Decoded, Economics & finance

The One-Firm Recovery: Dutch Growth Without ASML

The One-Firm Recovery: Dutch Growth Without ASML
The Dutch Economy in 2026 | Analytical Series, Paper 1

Executive Summary

On 30 October 2026, Statistics Netherlands (CBS) reported that the Dutch economy grew by 0.6 percent in the third quarter and by 2.1 percent year on year, its strongest annual pace since the end of 2022. The headline reads like the confirmation of a broad recovery. This paper argues that it is something narrower and more fragile: a recovery in which one company, ASML, now accounts for a share of measured growth that has no precedent in the Netherlands and only one modern precedent in Europe, Nokia in Finland around the turn of the century.

The question we ask is simple and, to our knowledge, has not been answered with a quarterly time series: what has the Dutch economy been doing once ASML is taken out? To answer it, we construct a counterfactual series of Dutch GDP excluding ASML's domestic value added, quarter by quarter, from 2022 to the third quarter of 2026. The series combines the CBS volume index of GDP with ASML's reported quarterly net sales, converted into Dutch value added through a ratio calibrated to independent estimates of the company's weight in the economy, and deflated into constant prices. We then use the series to decompose recent growth, to measure how concentrated Dutch growth has become, and to stress test 2027 under six deterministic scenarios and 200,000 simulated paths.

The main findings are as follows.

  1. The acceleration is almost entirely an ASML story. In the year to the third quarter of 2026, headline GDP grew by about 2.05 percent in our reconstruction. GDP excluding ASML grew by about 1.48 percent. ASML's direct contribution was roughly 0.59 percentage points, close to 29 percent of all annual growth. Under a deliberately conservative smoothed variant, the contribution is still 0.40 points and the underlying economy grows by 1.67 percent. Whatever the specification, the underlying economy did not accelerate in 2026; only the headline did.
  2. In the third quarter alone, ASML explains around 60 percent of growth. Of the 0.55 percent quarter-on-quarter rise in our reconstructed GDP, about 0.33 points come from the increase in ASML's value added. The rest of the economy grew by about 0.22 percent, its second weakest quarter in two years. CBS itself notes that industry produced about half of the quarter's growth and that the machinery sector stood out.
  3. The underlying growth rate of the Netherlands is about 1.4 to 1.5 percent, and has been for two years. Excluding ASML, annual growth was 1.12 percent in 2024, about 1.45 percent in 2025 and about 1.54 percent in 2026 on our projection. These are respectable numbers for a mature economy with a growing population, but they are not the numbers the headline suggests, and they imply per capita growth of roughly one percent.
  4. ASML's direct share of Dutch GDP is now about 1.6 percent for 2026 and will reach about 2.1 percent in the fourth quarter if the company meets the midpoint of its guidance. This is below Nokia's peak of about 4 percent of Finnish GDP in 2000, but it is rising quickly, and with supplier and investment effects included the weight approaches the levels Finland reached in the middle of the 2000s.
  5. 2027 contains a statistical trap. ASML's fourth-quarter guidance implies an annual run rate of about 59 billion euros, 32 percent above its expected 2026 total. If the company's 2027 sales grow by a healthy 10 percent, its quarterly sales must still fall by about 17 percent from the fourth-quarter peak. A strong year for ASML can therefore subtract from Dutch quarterly growth for several quarters in 2027.
  6. The downside is asymmetric and quantifiable. With non-ASML growth held at 1.3 percent, a normal semiconductor downturn (sales down 20 percent) would cut 2027 GDP growth to about 0.8 percent, a deep downturn (down 35 percent) to about 0.5 percent, and a 2009-type collapse to about 0.2 percent. In our Monte Carlo simulation, the median outcome for 2027 is about 1.5 percent growth, but there is an 11 percent probability of growth below 0.5 percent and a 16 percent probability that ASML alone subtracts more than half a percentage point.
  7. The growth is profit-intensive and therefore income-light. Between the third quarter of 2025 and the third quarter of 2026, ASML's sales rose by about 4.1 billion euros and its net income by about 2.8 billion euros. Close to seven of every ten additional euros of sales became profit, a large part of which accrues to an internationally dispersed shareholder base. GDP measures where value is produced, not who receives it. The household income channel of the ASML boom is far thinner than the GDP channel.

The policy conclusions follow directly. Fiscal planning should treat ASML-related revenue as cyclical windfall, not structural capacity. Wage negotiations and productivity debates should rely on an ex-ASML benchmark. CBS and the Netherlands Bureau for Economic Policy Analysis (CPB) would serve public debate by publishing a regular indicator of growth excluding the largest single firm contributions. And the government's strategic bet on Brainport Eindhoven should be understood for what it is: a concentration of national macroeconomic risk in one supply chain, one technology and one geopolitical fault line.

1. Introduction: Good News That Needs Careful Reading

For most of 2023 the Netherlands was in a technical recession. Output fell for three consecutive quarters, households absorbed the largest inflation shock in four decades, and manufacturing contracted as European gas prices and weak Chinese demand hit the industrial core. By the autumn of 2026 the picture appeared transformed. The CBS volume index of GDP (2021 equal to 100) rose from a trough of 103.8 in the third quarter of 2023 to 109.5 in the third quarter of 2026. Year-on-year growth reached 2.1 percent, ahead of the euro area, ahead of Germany, and well above the 1.5 percent the government had set as its structural ambition for the economy. The Netherlands, it seemed, had rediscovered its growth engine.

The difficulty is that the engine is unusually easy to identify. The same quarter in which Dutch GDP accelerated, ASML reported record net sales of 11.7 billion euros, a gross margin of 57.5 percent and net income of almost 5.0 billion euros, and guided to fourth-quarter sales of between 14.2 and 15.2 billion euros. Its full-year guidance of 44 to 45 billion euros implies growth of around 36 percent on 2025. The national accounts show the trace of this boom with little ambiguity: in the third quarter, CBS reported that the value added of manufacturing rose by 2.7 percent on the previous quarter, that the machinery sector in particular saw significant growth, and that exports of machinery and transport equipment rose markedly. Commentators, including ING, have estimated that ASML on its own accounts for about 1.6 percent of Dutch GDP in 2026, and around 2.2 percent once its domestic suppliers and investment are included.

These estimates of ASML's weight in the level of GDP are useful, but they leave the more important question unanswered. A firm that represents 1.6 percent of output can still account for a much larger share of output growth, if its own growth is much faster than that of the economy around it. And what matters for policy, for wage setting, for fiscal planning and for the assessment of the Dutch growth model, is not ASML's share of the level of GDP but the behaviour of everything else. Has the Dutch economy outside the lithography cluster recovered? Is it accelerating, decelerating, or simply drifting? How much of the improvement in 2026 would survive a turn in the semiconductor cycle?

This paper answers these questions with a counterfactual time series. Section 2 explains why a single company can move a national aggregate and why this matters more for growth than for levels. Section 3 sets out the data and the method, including its limits. Sections 4 to 6 present the results: the divergence between headline and underlying growth, the anatomy of the third quarter, and the nature of the growth that ASML generates. Section 7 places the current boom in the context of ASML's own history of drawdowns. Section 8 stress tests 2027. Section 9 compares the Dutch case with Finland's experience with Nokia. Section 10 draws the policy implications, and Section 11 concludes. A methodological appendix documents the equations and the robustness checks.

2. Why One Company Can Move a Nation

2.1 The granular economy

The conventional view of macroeconomics treats the economy as a large number of small firms, whose idiosyncratic shocks cancel out in the aggregate. If that were true, no single company could matter for GDP growth: the law of large numbers would wash out the fortunes of any individual firm. Xavier Gabaix's work on the "granular origins of aggregate fluctuations" overturned this view. When the size distribution of firms is fat tailed, as it is in every advanced economy, idiosyncratic shocks to the largest firms do not cancel out. They explain a material share of aggregate volatility, and the share rises as the economy becomes smaller and more concentrated.

The Netherlands is an almost perfect laboratory for granularity. It is a small, very open economy whose industrial structure is dominated by a handful of global firms. Some of these, like the large oil and chemical groups, have a large footprint in gross output but a smaller one in domestic value added, because much of what they process is imported. ASML is different. Its machines are designed, integrated and tested largely in Veldhoven, with a deep Dutch supplier network around Eindhoven, and its products sell for prices that embed very high margins. A euro of ASML sales therefore carries an unusually large share of domestic value added, and the company's growth rate, which swings between minus 40 percent and plus 80 percent across the cycle, is an order of magnitude more volatile than that of the economy around it.

2.2 Levels and growth are different questions

The distinction between a firm's share of the level of GDP and its share of the growth of GDP is the conceptual core of this paper. Suppose a company produces 1.5 percent of national value added and grows by 30 percent in a year, while the rest of the economy grows by 1.4 percent. The company then contributes 0.45 percentage points of growth (1.5 times 0.30), and the rest of the economy contributes about 1.38 points (98.5 times 0.014). Headline growth is 1.83 percent, of which the company explains a quarter, even though it represents only one sixtieth of output. If the company's sales then fall by 30 percent, the same arithmetic turns a contribution of plus 0.45 points into a contribution of minus 0.45 points: a swing of 0.9 points in national growth, from a firm that produces 1.5 percent of output.

This is exactly the situation of the Netherlands in 2026. ASML's share of GDP is modest. Its share of the change in GDP is not.

2.3 Gross domestic product is not gross national income

A further distinction matters for the interpretation of ASML-driven growth. GDP measures the value added produced on national territory. It does not tell us who receives the income generated. When value added consists mainly of wages paid to domestic workers, the two concepts move together. When it consists mainly of corporate profits, and when those profits belong to shareholders who are largely outside the country, GDP can rise substantially while national income rises much less. The Finnish research institute ETLA made precisely this point about Nokia: since a large majority of Nokia's shares were held abroad, the consequences of the rise and fall of its profits were felt mainly outside Finland, and Nokia's contribution to Finnish growth was overstated during the boom. Section 6 shows that the same mechanism is at work in the Netherlands today, and that it is stronger in 2026 than at any time in ASML's history, because the company's margins are at a record.

3. Data and Method

3.1 Sources

The analysis uses three sets of data. The first is the CBS quarterly volume index of GDP, seasonally adjusted, with 2021 equal to 100, as published in the first estimate for the third quarter of 2026. The series runs from the third quarter of 2022 to the third quarter of 2026 and incorporates the upward revision of the second quarter to 0.7 percent growth. The second is ASML's reported quarterly total net sales from the first quarter of 2021 to the third quarter of 2026, as published in the company's quarterly results releases, together with the midpoint of its fourth-quarter 2026 guidance (14.7 billion euros). The third is a set of annual macroeconomic aggregates: nominal GDP in 2021 (870.6 billion euros) and the GDP deflator path implied by CPB and Ministry of Finance projections, which we use to convert ASML's nominal value added into constant 2021 prices.

3.2 From ASML sales to Dutch value added

ASML does not report its Dutch value added. We therefore estimate it as a fixed ratio of group sales. The ratio must capture three things at once: the share of group sales that becomes value added rather than purchased inputs (a large part of the optical system, for instance, is bought from Zeiss in Germany); the share of group value added that is produced in the Netherlands rather than in the company's US, Asian or German operations; and the timing difference between revenue recognition and production.

We calibrate the ratio so that our estimate of ASML's share of Dutch GDP matches the independent estimate published by ING for 2025 (about 1.3 percent) and 2026 (about 1.6 percent). A ratio of 0.45, meaning that each euro of ASML group sales generates about 45 cents of Dutch value added, reproduces both figures closely. As a plausibility check, ASML's third-quarter 2026 results imply that group-level value added, the sum of profits before tax, personnel costs and depreciation, amounts to something in the region of 60 percent of sales. A Dutch share of around three quarters of that total is consistent with the company's footprint. We nevertheless treat the ratio as the main uncertainty of the exercise and report all key results for ratios of 0.35, 0.45 and 0.55.

3.3 Constructing the counterfactual

For each quarter, we compute the level of real GDP in 2021 prices, at an annualised rate, by applying the CBS volume index to the 2021 nominal level. We compute ASML's real Dutch value added, also annualised, by multiplying quarterly sales by four and by the value-added ratio, and dividing by the quarterly GDP deflator. GDP excluding ASML is the difference. From these levels we derive year-on-year and quarter-on-quarter growth rates for headline and ex-ASML GDP, ASML's share of GDP, and ASML's contribution to growth in percentage points. The equations are set out in the appendix.

3.4 Limits and robustness

Three limitations deserve emphasis. First, revenue recognition is lumpy. ASML books a machine when the customer accepts it, and its fourth quarters are systematically strong. The CBS seasonally adjusted series does not necessarily reproduce this pattern, because national accountants record production rather than accounting revenue and because the seasonal adjustment absorbs part of the recurrent fourth-quarter spike. This is visible in the data: in the quarters when ASML's sales surge, our ex-ASML series shows weak quarter-on-quarter growth, which suggests that part of the measured ASML surge is already smoothed in the official aggregate. To address this, we report a second specification in which ASML's sales are averaged over two quarters. Second, the value-added ratio may vary over time, in particular as margins rise. Third, our deflator is an annual path interpolated to quarters. None of these limitations changes the direction of the results; Table 1 shows how much they change their size.

Table 1. Robustness of the third-quarter 2026 decomposition

VA ratio ASML sales timing Headline y/y (%) Ex-ASML y/y (%) ASML contribution (pp) Share of growth ASML share of GDP (%)
0.35 Reported quarter 2.05 1.61 0.46 22% 1.32
0.35 Two-quarter average 2.05 1.76 0.31 15% 1.19
0.45 Reported quarter 2.05 1.48 0.59 29% 1.70
0.45 Two-quarter average 2.05 1.67 0.40 19% 1.53
0.55 Reported quarter 2.05 1.35 0.72 35% 2.08
0.55 Two-quarter average 2.05 1.58 0.49 24% 1.87

Source: author's calculations from CBS and ASML data.

The table yields a robust conclusion. Across all six specifications, ASML explains between 0.31 and 0.72 percentage points of the 2.05 percent year-on-year growth recorded in the third quarter of 2026, that is, between 15 and 35 percent of the total. Underlying growth lies between 1.35 and 1.76 percent. In no specification does the underlying economy reach the 2 percent mark suggested by the headline.

4. Results I: Two Economies

4.1 The divergence

Figure 1 plots headline GDP growth and our estimate of growth excluding ASML, both year on year. For most of the period since 2023 the two lines are close. They diverge in two episodes, at the end of 2024 and in the course of 2026, both of which coincide with surges in ASML's sales. The second divergence is the larger one, and it is still widening.

Figure 1. Two economies: headline growth and growth without ASML
Figure 1. Two economies: headline growth and growth without ASML. Year-on-year change in GDP volume. Source: CBS; ASML results releases; author's model (Dutch value added equal to 0.45 times ASML group sales, deflated).

The pattern is striking because it reverses the conventional narrative of 2026. In that narrative, the economy accelerated through the year as household consumption recovered and government spending on defence and public services increased. Our series tells a different story. The underlying economy grew by 1.30 percent in the year to the first quarter of 2026, 1.68 percent in the year to the second quarter, and 1.48 percent in the year to the third quarter. The acceleration of the headline, from 1.41 to 2.05 percent, did not come from the domestic economy.

4.2 The contribution

Figure 2 isolates ASML's direct contribution to year-on-year GDP growth. It was negative through most of 2024, when the company described the year as a transition and its sales fell in the first half. It turned positive at the end of 2024, faded again in 2025, and rose sharply in 2026, reaching about 0.59 percentage points in the third quarter.

Figure 2. ASML contribution to year-on-year GDP growth
Figure 2. ASML contribution to year-on-year GDP growth. Bars: ASML direct contribution in percentage points. Line: headline growth. Source: author's model.

Table 2 summarises the annual picture. ASML's contribution was slightly negative in 2024, small and positive in 2025, and substantial in 2026, when it accounts for about 0.40 percentage points of the 1.92 percent annual growth that our reconstruction projects for the year.

Table 2. Annual growth with and without ASML (base specification)

Year Headline GDP growth (%) GDP excluding ASML (%) ASML contribution (pp) ASML share of GDP (%)
2024 1.08 1.12 -0.03 1.14
2025 1.56 1.45 0.13 1.25
2026 (projection) 1.92 1.54 0.40 1.63

2026 assumes ASML Q4 sales of 14.7 billion euros and non-ASML growth of 0.35% q/q in Q4. Source: author's model.

Two observations follow. First, the underlying growth rate of the Dutch economy has been remarkably stable since mid-2024, at about 1.4 to 1.5 percent. Second, the gap between this underlying rate and the headline has become large enough to change the qualitative reading of the economy. A 2 percent economy in which wages, profits and tax receipts are broadly based is a different object from a 1.5 percent economy with a single fast-growing export cluster attached to it.

4.3 Which sectors carried the underlying economy?

The CBS production tables allow a cross-check of the counterfactual from the supply side. Table 2b compiles the quarter-on-quarter growth of value added by sector for the first three quarters of 2026, as published in the second- and third-quarter releases. The pattern confirms the reading of Figure 1.

Table 2b. Value added by sector, quarter-on-quarter volume change (%), 2026

Sector Q1 2026 Q2 2026 Q3 2026
Manufacturing -2.1 1.8 2.7
Trade, transport, hospitality 0.1 2.2 -0.3
Information and communication 1.9 0.9 1.3
Business services 0.0 1.0 0.6
Government, education, care 0.8 0.3 0.5
Financial institutions 2.7 -0.3 0.7
Real estate activities 0.4 0.1 0.4
Construction -1.1 -0.6 0.1
Energy 4.1 -0.3 3.6
Mining and quarrying -2.1 -3.2 6.9

Q1 from the CBS Q2 release; Q2 (revised) and Q3 from the CBS Q3 first estimate, 30 October 2026.

Manufacturing is the only large sector whose growth accelerated through the year, from a contraction of 2.1 percent in the first quarter to growth of 1.8 and 2.7 percent in the second and third. CBS attributes this acceleration in large part to machinery, with petroleum refining also contributing in the second quarter as the closure of the Strait of Hormuz reshaped European fuel flows. Construction contracted in the first two quarters and barely grew in the third, consistent with the constraints imposed by nitrogen permitting and grid congestion. Trade, transport and hospitality swung from a strong second quarter to a contraction in the third. Information and communication, business services and the public sector grew steadily but at moderate rates.

Read together, the sector data describe an economy in which the services core grows at a pace of roughly 0.3 to 0.6 percent a quarter, the building sector is stuck, and manufacturing growth is concentrated in a narrow set of activities. That is exactly the economy that our ex-ASML series measures.

Figure 9. Quarterly growth split between ASML and the rest of the economy
Figure 9. Quarterly growth split between ASML and the rest of the economy. Percentage points of quarter-on-quarter GDP growth. Source: author's model.

Figure 9 presents the same story in a single picture, splitting each quarter's growth between ASML and the rest of the economy since the end of 2023. The rest of the economy has produced positive growth in every quarter since the start of 2024, a genuine sign of resilience. But its contribution has fluctuated between about 0.1 and 0.7 points with no upward trend, while ASML's contribution has become the swing factor that determines whether a quarter looks strong or weak.

4.4 Per capita

The Dutch population is growing by roughly 0.5 to 0.6 percent a year, mainly through migration. Underlying growth of 1.5 percent therefore corresponds to per capita growth of about 0.9 to 1.0 percent. This is consistent with the CPB's own observation that the purchasing power of households will barely grow over the coming years. The disconnect between a headline growth rate above 2 percent and a public that perceives little improvement in living standards is not a puzzle. It is largely explained by the composition of growth.

5. Results II: The Anatomy of a Quarter and the Weight of a Firm

5.1 The third quarter of 2026

Figure 4 decomposes the quarter-on-quarter growth of the third quarter of 2026. In our reconstruction, GDP rose by 0.55 percent. The increase in ASML's real value added accounts for about 0.33 percentage points. The remainder of the economy grew by about 0.22 percent.

Figure 4. Anatomy of the 0.6% third quarter
Figure 4. Anatomy of the 0.6% third quarter. Quarter-on-quarter growth of reconstructed GDP, Q3 2026. Source: author's model.

Two independent pieces of evidence support this reading. CBS reports that industry accounted for about half of the quarter's growth, with machinery as the standout sector, and that investment fell slightly, so that the strength did not come from domestic capital spending. The production table published by CBS also shows that the trade, transport and hospitality sectors, the most direct barometer of domestic activity, contracted by 0.3 percent in the quarter, after a strong second quarter. Outside the high-technology cluster, the Dutch economy slowed in the summer of 2026.

5.2 The rising share

Figure 3 shows our estimate of ASML's direct share of Dutch GDP. It fluctuated between about 0.9 and 1.5 percent from 2022 to 2025, rose to 1.7 percent in the third quarter of 2026, and will reach about 2.1 percent in the fourth quarter if the company delivers the midpoint of its guidance. For 2026 as a whole, the share is about 1.6 percent, in line with the ING estimate on which our calibration rests.

Figure 3. ASML's estimated direct share of Dutch GDP
Figure 3. ASML's estimated direct share of Dutch GDP. The last point assumes fourth-quarter 2026 sales at the midpoint of ASML guidance (14.7 billion euros). Source: author's model.

5.3 The carry-over trap of 2027

The fourth-quarter guidance has a consequence that has received little attention. Sales of 14.7 billion euros in a single quarter correspond to an annual run rate of about 58.8 billion euros, 32 percent above ASML's expected total for 2026. This creates a large statistical carry-over into 2027, in both directions.

If ASML's 2027 sales are 10 percent higher than in 2026, at about 48.9 billion euros, its average quarterly sales will be about 12.2 billion euros, some 17 percent below the fourth-quarter 2026 level. If sales are flat on the year, the average quarter will be 24 percent below the fourth quarter. Only with annual growth above 30 percent would ASML's quarterly sales in 2027 stay at or above the end-2026 peak.

The implication for the Dutch quarterly growth profile is that a good year for ASML can still produce negative quarter-on-quarter contributions to GDP in the first half of 2027. If CBS's seasonal adjustment absorbs part of the fourth-quarter spike, as our robustness analysis suggests, the effect will be attenuated. But it will not disappear. Markets and policymakers who read the Dutch quarterly accounts in 2027 should expect an economy whose headline growth weakens even if ASML performs well, and should not mistake that arithmetic for a domestic downturn.

5.4 How much of Dutch volatility is ASML?

Over the eleven quarters from the fourth quarter of 2023 to the third quarter of 2026, the variance of ASML's quarterly contribution to growth equals about 19 percent of the variance of headline quarterly growth, and the correlation between the two is about 0.25. A firm producing between 1 and 2 percent of output thus accounts for close to a fifth of the measured volatility of the national growth rate. The covariance between ASML's contribution and the rest of the economy is negative in our series, which we interpret, as explained in Section 3.4, as partly reflecting the smoothing of ASML's revenue spikes in the official seasonally adjusted aggregate. The direct variance share is therefore an upper bound for measured volatility, but a lower bound for the economic exposure, since it ignores the supplier network.

6. Results III: What Kind of Growth Is It?

6.1 Profit-intensive growth

ASML's growth in 2026 is not only fast; it is exceptionally profitable at the margin. Between the third quarter of 2025 and the third quarter of 2026, total net sales rose from 7.5 to 11.7 billion euros, an increase of about 4.1 billion. Over the same period net income rose from about 2.1 to 5.0 billion euros, an increase of about 2.8 billion. Roughly 68 percent of the additional sales therefore became net profit. The gross margin rose from about 51.6 percent to 57.5 percent, and the company indicates that the effects of value-based pricing will become partly visible in 2027 and fully visible from 2028.

This matters for the Dutch economy because the composition of value added determines who benefits. When an increase in value added consists mainly of profits, the first-round effect on household income runs through dividends, share buybacks and capital gains, not through wages. ASML's share register is dominated by international institutional investors; Dutch households own a small fraction of it, directly or through pension funds. A large part of the profit increase recorded in Dutch GDP is therefore income that accrues abroad. In national accounting terms, the gain in gross national income is materially smaller than the gain in GDP.

6.2 Employment-light growth

The same logic applies to employment. ASML employs a large and highly paid workforce in the Netherlands, and its supplier network adds several times that number. But the company's employment does not rise in proportion to its sales: a 36 percent increase in sales in 2026 is achieved with a far smaller increase in headcount, through higher prices, a richer product mix and higher utilisation. Our estimate is that ASML's Dutch employees represent in the order of a quarter of one percent of Dutch employment, against a direct share of value added of about 1.6 percent. The ASML boom therefore raises measured labour productivity, which CBS reports rose by 2.4 percent in 2025, without raising the productivity of the median Dutch worker.

6.3 Why the distinction matters for wage setting

The Netherlands runs a coordinated system of wage bargaining in which aggregate productivity growth and the profitability of the corporate sector are reference points. If the headline productivity and profit figures are inflated by a single firm, they can send a misleading signal in two directions. Unions may cite aggregate profitability to justify wage claims that most firms, operating at the underlying 1.4 percent growth rate, cannot sustain. Employers in the rest of the economy may find their competitiveness compared unfavourably with an aggregate that they do not, in fact, belong to. A benchmark that excludes ASML, as this paper proposes, would provide a more accurate reference for the economy in which most Dutch workers are employed.

6.4 Tax receipts as a windfall

The fiscal channel is more direct. Corporate tax on ASML's Dutch profits, payroll taxes on its workforce, and the taxes paid by its suppliers flow into the Dutch budget. In a year in which the company's net income is likely to approach 16 to 17 billion euros at group level, and in which its effective tax rate has been in the region of 16 to 17 percent, the corporate tax paid by the group is likely to be in the order of three billion euros, much of it in the Netherlands. In a budget that the minority Jetten government is balancing against a 2.7 percent deficit, rising defence spending and cuts to health care and social security, revenue of this size and volatility should be treated as cyclical. We return to this point in Section 10.

6.5 A nation within the nation

The ASML boom also has a geography. The company's headquarters, its main assembly and test halls and the densest part of its supplier network are concentrated in the Brainport region around Eindhoven, in the province of North Brabant. In the regional accounts, this concentration means that the province's growth rate, its labour market and its housing market increasingly follow a different rhythm from the rest of the country. Tight housing and congested electricity networks in the region are not local inconveniences: they are the binding constraints on the expansion of the most productive cluster in the Dutch economy, and they are the reason the government's Beethoven programme was conceived.

The regional concentration has a second consequence that is easy to overlook. Because the cluster's demand for skilled labour, housing and grid capacity is so intense, its growth raises costs for every other firm in the region. Engineers hired by ASML and its suppliers are engineers not hired by medical technology firms, by defence electronics companies or by the energy transition. In periods of boom, the cluster can crowd out the very diversification that would reduce the economy's dependence on it. This is a familiar mechanism in resource economies, where it is known as the Dutch disease, a term coined to describe the effects of Groningen gas on Dutch manufacturing in the 1970s. It would be an irony of economic history if the Netherlands were to experience a technological variant of its own disease.

7. The Cycle Behind the Boom

7.1 A growth company with violent drawdowns

Figure 5 shows ASML's annual net sales since 2000 on a logarithmic scale. The long-run trend is spectacular: sales have grown at an average annual rate of about 10 percent over a quarter of a century, from 2.7 billion euros in 2000 to an expected 44.5 billion in 2026. But the path has included five years of falling sales, four of them by a fifth or more. Sales fell by about a third in 2001, by about a fifth in both 2003 and 2008, and by 46 percent in 2009, when the global financial crisis froze investment in chip production. In one year out of five, ASML's sales have contracted.

Figure 5. ASML net sales, 2000 to 2026
Figure 5. ASML net sales, 2000 to 2026. EUR billion, logarithmic scale. Orange bars mark years of falling sales; 2026 is the guidance midpoint. Source: ASML annual reports and results releases.

The volatility of annual sales growth, measured as the standard deviation of log changes, is about 31 percent. This is the volatility of an early-stage technology company, not of a mature industrial group, and it is the volatility that is now being imported into Dutch GDP.

7.2 What drives the current cycle

Three forces explain the 2026 boom, and each carries its own risk.

The first is the investment cycle in artificial intelligence. The build-out of data centres for training and running large models requires leading-edge logic and high-bandwidth memory chips, which in turn require ASML's extreme ultraviolet (EUV) systems. ASML expects EUV sales to rise by more than 45 percent in 2026, and its management describes order discussions extending to 2029 and 2030. The risk is the classic risk of capital-goods cycles: if the returns on AI investment disappoint, or if capacity is built faster than demand, orders for the equipment that produces the chips will fall faster than the demand for the chips themselves. This is the accelerator principle, and ASML sits at the top of the chain.

The second force is China. China's share of ASML's sales fell from 33 percent in 2025 to about 20 percent expected for 2026, as US export controls tightened and Chinese customers worked through earlier orders. But in the third quarter of 2026, sales to China rose by 109 percent on the previous quarter, lifting China's share of system sales from 14 to 23 percent. According to reports from the company's investor communications, Chinese memory and storage makers have been stockpiling lithography equipment to cover several years of expansion in case supply is cut off. Stockpiling borrows demand from the future. It is a source of strength in 2026 and a source of weakness later.

The third force is policy. The United States is pressing the Netherlands to restrict sales of less advanced lithography equipment to China, and the proposed MATCH Act in Washington would extend US controls extraterritorially. The Dutch government has publicly opposed extraterritoriality, but its room for manoeuvre is limited, as the Nexperia affair demonstrated in 2025. A decision to cut deep ultraviolet sales to China would remove a large block of ASML's revenue in a single administrative step.

7.3 Booms are not followed by busts on average, but the tail is fat

A natural question is whether years of very strong ASML growth tend to be followed by contractions. In the historical record since 2000, six years recorded growth above 30 percent. The years that followed them recorded average growth of about 11 percent, and none recorded a contraction; the weakest was plus 2 percent. On this evidence, a boom does not mechanically produce a bust the following year. But the sample is small, and the deepest contractions in the record coincided with external shocks (the collapse of the dot-com boom and the global financial crisis) rather than with the ordinary rhythm of ASML's own cycle. The relevant risk for 2027 is therefore not mean reversion but an external shock to AI investment, to China, or to US-Dutch export policy.

7.4 The supplier multiplier

ASML's direct value added is only part of its economic footprint. The company purchases a large share of its modules, mechatronic sub-assemblies, precision components and engineering services from suppliers, many of which are located in the Netherlands, and particularly in the Brainport region. ING's estimate that the company's total weight, including domestic suppliers and investment, reaches about 2.2 percent of GDP implies that each euro of ASML's direct value added is accompanied by roughly 0.375 euros of additional domestic value added elsewhere. We use this ratio throughout the stress tests.

The supplier multiplier is likely to be asymmetric over the cycle. In an upswing, suppliers expand capacity gradually, constrained by skilled labour, space and grid connections, so that part of the additional demand is met by imports or by longer delivery times. In a downswing, orders to suppliers are cut quickly, and smaller firms with thinner balance sheets absorb a disproportionate part of the adjustment. Several Brainport suppliers depend on ASML for a large majority of their turnover. The 1.375 factor is therefore probably a lower bound for the downside and an upper bound for the upside, which reinforces the asymmetry of the stress test results in Section 8.

8. Stress Test: 2027 Under Six Scenarios and 200,000 Simulations

8.1 Deterministic scenarios

We now translate ASML's possible 2027 trajectories into Dutch GDP growth. The method is straightforward. We hold the growth rate of the economy excluding ASML at 1.3 percent, in line with the CPB and private-sector consensus for underlying growth, and compute the change in ASML's real Dutch value added under each scenario for its 2027 sales. To capture the effect on domestic suppliers and investment, we scale the direct effect by a factor of 1.375, the ratio between ING's total estimate of ASML's economic footprint (2.2 percent of GDP) and its direct estimate (1.6 percent).

Figure 6. 2027 GDP growth under six ASML revenue paths
Figure 6. 2027 GDP growth under six ASML revenue paths. Non-ASML growth fixed at 1.3%; supplier spillover factor 1.375. Source: author's model.

Figure 6 and Table 3 present the results. If ASML's sales rise by 20 percent in 2027, the company adds about 0.39 percentage points to Dutch growth, which reaches about 1.67 percent. A rise of 10 percent adds 0.17 points. A flat year subtracts slightly, because of the higher deflator. A normal semiconductor downturn, with sales down 20 percent, subtracts about half a percentage point and brings growth down to 0.79 percent. A deep downturn of 35 percent subtracts 0.81 points. A collapse of the 2009 type, with sales down 46 percent, subtracts more than one percentage point and leaves Dutch growth at about 0.2 percent, at the edge of stagnation.

Table 3. 2027 stress test

Scenario ASML sales growth 2027 Direct effect (pp) Incl. suppliers (pp) Dutch GDP growth 2027 (%)
Boom continues (+20%) +20% +0.28 +0.39 1.67
Consensus-type (+10%) +10% +0.12 +0.17 1.45
Plateau (0%) +0% -0.04 -0.05 1.23
Normal downturn (-20%) -20% -0.35 -0.49 0.79
Deep downturn (-35%) -35% -0.59 -0.81 0.46
2009-type collapse (-46%) -46% -0.77 -1.06 0.22

Non-ASML growth fixed at 1.3%. Source: author's model.

The asymmetry is the key result. The plausible upside for Dutch GDP from ASML in 2027 is about 0.4 percentage points. The plausible downside is more than twice as large. Because ASML's base is now so high, every percentage point of its growth is worth more to Dutch GDP than at any time in the past, in both directions.

8.2 Monte Carlo simulation

Deterministic scenarios do not tell us how likely each outcome is. To estimate the probability distribution of 2027 growth, we ran 200,000 simulations. In each, ASML's 2027 sales growth is drawn at random from the 25 annual growth rates recorded between 2001 and 2025, a bootstrap that preserves the fat tails of the historical distribution. Non-ASML growth is drawn from a normal distribution with a mean of 1.3 percent and a standard deviation of 0.6 points, reflecting the usual uncertainty of macroeconomic forecasts one year ahead.

Figure 7. 200,000 simulated 2027 outcomes
Figure 7. 200,000 simulated 2027 outcomes. Distribution of simulated Dutch GDP growth in 2027. ASML growth bootstrapped from 2001-2025 history; non-ASML growth normal with mean 1.3% and standard deviation 0.6 points. Source: author's Monte Carlo.

The median outcome for Dutch GDP growth in 2027 is about 1.48 percent. The central 50 percent of outcomes lies between 0.94 and 2.03 percent, and the central 90 percent between 0.13 and 3.17 percent. The probability that growth falls below 0.5 percent is about 11 percent; the probability that it is negative is about 4 percent. The probability that ASML alone subtracts more than half a percentage point is 16 percent. Table 3b reports the full distribution.

Table 3b. Distribution of simulated 2027 Dutch GDP growth (200,000 draws)

Statistic Value
5th percentile 0.13%
10th percentile 0.43%
25th percentile 0.94%
50th percentile 1.48%
75th percentile 2.03%
90th percentile 2.60%
95th percentile 3.17%
Probability growth below 1.0% 27%
Probability growth below 0.5% 11%
Probability growth below zero 4%
Probability ASML subtracts more than 0.5 pp 16%
Probability ASML adds more than 0.5 pp 28%

Source: author's Monte Carlo simulation.

These probabilities should be read as conditional on history being a fair guide to ASML's future volatility. That assumption is conservative in one respect, since the company's dominant position in EUV and its long order book reduce the probability of a 2009-type collapse, and aggressive in another, since the concentration of demand in AI investment and the politicisation of the China market create risks that did not exist in earlier cycles. On balance, we regard the simulated distribution as a reasonable description of the uncertainty that the Netherlands has imported by tying a growing share of its growth to one firm.

8.3 How to read the Dutch quarterly accounts in 2027

The combination of the carry-over trap described in Section 5.3 and the fat-tailed distribution described above suggests a practical reading guide for the quarterly releases of 2027.

  1. Compare each quarterly GDP release with ASML's results for the same quarter, published about two weeks earlier. A fall in ASML's sales from the fourth-quarter 2026 peak will mechanically subtract from Dutch growth, even if the company's annual performance is strong.
  2. Look first at the production table, not the expenditure table. The machinery sector's value added is the closest official proxy for ASML's contribution. Growth in trade, transport and hospitality, business services and construction is the better gauge of the underlying economy.
  3. Treat a weak first half of 2027 with caution before concluding that the economy has turned. If the underlying sectors continue to grow at about 0.3 to 0.4 percent a quarter, a soft headline is an arithmetic echo of the 2026 boom, not a recession signal.
  4. Watch ASML's order intake rather than its sales. Bookings lead sales by several quarters. A sustained fall in bookings, particularly for EUV systems, would be the earliest reliable warning of the downturn scenarios in Table 3.
  5. Watch Washington and Beijing as closely as Veldhoven. A tightening of export rules on deep ultraviolet systems, or a Chinese decision to stop stockpiling, would shift the distribution of outcomes to the left more quickly than any domestic indicator.

9. The Nokia Mirror

9.1 What Finland teaches

No European economy has experienced a single-firm dependence comparable to the Finnish experience with Nokia. According to ETLA, Nokia represented about 4 percent of Finnish GDP in 2000 and contributed around half of Finland's growth in that year, and about a quarter of Finnish growth over the decade from 1998 to 2007. Its share of GDP had fallen to 2.6 percent by 2008, to 1.6 percent in 2009, and to about half a percent by 2011. In 2009, Nokia alone subtracted about one percentage point from Finnish growth, and the electronics industry as a whole about two points of an 8 percent contraction of GDP.

Figure 8. The one-firm economy, then and now
Figure 8. The one-firm economy, then and now. Nokia: share of Finnish GDP (ETLA, Yle). ASML: share of Dutch GDP (author's model calibrated to ING). Different methods; orders of magnitude only.

Figure 8 compares the two trajectories. ASML's direct share of Dutch GDP is still well below Nokia's peak. But the trajectory is upward, and on the broader measure that includes suppliers and investment, ASML's footprint is approaching the share that Nokia held in Finland in the late 2000s. The comparison should be read with care, because the two sets of estimates use different methods. It is a comparison of orders of magnitude, not of decimals.

9.2 What is different

The differences matter as much as the similarities. Nokia sold a consumer product in a market with low barriers to entry, and it lost its position when the smartphone redefined the market. ASML holds a technological monopoly in EUV lithography that no competitor has come close to challenging, and its customers, the leading chip manufacturers, have no alternative supplier. Its risk is not obsolescence in the short term, but cyclicality and geopolitics. Nokia's decline was a structural loss of market share; an ASML downturn would most likely be a cyclical fall in volumes from which the company would recover. For the Dutch economy, this is a more favourable profile than Finland faced.

But the macroeconomic lesson is the same. Finland's public finances, its regional economy around Oulu and Salo, and its productivity statistics were all shaped by one company, and the adjustment when that company's fortunes turned was long and painful. Finland's growth after 2008 was among the weakest in the euro area for a decade, and part of the explanation was that the economy outside Nokia had been weaker than the headline numbers suggested during the boom. That is the warning contained in Figure 1.

9.3 An objection: is it fair to exclude ASML?

A reasonable objection to the exercise in this paper is that ASML's growth is real Dutch growth. The machines are built in the Netherlands, the engineers are paid in the Netherlands, and the company's success is the product of decades of Dutch public and private investment in research, education and industrial networks. Why should an analysis subtract the best part of the economy?

The answer is that the exclusion is a diagnostic device, not a value judgement. Central banks routinely look at core inflation, which excludes food and energy, not because food and energy prices are unimportant but because their volatility obscures the underlying trend that policy must respond to. In the same way, growth excluding ASML is not a claim that ASML does not count. It is a measure of the trend in the rest of the economy, which is where most employment, most household income and most tax bases are located, and which is the part of the economy that most domestic policy instruments can actually influence. A complete reading of the Dutch economy requires both numbers: the headline, which tells us what the country produces, and the underlying rate, which tells us how most of the country is doing.

10. Policy Implications

10.1 For fiscal policy: treat ASML revenue as windfall

The Dutch fiscal framework separates trend-based spending ceilings from the cyclical behaviour of revenue. That framework was designed for an economy in which the cycle is broad and roughly symmetric. In an economy where a measurable share of corporate tax revenue depends on one firm with a volatility of 30 percent a year, the framework should be extended. We propose that the Ministry of Finance identify, in each budget, an estimate of the corporate tax and payroll tax receipts attributable to the ASML cluster, and that receipts above a moving average be treated explicitly as windfall, unavailable for structural spending commitments. In a year in which the government is financing a permanent increase in defence spending with a combination of a new surcharge and cuts to social security, the temptation to rely on buoyant corporate tax receipts will be strong. It should be resisted.

10.2 For statistical policy: publish growth excluding the largest contributors

CBS already publishes growth by sector and has, on occasion, identified the role of individual sectors in quarterly growth. We propose a regular indicator, published with each quarterly estimate, of GDP growth excluding the contribution of the largest single firms, or at least excluding the semiconductor equipment industry. Ireland offers a precedent: the distortions created by multinational accounting led the Central Statistics Office to develop modified gross national income (GNI star) as a more reliable indicator of domestic economic activity. The Dutch case is less extreme, but the principle is the same. An economy in which one firm can explain 60 percent of quarterly growth needs a measure that tells citizens and policymakers what the rest of the economy is doing.

10.3 For wage policy: use the right benchmark

As argued in Section 6.3, the social partners should use an ex-ASML benchmark when they refer to aggregate productivity and profitability in wage negotiations. The CPB, which provides the central projections used in Dutch wage bargaining, could publish its productivity and profitability projections both including and excluding the semiconductor cluster.

10.4 For industrial policy: manage concentration consciously

The government's Beethoven programme, which commits around 2.5 billion euros of public investment to infrastructure, housing and education around Eindhoven to anchor ASML in the Netherlands, is a rational response to a real threat: the company's repeated warnings that it could expand abroad if Dutch conditions deteriorate. But the policy has a macroeconomic dimension that is rarely acknowledged. Each euro invested to anchor ASML increases the share of Dutch growth that depends on one supply chain, one technology and one geopolitical relationship. This is not an argument against the policy. It is an argument for pairing it with deliberate efforts to diversify, for instance by supporting the growth of the broader Brainport supplier base into adjacent markets such as medical technology, photonics, defence electronics and precision manufacturing for the energy transition, so that the regional ecosystem is less dependent on a single customer.

10.5 For foreign economic policy: price the China risk

The debate about export controls on lithography equipment is usually framed as a question of national security and alliance management. It is also a macroeconomic question. Our scenarios suggest that a sudden loss of ASML's China business, equivalent to about one fifth of its sales, would subtract about 0.35 percentage points from Dutch GDP growth directly, and about half a percentage point once supplier effects are included, in the year it occurs, before any compensation from other markets. That is a cost that Dutch negotiators should quantify and put on the table, both in The Hague and in Washington, rather than treat as an afterthought. Alignment with the United States on technology security is a defensible strategic choice for the Netherlands; it is more defensible still when its economic price is measured and planned for.

10.6 For financial stability: watch the wealth channel

ASML is the largest company on the Amsterdam stock exchange and one of the largest holdings of Dutch pension funds and retail investors. A cyclical downturn in the company's results would therefore reach Dutch households not only through jobs and suppliers but also through the value of their savings and pension assets. The Dutch pension system is in the middle of its transition to the new contract regime, under which the value of individual pension entitlements will respond more directly to asset prices. In that system, a sharp fall in the valuation of a single heavily held stock would be visible to members far more quickly than under the old defined-benefit framework. The Dutch central bank (DNB), in its financial stability reporting, could usefully assess the concentration of Dutch household and pension wealth in the semiconductor sector, alongside its existing work on interest-rate and real-estate exposures.

10.7 For the European dimension

Finally, the Dutch case has implications beyond the Netherlands. ASML is one of the few European firms with the global market power that the European Commission's competitiveness agenda seeks to foster. Its growth contributes to euro-area growth, its exports support the euro-area trade surplus, and its position is a central asset in Europe's negotiations with both Washington and Beijing on technology controls. A downturn in ASML would be a Dutch macroeconomic event, but it would also be a European strategic one. The Netherlands has an interest in ensuring that the European debate on economic security takes explicit account of the macroeconomic concentration that comes with strategic champions.

11. Conclusion

The Dutch economy in 2026 is two economies. One is a world-class high-technology cluster around Eindhoven, growing at more than 30 percent a year on the back of the global investment boom in artificial intelligence, earning record margins and exporting machines that no other company on earth can make. The other is everything else: a mature, open, service-heavy economy growing at about 1.4 to 1.5 percent a year, with modest per capita gains, a tight but cooling labour market, constrained housing and electricity networks, and an industrial base under pressure from energy costs. The headline growth rate averages the two. It describes neither.

Our counterfactual series shows that the acceleration of Dutch growth in 2026 is almost entirely attributable to ASML, that the firm explains around 60 percent of growth in the third quarter and close to 30 percent of growth over the year, and that the underlying economy has not accelerated at all. Our stress tests show that the Netherlands has imported a significant and asymmetric risk into its national growth rate, and that 2027 carries a built-in statistical drag even in favourable scenarios.

None of this diminishes the achievement that ASML represents. The company is the most valuable technology firm in Europe and one of the very few European firms that sit at a true global chokepoint. But a nation that depends on a champion must learn to read its own statistics with the champion removed. For the Netherlands, that is now a macroeconomic necessity, not an academic exercise.

About this series. This paper is the first in a ten-part series on the Dutch economy in 2026 published by Global Strategy Files. Forthcoming papers examine gas storage and the Groningen option, the pension reform and mortgage rates, grid congestion, the sale of rental homes, the Rotterdam and Chemelot chemical cluster, the domestic multiplier of defence spending, the Beveridge curve, the Netherlands between Washington and Beijing, and the Dutch inflation premium.

Methodological Appendix

A.1 Equations

Let G(t) be the CBS volume index of GDP in quarter t (2021 equal to 100), Y21 the nominal GDP of 2021 (870.6 billion euros), S(t) ASML's total net sales in quarter t in billions of euros, r the value-added ratio and P(t) the GDP deflator (2021 equal to 1), interpolated from annual values centred on mid-year.

Real GDP, annualised, in 2021 prices: Y(t) = G(t) / 100 x Y21.

ASML real Dutch value added, annualised: A(t) = 4 x S(t) x r / P(t).

GDP excluding ASML: X(t) = Y(t) minus A(t).

ASML share of GDP: s(t) = A(t) / Y(t).

ASML contribution to year-on-year growth, in percentage points: c(t) = [A(t) minus A(t minus 4)] / Y(t minus 4) x 100.

In the smoothed specification, S(t) is replaced by the average of S(t) and S(t minus 1).

A.2 Calibration of r

With r equal to 0.45, ASML's estimated share of Dutch GDP is 1.25 percent on average in 2025 and 1.63 percent in 2026 (including the fourth-quarter projection), compared with ING estimates of 1.3 and 1.6 percent respectively.

A.3 Scenario and simulation settings

Underlying non-ASML growth in 2027: 1.3 percent (scenarios); normal with mean 1.3 and standard deviation 0.6 (simulation). Deflator growth in 2027: 2.2 percent. Supplier and investment spillover factor: 1.375. ASML 2026 sales: 44.45 billion euros. Historical ASML annual sales, 2000 to 2025, from the company's annual reports; bootstrap of 25 annual log growth rates, 200,000 draws, fixed random seed.

A.4 Quarterly data

Table A1. Quarterly series, base specification

Quarter CBS GDP index ASML sales (EUR m) ASML share (%) Headline y/y (%) Ex-ASML y/y (%) ASML contrib. (pp)
2022 Q3 105.7 5,778 1.04
2022 Q4 105.5 6,430 1.14
2023 Q1 105.0 6,746 1.18
2023 Q2 104.6 6,902 1.19
2023 Q3 103.8 6,673 1.14 -1.80 -1.90 0.08
2023 Q4 104.1 7,237 1.22 -1.33 -1.41 0.07
2024 Q1 104.5 5,290 0.88 -0.48 -0.17 -0.31
2024 Q2 105.3 6,243 1.02 0.67 0.85 -0.17
2024 Q3 105.9 7,467 1.19 2.02 1.97 0.08
2024 Q4 106.3 9,263 1.46 2.11 1.86 0.27
2025 Q1 106.6 7,742 1.21 2.01 1.67 0.36
2025 Q2 106.9 7,692 1.19 1.52 1.34 0.19
2025 Q3 107.3 7,516 1.15 1.32 1.37 -0.03
2025 Q4 107.8 9,717 1.47 1.41 1.41 0.02
2026 Q1 108.1 8,767 1.31 1.41 1.30 0.12
2026 Q2 108.9 9,326 1.37 1.87 1.68 0.21
2026 Q3 109.5 11,662 1.70 2.05 1.48 0.59

Source: CBS, ASML, author's model.

A.5 A note on the CBS year-on-year figures

Year-on-year growth rates computed from the published, rounded, seasonally adjusted index can differ by one or two tenths from the growth rates CBS publishes, which are calculated from unrounded and, for annual comparisons, unadjusted data. For example, the index implies a fall of 1.8 percent in the year to the third quarter of 2023, against 1.5 percent in the published series. The differences are small relative to the effects measured in this paper and do not affect the conclusions.

References

Statistics Netherlands (CBS), "Dutch economy grew by 0.6 percent in Q3 2026", first estimate, 30 October 2026, and "Dutch economy grew by 0.4 percent in Q2 2026", 30 July 2026.

ASML Holding NV, quarterly results releases, 2022 to 2026, including "ASML reports 11.7 billion euros total net sales and 5.0 billion euros net income in Q3 2026", 14 October 2026; annual reports, 2000 to 2025.

ING Research, estimates of ASML's contribution to Dutch GDP, as reported in October 2026.

CPB Netherlands Bureau for Economic Policy Analysis, Macro Economic Outlook 2027 (September 2026) and Central Economic Plan 2025 key figures.

Ministry of Finance, "Het Blauwe Boekje 2025-2026".

ETLA, The Research Institute of the Finnish Economy, studies on Nokia's role in the Finnish economy (2000, 2010, 2012).

Gabaix, X. (2011), "The Granular Origins of Aggregate Fluctuations", Econometrica 79(3).

South China Morning Post and company investor communications on ASML's China sales, October 2026.

Driss EFFINA, PhD
Independent Researcher, Global Strategy Files

Free preview
The Q4 2026 Global Outlook

The opening pages of our four country reports: United States, United Kingdom, Japan and China. A free PDF, sent to your inbox.

You will also receive occasional analysis from Global Strategy Files. Unsubscribe in one click, at any time.