Common year, source vintage and general-government category do not eliminate residual national coverage, debt-instrument, valuation and cash/accrual limitations. Member subsectors remain exactly as provided by IMF; some national coverage is incomplete or differently classified. Debt valuation mixes nominal and face value. Fiscal reporting mixes cash and accrual conventions; complete harmonisation is not established.

The aggregate follows IMF within-country debt consolidation without re-consolidating cross-member government claims. Separate EU-institution liabilities are not added. Onlent EU funds may already appear as member liabilities; this is not a total EU institutional balance sheet, a joint repayment resource or an estimate of common borrowing capacity.

Austria’s 2025 fiscal observations are IMF staff estimates; Bulgaria’s fiscal actuals are preliminary, and other reported actuals are revisable. The EU27 row therefore has mixed actual/staff-estimate status, even though the selected US, Italy and Netherlands debt rows are reported actuals.

A debt stock relative to annual GDP does not measure investment returns, financing costs or the ability to service additional borrowing. No new projects, scenarios or future values are calculated here. The approved essay’s argument must be assessed separately from this descriptive chart.

Download existing CXV model inputs (JSON)IMF October 2026 WEO source workbook

Source: International Monetary Fund, World Economic Outlook; 2025 baseline · IMF October 2026. Frozen input snapshot observed 2026-11-11. Only debt_pct_gdp is plotted.

Nominal and PPP: economic-size context

Switch the GDP comparison between market-rate monetary values and purchasing-power-parity output values. The debt ratios above do not change. Converting debt and GDP with the same factor cancels in the ratio; converting only GDP would mix incompatible units. EU27 keeps its original nominal-GDP fiscal weights, never PPP weights.

2025 GDP · trillions of US dollars at market exchange rates. Current-price monetary valuation; not a cross-country volume comparison.

2025 GDP: nominal EU27 and US comparisonZero-origin fixed 0–40 trillion scale in US dollars at market exchange rates. This chart compares economic output, not fiscal debt ratios. The table supplies all four country or aggregate values in both bases.
The currency unit changes with the switch. The scale stays at 0–40 trillions; GDP is output, not tax revenue. EU27 is a country sum, not pooled resources.
2025 GDP · both valuation bases remain available without JavaScript
EconomyNominal (US$ trillions)PPP (international $ trillions)
EU2721.3029.51
United States30.8630.86
Italy2.553.74
Netherlands1.321.52

IMF October 2026, 2025 NGDPD and PPPGDP. PPP values are officially published, but this workbook supplies no actual/estimate classification for those PPP observations. Nominal observations are reported actuals, still revisable. Values are rounded for display; PPP is neither a tradable exchange rate nor an estimate of government repayment resources.

IMF source workbookWorld Bank: appropriate uses of PPP

Explore a conditional 2035 growth scenario

Illustration, not a forecast. This extends our hypothetical 40% emerging-market share of selected high-investment design by 2035. The slider varies assumed emerging-market real growth and its assumed contribution to US/EU export and innovation gains. Growth does not automatically determine design share; the 20% → 40% design-share path is held fixed independently.

5.0% / year

0–8%, in 0.5-point steps; no silent rounding.

Restores the default 5% growth scenario, nominal GDP-size view and hidden trade overlay. This does not switch to the separate no-design-share-migration reference or remove the assumed design shift.

Range 0–8%, in 0.5-point steps. Keyboard arrows change the assumption; Home/End reach the endpoints. Default 5%. Controls work locally without transmitting or storing your choice. Without JavaScript, the default scenario, its chart, table and assumptions remain visible.

Assumed cumulative emerging-market real-output growth: 62.9% over ten years.

United States · 2035 debt / nominal GDP

152.7%

+1.00 percentage points vs no-design-share-migration reference (151.7%).

Assumed incremental domestic export/innovation value added by 2035: 0.40% of reference GDP.

EU27 · 2035 debt / nominal GDP

86.4%

+0.73 percentage points vs no-design-share-migration reference (85.7%).

Assumed incremental domestic export/innovation value added by 2035: 0.60% of reference GDP.

Copies a public link with your growth assumption, nominal/PPP GDP-size basis and trade-overlay choice. The same settings produce the same link. Version 1 identifies the current assumption set, not a permanently immutable model archive. PPP never changes the debt denominator. Shared settings may appear in recipients’ browser history, server logs or existing site analytics when the link is opened. This feature adds no tracking, saved records or login. Nothing is posted automatically; copying requires JavaScript.

Historical trade context

Annual exports minus imports of goods and services divided by nominal GDP—not goods-only trade, the current account or the government budget balance. Fixed EU27 membership excludes the UK. Historical 2015–2024 values stay on their real calendar years, using a separate right axis; they are never projected to 2035 or added to government debt.

Historical trade data and sources
2015–2024 national-accounts trade balance / nominal GDP (%) · historical, revisable
YearUnited StatesEU27
2015-2.8664.340
2016-2.6764.360
2017-2.7704.180
2018-2.8713.666
2019-2.6803.376
2020-2.8943.505
2021-3.5423.688
2022-3.5241.763
2023-2.7423.508
2024-3.0004.125

Selected historical decade 2015–2024, retrieved 13 November 2026 from BEA NIPA and Eurostat ESA 2010; all 20 observations available and revisable. Fixed EU27_2020 membership, including Croatia and excluding the UK throughout. Published national-accounts net exports, not goods-only trade or current account. EU gross trade includes intra-EU flows, which cancel in the net balance; no exports-plus-imports aggregation. Different official statistical conventions and revisions can remain. Historical context, not a forecast or government borrowing capacity; no trade observations after 2024.

A national trade surplus is not a government fiscal surplus. The different debt and trade axes do not imply a causal relation or borrowing capacity. Historical context does not change any scenario calculation. PNG/PDF exports remain fiscal-scenario snapshots; the optional historical trade overlay is not included in those files.

Conditional US and EU27 debt paths, 2025–2035, at 5.0% assumed emerging-market annual real growthGross general-government debt divided by nominal GDP, percent. Solid lines are conditional design-shift outcomes; dashed lines are the no-share-migration references. Fixed 0–240% vertical axis. PPP valuation never changes these fiscal paths. All annual values are available in the expandable table.
US: orangeEU27: blueSolid: scenario · dashed: reference · fixed 0–240% debt scale

On narrow screens, scroll the plot horizontally for readable axis labels. Hover or tap the plot for an exact annual readout. Keyboard: focus the chart, then use Left/Right or Home/End; Escape dismisses. Values update with the growth slider. The table remains available without JavaScript.

Select a year in the plot to inspect the unrounded scenario and reference values.

What your selected scenario implies

At 5.0% assumed annual emerging-market real growth, emerging output grows by 62.9% over 2025–2035. The hypothetical design share still moves from 20% to 40% independently of this setting. This is a conditional model illustration, not a prediction of countries’ growth.

2035 US–EU debt gap

The modeled US debt-to-GDP ratio is 66.23 percentage points higher than the EU27 aggregate. This is US minus EU27, calculated before display rounding—not a percentage change or a difference in currency debt amounts.

The no-migration reference gap is 65.96 percentage points; this scenario changes that signed gap by +0.27 percentage points. Each ratio uses its own region’s nominal GDP. EU27 is not a pooled treasury, and this gap alone is not a measure of borrowing capacity or debt affordability.

United States

Assumed export/innovation gains reach 0.400% of reference GDP, against 0.375% of unrecovered design-displacement losses. Modeled 2035 domestic output is therefore 0.025% higher than the no-migration reference.

With the frozen revenue yield and an additional policy budget cost of 0.200% of reference GDP, the annual worsens by 0.192 percentage points of reference GDP in 2035. This is a flow difference, not cumulative debt or measured tax revenue.

The model’s 2035 gross debt / nominal GDP is 152.7%, 1.00 percentage points above its unchanged 151.7% reference. A lower ratio does not by itself mean debt is falling, affordable or safely below a policy limit.

EU27 aggregate

Assumed export/innovation gains reach 0.600% of reference GDP, against 0.500% of unrecovered design-displacement losses. Modeled 2035 domestic output is therefore 0.100% higher than the no-migration reference.

With the frozen revenue yield and an additional policy budget cost of 0.200% of reference GDP, the annual worsens by 0.154 percentage points of reference GDP in 2035. This is a flow difference, not cumulative debt or measured tax revenue.

The model’s 2035 gross debt / nominal GDP is 86.4%, 0.73 percentage points above its unchanged 85.7% reference. A lower ratio does not by itself mean debt is falling, affordable or safely below a policy limit.

Faster emerging-market growth can increase the assumed demand gains and offset some modeled displacement, but the transmission is analyst-selected, not an estimated trade elasticity. revenue/GDP is an aggregate proxy, not a marginal tax rate. At 0% growth, demand gains disappear while design displacement and policy costs remain. EU27 is not a pooled treasury; valuation and the historical trade overlay do not feed into these debt calculations.

The compares goods-and-services exports and imports, not public revenue or government spending. Hover, focus or tap the underlined terms to inspect their meaning; press Escape to dismiss an explanation.

Calculation basis: frozen IMF October 2026 fiscal inputs and the explicit assumptions in the annual-values disclosure below. Values are rounded here for interpretation, not forecast precision.

Economic terms and sources
Purchasing Power Parity (PPP)
A common-price conversion used to compare the volume of goods and services produced or purchased across economies. It is not a tradable exchange rate or money available to repay debt. Changing this display basis does not change the scenario’s debt-to-GDP ratios. World Bank: appropriate uses of PPP
trade balance
Exports minus imports of goods and services. Here the historical series is national-accounts net exports divided by nominal GDP, not the current account, goods-only trade or a government budget balance. The historical overlay is context, not an input to the debt calculation. BEA: net exports of goods and services
General-government
The government sector, including central/federal, state or regional, local government and social security funds where applicable. The EU27 figure aggregates member-government accounts; it is not one pooled treasury and excludes separate EU-institution liabilities. Source coverage and valuation qualifications still apply. IMF: government finance statistics
primary balance
The government’s revenue less expenditure excluding interest. This scenario uses a net-primary-compatible balance derived consistently with its net-interest proxy; it is an annual flow, not the accumulated debt stock. A positive modeled balance is a surplus. ECB: debt dynamics and primary balances
Export your selected scenario

Exports capture the growth setting and historical-trade choice at the moment you click. PNG/PDF: when enabled, the 2015–2024 trade lines, separate axis and sources are included alongside 2025–2035 debt paths. Reference lines and scenario warnings remain; menus and hover tooltips are not exported. CSV: 22 annual debt rows (USA and EU27, 2025–2035), unrounded scenario/reference values and percentage-point differences, units, frozen source vintage and warnings; regional mechanism values appear only on 2035 rows. Historical trade is omitted from CSV regardless of the overlay and is not a causal debt-model input. Files are created locally in your browser; your settings are not uploaded. All exports require JavaScript. “Download started” means a browser download was requested, not confirmation that a file was saved.

Inspect the assumed fiscal transmission and annual values

Analyst-selected assumptions, not measured coefficients: baseline nominal US/EU growth 4%; effective net financing-cost proxy rises gradually by 1 percentage point with a six-year time constant. Selected design exposure is 3% of reference GDP for the US and 4% for EU27; half of displaced activity is replaced. Additional policy cost reaches 0.2% of reference GDP in each region. No Italy/Netherlands projection is inferred.

At 5% assumed emerging-market growth, incremental US/EU gains reach 0.4% / 0.6% of reference GDP. The slider scales these gains by [(1 + selected growth)^10 − 1] / [(1.05)^10 − 1]. Gains and costs phase in linearly with the hypothetical design shift; this demand relationship is not empirically calibrated. Growth at zero sets demand gains to zero, but does not remove the independently assumed design displacement or policy cost.

The frozen 2025 total-revenue/GDP yield applies to modeled domestic output; the net-primary-compatible spending residual follows reference GDP. Annual debt = prior debt + modeled net interest − primary balance; debt ratios use each year’s nominal GDP. No PPP denominator substitution, new inflation/FX sensitivity, debt-maturity model, automatic policy response or risk-premium feedback is estimated. This is not an IMF forecast. General-government coverage and EU aggregation limits above remain.

Conditional debt / nominal GDP (%), 2025–2035
YearUS scenarioUS referenceEU27 scenarioEU27 reference
2025124.4%124.4%82.6%82.6%
2026126.7%126.7%82.6%82.6%
2027129.2%129.2%82.7%82.7%
2028131.9%131.8%83.0%82.9%
2029134.6%134.4%83.3%83.2%
2030137.4%137.2%83.7%83.5%
2031140.3%140.0%84.1%83.9%
2032143.3%142.8%84.6%84.3%
2033146.4%145.7%85.2%84.7%
2034149.5%148.7%85.8%85.2%
2035152.7%151.7%86.4%85.7%

Frozen fiscal basis: existing CXV general-government inputs. Emerging-market growth and the design-share path are hypothetical aggregates; countries differ. These results illustrate a possible mechanism, not inevitable Western decline or guaranteed fiscal gains from faster emerging-market growth.

Why a surplus can support the investment argument—but does not settle it

An external surplus means that an economy's aggregate saving exceeds its domestic investment. That can coexist with successful exporters, precautionary saving, demographic pressures or insufficient attractive investment opportunities at home. It is not evidence, by itself, that Europe is investing too little. Nor is it money sitting in a public EU account. The IMF stresses that external deficits and surpluses must be judged by their causes rather than labelled automatically good or bad. 6

Nevertheless, this creates a serious policy question: could Europe channel more of its savings into productive European projects instead of leaving worthwhile projects unfunded? Draghi's competitiveness report identifies weak productive investment despite ample private savings and argues for capital-market integration alongside public support. It proposes joint funding for European public goods such as breakthrough innovation and cross-border grids, subject to political and institutional conditions. 8

Additional investment can initially increase imports of machinery, materials or technology and reduce an external surplus. That would not automatically be a failure. The relevant question is whether those purchases create durable productive capacity, useful services and future income.

Trust must connect to a credible repayment arrangement

You are right that borrowing allows future income to support investment today. But a government repays through taxes, fees and other fiscal resources—not directly through national GDP or a company's worldwide sales. An investment can have a high social return without generating enough revenue for the particular institution that borrowed.

For example, a European research programme might improve products sold worldwide. Its benefits could include European wages and tax receipts, but also cheaper products for foreign buyers or profits taxed elsewhere. The borrowing decision therefore needs to identify who receives the benefits, who pays the financing and operating costs, and what happens if the technology fails.

A small arithmetic example illustrates the distinction. A one-off loan equal to 1% of GDP, at a hypothetical 4% interest rate, costs 0.04% of GDP in annual interest. If government captured a hypothetical 40% of additional output in revenue, a permanent 0.1% of GDP output gain would cover that interest alone. This is not a project forecast: it excludes principal repayment, maintenance, implementation delays and risk. It shows why “growth-enhancing” and “self-financing” are not synonyms.

The ECB's 2026 analysis supports productive public investment, particularly when it complements private capital. Yet its illustrative German investment simulations also show that output can rise while the debt ratio rises. Better investment is not guaranteed to pay for itself, and efficiency, financing costs and spending composition materially affect the result. 7

What Europe should finance—and at which level

My assessment is that the strongest case is for European-scale capacity that individual countries or private firms would otherwise underprovide. Examples include interconnected electricity networks, shared research infrastructure, advanced technical training, and competitively selected early-stage innovation. These can reduce costs or enable production across many industries, rather than subsidising only one existing producer. Draghi's report explicitly links common funding to such cross-border public goods. 8

For aviation, electric vehicles and other specialised industries, public support should address a demonstrable technology, coordination or infrastructure problem. Financing shared research, testing or complementary infrastructure can be more defensible than indefinitely paying firms to preserve market share. Support should remain open to new entrants and be assessed against alternatives, including collaboration with emerging-market producers. This is a policy judgement, not a claim that every existing subsidy fails.

The choice is not exclusively public debt. Private investment, public loans, guarantees, procurement and grants serve different purposes. Loans require a plausible repayment stream; grants may be justified for knowledge spillovers. Guarantees and public lending still carry risk and must not be treated as costless. Europe's investment strategy should combine these instruments rather than turn the whole investment need into sovereign borrowing. Parliament's 2026 study also warns that promised private-capital leverage may not materialise consistently. 10

Crucially, EU27 is not one borrower with one pooled tax base. In our same baseline, Italy's debt ratio is about 137%, while the Netherlands' is about 45%. The aggregate cannot transfer one government's financing room to another. National borrowing must fit each country's fiscal position; common borrowing requires an agreed EU-level repayment resource, governance and allocation of costs. It is not created by adding up national GDP. 1

The existing fiscal framework retains the 3% deficit and 60% debt reference values, with country-specific expenditure paths. Qualifying reforms and investments can support extending the adjustment period from four to up to seven years, but there is no unlimited exemption for expenditure labelled “investment”. Those are the current rules; a broader permanent common-investment facility would require the relevant political and legal decisions. 9

The answer to your proposal

Europe should be more willing to borrow for well-chosen productive investment, but less willing to use borrowing as an indefinite substitute for competitiveness. The test is whether a project adds useful capacity beyond what would otherwise happen, can actually be delivered, and has a financing arrangement that remains credible if growth or revenues disappoint.

For cross-border benefits, common European financing has a stronger rationale than competing national subsidy packages. It can help exploit Europe's market scale without requiring every country to duplicate the same facilities. Public returns, project milestones and the conditions for stopping unsuccessful support should be visible from the beginning.

The danger is not only borrowing too much. It is also refusing worthwhile investment until productive capabilities deteriorate and the future tax base becomes weaker. But avoiding that danger does not mean copying US debt levels. Trust is best preserved by making the future gains—and the obligations used to obtain them—credible.

References

  1. [1]IMF October 2026 WEO workbook; frozen 2025 fiscal and PPP inputs used in our existing models
  2. [2]Eurostat population and population change statistics, July 2026 data
  3. [3]US Census Bureau Vintage 2025 population estimates, released January 2026
  4. [4]Eurostat EU current account in the second quarter of 2026
  5. [5]BEA US international transactions and investment position, second quarter 2026
  6. [6]IMF Finance and Development: Current Account Deficits
  7. [7]ECB 2026: Boosting efficiency in public investment in times of fiscal constraint
  8. [8]Draghi 2024: The future of European competitiveness, Part A
  9. [9]Council of the EU: Economic governance framework and country-specific fiscal rules
  10. [10]European Parliament 2026: Investment needs in the Draghi and Letta reports and implications for the EU budget