## _cr08172 - Appendix I

## Source details

**Canonical URL:** [_cr08172 - Appendix I](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2008/_cr08172.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2008/_cr08172.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2008/_cr08172.pdf.json)

---

### A. Introduction and key findings
- Overall competitiveness: "Overall competitiveness of the Dutch economy seems adequate, but domestically produced exports have lost market share recently."
- External indicators: "large current account surplus, robust exports, and strong growth."
- Real exchange rates and unit labor costs:
  - "After the sizable real appreciation in 2001-03, REER measures based on different price indices have been relatively stable, although ULC-based measures suggest a growing gap with some competitors."
  - "Manufacturing unit labor costs have fallen for the last four years, but at a lower pace than for some trading partners."
- Market-share and exposure:
  - Domestic exports’ growth aligned with adjusted market growth until 2000; thereafter Dutch market share declined.
  - Foreign trade represents "about 80 percent of GDP", making market-share loss a concern.
- Stylized summary conclusions (verbatim list):
  - "(1) Dutch exports trend has been somewhat below Germany’s in the 2000s."
  - "(2) Dutch traditional lead over Germany in exports of manufactured goods, machinery and transport equipment was lost in the 2000s."
  - "(3) The Netherlands is relatively more exposed to supply-driven shocks while Germany is more exposed to demand-driven shocks."
  - "(4) Following an increase in unit labor costs in manufacturing (ULCM), the adjustment in the Netherlands is comparatively less flexible than in Germany. The Netherlands adjusts relatively less via price and wage changes, and more via employment changes."
  - "(5) The same features are also evident when the two countries are faced with an upward, supply-driven, terms of trade (TOT) shock."
  - "(6) The Netherlands profits relatively less from a demand-driven increase in the TOT, but accordingly, suffers a relatively lower output fall following a demand-driven ULCM increase."

### B. Data, sample, and stationarity treatment
- Panels and series:
  - Panel 1: "396 quarterly macroeconomic series and 106 series of trade by country (for a total of series N = 502)."
  - Panel 2: "396 quarterly macroeconomic series and 110 series of trade by SITC, Revision 3, category of products (for a total of series N = 506)."
- Sample period: "1981:Q1–2006:Q4 (i.e., T = 104)."
- Countries: "France, Germany, Japan, the Netherlands, the United Kingdom, and the United States."
- Additional variables: "crude oil prices, a commodity industrial inputs price index, world demand, and world reserves."
- Treatment of non‑stationarity:
  - Seasonally adjusted series.
  - "the Corbae-Ouliaris Ideal Band-Pass Filter was used to make the series stationary (Corbae and Ouliaris, 2006)."
  - Rationale: low power of unit root tests; first differencing removes significant variance; the filter "is consistent, is not subject to end-point problems, and has no finite sampling error."

### C. Stylized facts on export behavior
- Cyclical behavior:
  - "Dutch and German exports display a similar cyclical behavior by destination and by product, although Germany’s is more volatile."
  - "In general, the Dutch export cyclical component is less volatile than Germany’s."
- Trend behavior since 2000:
  - "Starting in 2003-04, the Netherlands lost its trade export growth advantage over Germany vis-à-vis the euro area, the EU, Japan, and the UK."
  - The Netherlands gained from eastward EU expansion (starting in 2000) and Chinese rapid growth after 2003, but lost its growth advantage over Germany in several product groups:
    - Manufactured goods (from 2000).
    - Machinery and transport equipment, miscellaneous manufactured goods, and crude materials except fuels (from 2003).
  - Exception: "animal and vegetable oils, fats and waxes" preserved relative advantage.
- Descriptive conclusion: "there has been since 2000-03 an underperformance of Dutch exports relative to its own past and relative to Germany."

### D. Dynamic behavior: model, identification, and shock interpretation
- Model framework:
  - "large dimensional approximate dynamic factor model in the tradition of Stock and Watson (2002)."
  - Series decomposed into common and idiosyncratic components; static factor model treats lagged/dynamic factors as additional static factors.
  - Number of factors selected using Bai and Ng (2002) criteria: four factors were chosen.
- Identification and estimation:
  - Identification scheme (following Eickmeier, 2007 and sign-restriction literature Peersman (2005), Uhlig (2003)):
    1. maximize explained variance of forecast error of chosen variable and calculate impulse-response functions;
    2. assume identified shocks are linearly correlated to a vector of fundamentals;
    3. identify orthogonal shocks by rotation using a sign-identification strategy imposing inequality restrictions on impulse responses based on aggregate demand and supply framework.
  - Targeted variables: common components of the Netherlands’s and Germany’s ULCM and TOT.
  - Horizons: impulse-response functions for first five years; maximization over 0 to 19 (five years, 20 quarters).
  - Only two structural shocks identified for each variable of interest (ULCM and TOT): a supply shock and a demand shock.
  - Bootstrap with 500 draws used to mitigate small-sample bias; median response and a 90 percent bootstrapped confidence band are estimated.
- Shock definitions and economic interpretation:
  - Increase in ULCM can reflect a fall in labor productivity → interpreted as a supply shock; or an increase in labor compensation → interpreted as a demand shock.
  - Rise in TOT may reflect deterioration of competitiveness (persistent shock) or strong world demand (transient shock).
  - Sign restrictions imposed contemporaneously and during the first year after the shock (explicit restrictions preserved below in identification subsection).

### E. Sign restrictions and identification details
- Sign restrictions (contemporaneous and during first year) used to distinguish supply vs demand shocks:
  - ULCM: supply shock ≥ 0; demand shock ≤ 0
  - Output: supply shock ≤ 0; demand shock ≥ 0
  - Real wages: supply shock ≤ 0; demand shock ≥ 0
  - Terms of trade: supply shock ≥ 0; demand shock ≥ 0
  - Consumption: supply shock ≥ 0; demand shock ≤ 0
  - Current account: supply shock ≤ 0; demand shock ≥ 0
- Rotation and Monte Carlo:
  - Orthogonal shocks are rotated using a 2×2 orthogonal matrix P; grid of 30-degree increments producing 12 angles between 0 and π is used.
  - Only rotations producing plausible impulse responses and variance decompositions consistent with the sign restrictions are accepted.

### F. Key econometric findings and variance decompositions
- Two identified shocks explain "99 percent of the error variance of the common components of Dutch and German ULCM and TOT over 20 quarters."
- Variance shares of common components:
  - ULCM variance shares: Netherlands: "about 68 percent"; Germany: "about 73 percent."
  - TOT variance shares: Netherlands: "up to 20 percent"; Germany: "up to 43 percent."
  - Interpretation: "Dutch TOT are more heavily influenced by idiosyncratic factors than Germany’s; TOT are relatively less significant channels of shock transmission."
- Selected forecast-error variance highlights (forecast horizon 20 quarters; bootstrap 500 draws):
  - Table I-3 (Netherlands, ULCM shocks) selected entries:
    - GDP: Supply 0.87; Demand 0.49
    - Productivity: Supply 0.74; Demand 0.37
    - Unit labor cost of the manufacturing sector: Supply 0.68; Demand 0.66
    - Exports total: Supply 0.84; Demand 0.36
    - EXP SITC 3 (Mineral fuels, lubricants and related materials): Supply 0.50; Demand 0.09
    - EXP SITC Total: Supply 0.91; Demand 0.24
  - Table I-4 (Germany, ULCM shocks) selected entries:
    - GDP: Supply 0.70; Demand 0.13
    - Private investment: Supply 0.93; Demand 0.05
    - Unit labor cost of the manufacturing sector: Supply 0.74; Demand 0.34
    - Exports total: Supply 0.69; Demand 0.15
    - EXP SITC Total: Supply 0.92; Demand 0.37
  - Table I-5 (Netherlands, TOT shocks) selected entries:
    - GDP: Supply 0.70; Demand 0.46
    - Consumer prices: Supply 0.92; Demand 0.04
    - Terms of trade: Supply 0.42; Demand 0.11
    - Exports total: Supply 0.69; Demand 0.54
    - EXP SITC Total: Supply 0.92; Demand 0.05
  - Table I-6 (Germany, TOT shocks) selected entries:
    - GDP: Supply 0.87; Demand 0.39
    - Consumer prices: Supply 0.92; Demand 0.05
    - Terms of trade: Supply 0.20; Demand 0.62
    - EXP SITC Total: Supply 0.91; Demand 0.08

### G. Comparative exposure and dynamic responses (Netherlands vs Germany)
- Main comparative result: "Netherlands is relatively more exposed to supply-driven shocks; Germany is more exposed to demand-driven shocks."
- Responses to a supply-driven (productivity) increase in ULCM:
  - Dutch output falls driven by declines in consumption, investment and exports; fall is larger in the Netherlands than in Germany.
  - Larger appreciation of the real exchange rate and a milder fall in the CPI in the Netherlands than in Germany.
  - Real compensation of employees: stable in the Netherlands, falls in Germany.
  - Dutch total employment falls for a longer period than German employment.
  - Dollar value of exports to all destinations increases more in Germany than in the Netherlands; value of Dutch exports to China falls.
  - Euro value of exports per product: Dutch falls more markedly for "mineral fuels and lubricants, and chemicals and related products."
- Flexibility and adjustment:
  - Following an increase in ULCM in manufacturing (ULCMM), "the Netherlands displays less flexibility to adjust than Germany":
    - Adjusts relatively less via price and wage changes, and more via employment and output changes.
    - Evidence of relatively larger downward wage rigidity in the Netherlands.
  - Similar features observed for upward, supply-driven, TOT shocks.
- Responses to a demand-driven increase in TOT:
  - "The Netherlands profits relatively less from demand-driven TOT increases."
  - Result: relatively lower fall in output following a demand-driven increase in ULCM for the Netherlands.
  - Larger appreciation of the real exchange rate and smaller fall in the CPI in the Netherlands than in Germany.
  - Real compensation to employees falls in Germany but increases in the Netherlands despite falling labor productivity.
  - Total employment increases relatively less in the Netherlands.
  - Dollar and euro export values: increases for the Netherlands except exports to China; increases are larger for Germany across many product categories. Dutch exports of fuels and lubricants fall; German exports are flat.

### H. Policy implications on product, labor markets and competitiveness
- "It is difficult to overestimate the relevance of product and factor markets flexibility for the open Dutch economy."
- Policy recommendations (explicit and implied):
  - Increase flexibility in labor, product and service markets to cope with foreign and domestic shocks in an increasingly globalized economy.
  - Boost productivity via research and development.
  - Reduce relatively high Employment Protection Legislation (EPL).
  - Raise labor market participation.
- Overall assessment: "Export performance relative to Germany has deteriorated somewhat in recent years, but the short answer to whether the Netherlands is suffering from a competitiveness problem is: no."
- Additional conclusion: "the Netherlands adapts less quickly to upward output pressures due to strong world demand; it is relatively less flexible to adjust than Germany and must undergo larger quantity adjustments because prices and wages appear more rigid downward."

### I. Trend exports — regional and product group averages (selected exact figures)
- Table I-1: Trend Exports per Region (Average annual percent change) — Netherlands (five period columns preserved in order)
  - Netherlands to EU: 2.0 1.8 1.4 2.8 2.2
  - Netherlands to Asia: 2.4 2.7 2.1 2.4 2.9
  - Netherlands to Japan: 2.3 3.9 1.8 0.9 0.6
  - Netherlands to China: 3.9 3.5 3.8 6.1 6.9
  - Netherlands to Euro: 1.9 1.8 1.3 2.8 2.2
  - Netherlands to Accession Countries: 3.9 2.4 4.5 5.3 5.5
  - Netherlands to United States: 2.4 2.8 1.5 3.2 3.4
  - Netherlands to United Kingdom: 2.0 2.3 1.5 2.4 1.7
  - Netherlands to ROW: 1.5 0.4 1.0 3.5 3.3
- Table I-1: Trend Exports per Region — Germany (preserved)
  - Germany to EU: 1.9 2.5 0.9 2.2 2.4
  - Germany to Asia: 2.3 2.6 1.1 2.7 4.2
  - Germany to Japan: 2.2 3.9 0.3 1.5 2.1
  - Germany to China: 3.7 3.2 2.5 5.3 6.3
  - Germany to Euro: 1.8 2.5 0.6 2.1 2.2
  - Germany to Accession Countries: 3.2 2.1 4.4 3.0 3.1
  - Germany to United States: 2.3 2.8 1.9 2.0 2.9
  - Germany to United Kingdom: 2.0 3.0 1.2 1.9 2.0
  - Germany to ROW: 1.6 1.3 0.6 3.2 3.9
- Table I-2: Trend Exports per Product SITC — Netherlands (1980-2006; 1980-1989; 1990-1999; 2000-2006; 2003-2006)
  - Total: 1.8 1.7 1.6 2.3 2.3
  - Food and live animal - SITC 0: 1.3 1.6 0.8 1.7 1.7
  - Beverages and tobacco - SITC 1: 2.0 2.0 2.2 1.2 1.7
  - Crude materials, inedible, except fuels - SITC 2: 2.0 2.4 1.2 2.6 2.7
  - Mineral fuels, lubricants and related materials - SITC 3: 1.1 -1.0 0.8 4.3 4.0
  - Animal and vegetable oils, fats and waxes - SITC 4: 1.5 0.8 1.8 2.0 3.3
  - Chemicals and related products - SITC 5: 1.9 2.1 1.4 2.5 2.4
  - Manufactured goods - SITC 6: 1.6 2.1 0.9 1.8 2.3
  - Machinery and transport equipment - SITC 7: 2.5 2.7 2.4 2.0 1.7
  - Miscellaneous manufactured articles - SITC 8: 2.3 3.1 1.9 1.9 1.7
  - Commodities and transactions - SITC 9: 0.9 -1.1 1.3 2.9 3.7
- Table I-2: Trend Exports per Product SITC — Germany (preserved)
  - Total: 1.9 2.3 1.0 2.5 2.8
  - Food and live animal - SITC 0: 1.5 1.8 0.7 2.1 1.9
  - Beverages and tobacco - SITC 1: 2.1 2.3 1.4 3.2 3.6
  - Crude materials, inedible, except fuels - SITC 2: 1.8 2.1 0.5 2.8 3.4
  - Mineral fuels, lubricants and related materials - SITC 3: 1.3 -1.3 0.6 4.7 4.7
  - Animal and vegetable oils, fats and waxes - SITC 4: 0.7 0.5 1.1 0.7 1.7
  - Chemicals and related products - SITC 5: 2.0 2.3 1.1 2.8 3.1
  - Manufactured goods - SITC 6: 1.6 2.0 0.6 2.4 2.9
  - Machinery and transport equipment - SITC 7: 2.0 2.5 1.1 2.4 2.5
  - Miscellaneous manufactured articles - SITC 8: 2.0 2.9 0.7 2.5 2.6
  - Commodities and transactions - SITC 9: 3.0 1.3 3.5 2.9 4.5

### J. Corporate income tax (CIT) interaction, rate-revenue puzzle, and fiscal projections
- Historical patterns and interaction:
  - Statutory CIT rates in western Europe fell from "about 50 percent in 1982 to about 30-35 percent in 2006."
  - Empirical estimates imply strong strategic interaction:
    - Devereux, Lockwood, and Redoano (2004): a 1 percentage point reduction in average statutory CIT rates of other countries reduces home country rate by 0.7–0.8 percentage points.
    - Redoano (2007): parameter estimates for neighboring countries are "positive and significantly above 1" for western Europe.
  - Implication for the Netherlands: a 1 percentage point reduction in neighboring western European countries tends to reduce the Netherlands rate by about 0.8–1.0 percentage point.
- The CIT rate–revenue puzzle (Netherlands):
  - Observed: statutory rate declined strongly, yet CIT revenues (share of GDP and of total tax revenue) have been volatile but trending up since early 1980s.
  - Legal base broadening evidence: effective CIT rates at all profitability levels declined in line with statutory rates, implying limited legal base broadening.
  - Economic base broadening evidence:
    - Pretax corporate profits to GDP expanded from "about 11 percent of GDP in 1979 to 16 percent in 2006."
    - Nonfinancial corporate investment share to GDP declined; financial sector investment did not compensate.
  - Income tax base shifting:
    - Applying De Mooij and Nicodème (2008) methodology to the Netherlands: income tax base shifting accounts on average for "about 35 percent of CIT revenue."
    - Implication: an ex-ante reduction in the CIT rate equivalent to one euro costs on average only 65 euro cents in CIT revenue lost ex-post; 35 euro cents regained through income shifting (eroding PIT revenue).
  - International profit shifting and FDI:
    - De Mooij (2005) estimates: with FDI elasticity -2.4, 1 percentage point rate reduction broadens economic base by 0.8 percent via FDI inflows (about 25 percent of ex-ante loss regained); with profit shifting elasticity -2.0, 30 percent of ex-ante loss regained via profit shifting.
    - Empirical net FDI flows show outflows from the Netherlands have outpaced inflows; net FDI outflows increased in the last decade.
  - Corporate-tax features attracting profit shifting: dividend and capital gains exemptions on foreign shares; no withholding taxes on interest; recent abolition of capital duty and reduced dividend withholding tax.
- Fiscal scenarios and staff estimates:
  - Context: Netherlands CIT rate declined from 35 percent to 25.5 percent over the last decade; average NMS rates fell from 32 percent in 1995 to below 18 percent in 2007.
  - Strategic interaction suggests possible decline of Netherlands CIT rate to "less than 20 percent in the medium term."
  - Staff fiscal scenarios (CIT-to-GDP ratio, percent):
    - 2006:
      - CIT ratio (unadjusted): 3.7
      - Gain from international profit shifting: 0.6
      - CIT ratio (adjusted for profit shifting): 3.1
      - Gain from income shifting: 0.6
      - CIT ratio (adjusted for income and profit shifting): 2.5
    - Medium-Term:
      - CIT ratio (unadjusted): 3.2
      - Gain from international profit shifting: 0.6
      - CIT ratio (adjusted for profit shifting): 2.6
      - Gain from income shifting: 0.6
      - CIT ratio (adjusted for income and profit shifting): 2.0
  - Staff summary: "With unchanged tax base, reduced CIT rate will lower the ratio of CIT revenues to GDP from 3.7 percent in 2006 to about 2 percent of GDP in the medium term." Allowing for base-broadening could limit the fall to "about 3 percent of GDP" (text phrasing retained).
- Factors that can reduce revenue offsets:
  - CCCTB (consolidated corporate tax base) and CIT harmonization would limit profit shifting and reduce revenue offsets.
  - If tax differentials with NMS persist, Netherlands could lose revenue as multinationals relocate profit or investment to NMS.
  - Size of potential revenue loss from profit shifting and relocation is "difficult to assess."

### K. Policy options and fiscal strategy in face of tax competition and aging
- Formal CIT rate harmonization:
  - "at present, this is not a realistic option."
  - Gains from harmonization likely small; member states unlikely to cede residual tax sovereignty.
- Domestic policy constraints and aging:
  - Demographic aging implies shrinking working-age population; productivity must become predominant growth source.
  - Raising PIT or social payroll contributions is "not the optimal strategy" given high labor tax burden.
- Recommended fiscal direction:
  - "The best strategy to address the potential decline in revenue as a share of GDP in the medium term is to broaden the tax base."
  - Shift away from mobile tax bases toward greater reliance on consumption and property taxes.
  - Consumption taxes (VAT): "Increasing the VAT by 1–2 percentage points, assuming constant elasticities, would deliver revenues of about ½–1½ percent of GDP, enough to compensate for the expected decline in CIT revenues over the medium term."
  - Property taxation: scope exists to raise property taxation and relax preferential treatment of saving instruments; would improve neutrality toward different forms of wealth.
  - Environmental taxes: Netherlands already relies relatively more on environmental taxes; scope exists to tax products on pollution created by fabrication processes.
- Trade-offs emphasized:
  - Continued CIT rate reductions risk eroding CIT revenue unless offset by legal base broadening or expansion of economic base.
  - Income tax base shifting has materially offset rate-cut revenue losses, but erodes PIT revenue.
  - CCCTB and harmonization would reduce profit-shifting gains and change the calculus of rate cuts.

*Source: Appendix I and selected sections of IMF Country Report _cr08172.*

### Appendix      I ........................................................................................................

### _cr08172 - Appendix      I

### A. Introduction and Key Findings
- Overall competitiveness: "Overall competitiveness of the Dutch economy seems adequate, but domestically produced exports have lost market share recently."
- External indicators: "large current account surplus, robust exports, and strong growth."
- Real exchange rates: "After the sizable real appreciation in 2001-03, REER measures based on different price indices have been relatively stable, although ULC-based measures suggest a growing gap with some competitors."
- Unit labor costs: "Manufacturing unit labor costs have fallen for the last four years, but at a lower pace than for some trading partners."
- Market-share pattern: "domestically produced exports’ growth was in line with adjusted market growth until 2000. Afterward, the Dutch market share declined."
- Trade exposure: "foreign trade representing about 80 percent of GDP" makes the loss of market share a concern.
- Stylized summary conclusions:
  - "(1) Dutch exports trend has been somewhat below Germany’s in the 2000s."
  - "(2) Dutch traditional lead over Germany in exports of manufactured goods, machinery and transport equipment was lost in the 2000s."
  - "(3) The Netherlands is relatively more exposed to supply-driven shocks while Germany is more exposed to demand-driven shocks."
  - "(4) Following an increase in unit labor costs in manufacturing (ULCM), the adjustment in the Netherlands is comparatively less flexible than in Germany. The Netherlands adjusts relatively less via price and wage changes, and more via employment changes."
  - "(5) The same features are also evident when the two countries are faced with an upward, supply-driven, terms of trade (TOT) shock."
  - "(6) The Netherlands profits relatively less from a demand-driven increase in the TOT, but accordingly, suffers a relatively lower output fall following a demand-driven ULCM increase."

### B. Data and Treatment of Non‑Stationarity
- Data panels:
  - Panel 1: "396 quarterly macroeconomic series and 106 series of trade by country (for a total of series N = 502)."
  - Panel 2: "396 quarterly macroeconomic series and 110 series of trade by SITC, Revision 3, category of products (for a total of series N = 506)."
- Sample period: "1981:Q1–2006:Q4 (i.e., T = 104)."
- Countries included: "France, Germany, Japan, the Netherlands, the United Kingdom, and the United States."
- Additional global variables: "crude oil prices, a commodity industrial inputs price index, world demand, and world reserves."
- Seasonal adjustment: "Variables have been seasonally adjusted."
- Stationarity treatment: "the Corbae-Ouliaris Ideal Band-Pass Filter was used to make the series stationary (Corbae and Ouliaris, 2006)."
- Rationale for filter:
  - "available unit root tests have low power and often the decision on the degree of integration of the series has to be based on subjective judgment."
  - "first differencing removes a significant part of the variance of economic time series."
  - "the Corbae and Ouliaris filter is consistent, is not subject to end-point problems, and has no finite sampling error."

### C. Stylized Facts on Export Behavior
- Cyclical behavior:
  - "Dutch and German exports display a similar cyclical behavior by destination and by product, although Germany’s is more volatile."
  - "The business cycle in main trading partners correlates well with exports."
  - "In general, the Dutch export cyclical component is less volatile than Germany’s."
- Trend behavior since 2000:
  - "Starting in 2003-04, the Netherlands lost its trade export growth advantage over Germany vis-à-vis the euro area, the EU, Japan, and the UK."
  - "The Netherlands has taken advantage of the eastward expansion of the EU starting in 2000 and of Chinese rapid growth after 2003 (Table I-1)."
  - "The Netherlands has lost its traditional growth advantage over Germany in manufactured goods starting in 2000, and machinery and transport equipment, miscellaneous manufactured goods, and crude materials except fuels, starting in 2003 (Table I-2)."
  - "With the exception of animal and vegetable oils, fats and waxes, a similar pattern is present in all other SITC, Revision 3, categories."
- Overall descriptive conclusion: "there has been since 2000-03 an underperformance of Dutch exports relative to its own past and relative to Germany. The change in export performance is recent, but raises some questions about the future competitiveness of the Dutch economy."

### D. Dynamic Behavior: Model and Shock Identification
- Methodology: "large dimensional approximate dynamic factor model in the tradition of Stock and Watson (2002)."
- Model intuition: "A vector of time series can be represented as the sum of two latent components, a common component and an idiosyncratic component."
- Common vs idiosyncratic:
  - "The common component... is driven by few common shocks... effects of common shocks differ from one variable to another due to different factor loadings."
  - "The idiosyncratic component is driven by idiosyncratic shocks, which are specific to each variable."
- Static factor treatment: "The static factor model used here differs from the dynamic factor model in that it treats lagged or dynamic factors as additional static factors."
- Estimation and identification:
  - "estimating the common components and identifying a reduced number of structural shocks that explain the common components of the variables of interest."
  - Identification scheme features (following Eickmeier, 2007):
    1. "it maximizes the explained variance of the forecast error of the chosen variable and calculates impulse-response functions;"
    2. "it assumes that identified shocks are linearly correlated to a vector of fundamentals;"
    3. "it identifies orthogonal shocks by rotation using a sign-identification strategy that imposes inequality restrictions on the impulse-response functions of variables based on a typical aggregate demand and aggregate supply framework."
  - "Only those shocks that have a structural meaning are chosen."

### E. Figures and Tables Referenced (content indicators only)
- Tables listed include: I-1, I-2, I-3, I-4, I-5, I-6 (trend exports, forecast error variances).
- Figures listed include: I-1 (ULCM, the Netherlands), I-2 (ULCM, Germany), I-3 (TOT, the Netherlands), I-4 (TOT, Germany), and various trend export charts by destination and product.

*Appendix I of IMF Country Report _cr08172 (pages excerpted).*

### 16.      Given recent Dutch ULCM developments and that the Netherlands is the

### _cr08172 - 16.      Given recent Dutch ULCM developments and that the Netherlands is the

### Shock definitions and economic interpretation
- Increase in ULCM can reflect:
  - a fall in labor productivity → interpreted as a supply shock.
  - an increase in labor compensation → interpreted as a demand shock.
- Rise in the Terms of Trade (TOT) can reflect:
  - a deterioration of competitiveness (structural factors) → persistent shock leading to increased consumption and investment and a current account deficit.
  - strong world demand for the country’s products → transient shock where consumers largely save the windfall and the current account moves into surplus.
- Sign restrictions for shock identification were imposed contemporaneously and during the first year after the shock.

### Estimation procedure
- Series assumed to follow an approximate dynamic factor model.
- Number of factors selected using Bai’s and Ng’s (2002) criteria: four factors were chosen.
- Structural shocks identified via structural VAR literature approach with sign restrictions on impulse responses following Peersman (2005) and Uhlig (2003).
- Identification targeted the common components of the Netherlands’s and Germany’s ULCM and TOT to find shocks that maximize explanation of forecast error variance over 20 periods.
- Only two structural shocks were identified for each variable of interest (ULCM and TOT): a supply shock and a demand shock.
- Bootstrap used to mitigate small-sample bias; impulse-response functions calculated for the first five years.
- Both the median response and a 90 percent bootstrapped confidence band are estimated.

### Key econometric findings
- The two identified shocks explain 99 percent of the error variance of the common components of Dutch and German ULCM and TOT over 20 quarters.
- Variance shares of ULCM common components:
  - Netherlands: about 68 percent.
  - Germany: about 73 percent.
- Variance shares of TOT common components:
  - Netherlands: up to 20 percent.
  - Germany: up to 43 percent.
  - Interpretation: Dutch TOT are more heavily influenced by idiosyncratic factors than Germany’s; TOT are relatively less significant channels of shock transmission.

### Comparative exposure and dynamics
- Main result: Netherlands is relatively more exposed to supply-driven shocks; Germany is more exposed to demand-driven shocks.
- Supply shocks to ULCM are relatively more important than demand shocks; for TOT shocks the opposite holds.
- Supply-driven (productivity) increase in ULCM:
  - Dutch output falls driven by declines in consumption, investment and exports; the fall is larger in the Netherlands than in Germany.
  - Larger appreciation of the real exchange rate and a milder fall in the CPI in the Netherlands than in Germany.
  - Real compensation of employees: stable in the Netherlands, falls in Germany.
  - Dutch total employment falls for a longer period than German employment.
  - Dollar value of exports to all destinations increases more in Germany than in the Netherlands; value of Dutch exports to China falls.
  - Euro value of exports per product: Dutch falls more markedly for mineral fuels and lubricants, and chemicals and related products.
- Following an increase in ULCM in manufacturing (ULCMM), the Netherlands displays less flexibility to adjust than Germany:
  - Adjusts relatively less via price and wage changes, and more via employment and output changes.
  - Evidence of relatively larger downward wage rigidity in the Netherlands.
  - Same features observed for upward, supply-driven, TOT shocks.
- Demand-driven increase in TOT:
  - The Netherlands profits relatively less from demand-driven TOT increases.
  - Result: relatively lower fall in output following a demand-driven increase in ULCM for the Netherlands.
  - Larger appreciation of the real exchange rate and smaller fall in the CPI in the Netherlands than in Germany.
  - Real compensation to employees falls in Germany but increases in the Netherlands despite falling labor productivity.
  - Total employment increases relatively less in the Netherlands.
  - Dollar and euro export values: increases for the Netherlands except exports to China; increases are larger for Germany across many product categories (animal and vegetable oils and fats, chemicals and related products, manufactured goods, machinery and transport equipment). Dutch exports of fuels and lubricants fall; German exports are flat.

### Policy implications and conclusions
- The Dutch economy is significantly affected by economic activity in the rest of the world.
- Export performance relative to Germany has deteriorated somewhat in recent years, but the short answer to whether the Netherlands is suffering from a competitiveness problem is: no.
- Trend export growth:
  - Dutch trend overall export growth is somewhat below Germany’s in the 2000s.
  - The Netherlands preserved relative advantage vis-à-vis Germany in animal and vegetable oils, fats and waxes.
  - The Netherlands lost its traditional lead over Germany in exports of manufactured goods, machinery and transport equipment.
- Policy recommendation (implied): prevent erosion of the competitive position by increasing flexibility in labor, product and service markets to cope with foreign and domestic shocks in an increasingly globalized economy.
- Overall: the Netherlands adapts less quickly to upward output pressures due to strong world demand; it is relatively less flexible to adjust than Germany and must undergo larger quantity adjustments because prices and wages appear more rigid downward.

*Source: _cr08172 - 16.      Given recent Dutch ULCM developments and that the Netherlands is the*

### 28.      It is difficult to overestimate the relevance of product and factor markets

### 28.      It is difficult to overestimate the relevance of product and factor markets

### Summary finding: relevance of product and factor markets flexibility
- "It is difficult to overestimate the relevance of product and factor markets flexibility for the open Dutch economy."
- Importance noted for:
  - trade flows and relative price changes in the international transmission of disturbances;
  - policy constraints imposed by the euro area.
- Main conclusion: "The Netherlands will benefit from further structural reforms that increase good, service, and labor markets flexibility."

### Policy recommendations (explicit)
- Boost productivity via research and development.
- Reduce its relatively high EPL.
- Raise labor market participation.

### Trend exports — regional averages (selected figures preserved exactly)
- Table I-1: Trend Exports per Region (Average annual percent change)
  - Netherlands to EU: 2.0 1.8 1.4 2.8 2.2
  - Netherlands to Asia: 2.4 2.7 2.1 2.4 2.9
  - Netherlands to Japan: 2.3 3.9 1.8 0.9 0.6
  - Netherlands to China: 3.9 3.5 3.8 6.1 6.9
  - Netherlands to Euro: 1.9 1.8 1.3 2.8 2.2
  - Netherlands to Accession Countries: 3.9 2.4 4.5 5.3 5.5
  - Netherlands to United States: 2.4 2.8 1.5 3.2 3.4
  - Netherlands to United Kingdom: 2.0 2.3 1.5 2.4 1.7
  - Netherlands to ROW: 1.5 0.4 1.0 3.5 3.3
  - Germany to EU: 1.9 2.5 0.9 2.2 2.4
  - Germany to Asia: 2.3 2.6 1.1 2.7 4.2
  - Germany to Japan: 2.2 3.9 0.3 1.5 2.1
  - Germany to China: 3.7 3.2 2.5 5.3 6.3
  - Germany to Euro: 1.8 2.5 0.6 2.1 2.2
  - Germany to Accession Countries: 3.2 2.1 4.4 3.0 3.1
  - Germany to United States: 2.3 2.8 1.9 2.0 2.9
  - Germany to United Kingdom: 2.0 3.0 1.2 1.9 2.0
  - Germany to ROW: 1.6 1.3 0.6 3.2 3.9
- Note: "Numbers in bold indicate a higher growth rate of Dutch trend exports." (bolding not reproduced here; values preserved)

### Trend exports — product groups (selected figures preserved exactly)
- Table I-2: Trend Exports per Product SITC (Average annual percent change)
  - Netherlands (1980-2006; 1980-1989; 1990-1999; 2000-2006; 2003-2006):
    - Total: 1.8 1.7 1.6 2.3 2.3
    - Food and live animal - SITC 0: 1.3 1.6 0.8 1.7 1.7
    - Beverages and tobacco - SITC 1: 2.0 2.0 2.2 1.2 1.7
    - Crude materials, inedible, except fuels - SITC 2: 2.0 2.4 1.2 2.6 2.7
    - Mineral fuels, lubricants and related materials - SITC 3: 1.1 -1.0 0.8 4.3 4.0
    - Animal and vegetable oils, fats and waxes - SITC 4: 1.5 0.8 1.8 2.0 3.3
    - Chemicals and related products - SITC 5: 1.9 2.1 1.4 2.5 2.4
    - Manufactured goods - SITC 6: 1.6 2.1 0.9 1.8 2.3
    - Machinery and transport equipment - SITC 7: 2.5 2.7 2.4 2.0 1.7
    - Miscellaneous manufactured articles - SITC 8: 2.3 3.1 1.9 1.9 1.7
    - Commodities and transactions - SITC 9: 0.9 -1.1 1.3 2.9 3.7
  - Germany (1980-2006 etc.; values preserved):
    - Total: 1.9 2.3 1.0 2.5 2.8
    - Food and live animal - SITC 0: 1.5 1.8 0.7 2.1 1.9
    - Beverages and tobacco - SITC 1: 2.1 2.3 1.4 3.2 3.6
    - Crude materials, inedible, except fuels - SITC 2: 1.8 2.1 0.5 2.8 3.4
    - Mineral fuels, lubricants and related materials - SITC 3: 1.3 -1.3 0.6 4.7 4.7
    - Animal and vegetable oils, fats and waxes - SITC 4: 0.7 0.5 1.1 0.7 1.7
    - Chemicals and related products - SITC 5: 2.0 2.3 1.1 2.8 3.1
    - Manufactured goods - SITC 6: 1.6 2.0 0.6 2.4 2.9
    - Machinery and transport equipment - SITC 7: 2.0 2.5 1.1 2.4 2.5
    - Miscellaneous manufactured articles - SITC 8: 2.0 2.9 0.7 2.5 2.6
    - Commodities and transactions - SITC 9: 3.0 1.3 3.5 2.9 4.5
- Note: "Numbers in bold indicate a higher growth rate of Dutch trend exports." (bolding not reproduced here; values preserved)

### Forecast error variance decompositions — selected highlights (values preserved exactly)
- Table I-3: Forecast Error Variance of the Common Components of Netherlands Variables Explained by the Supply and Demand Shock to Unit Labor Costs in Manufacturing, 1981-2006 (forecast horizon 20 quarters; bootstrap 500 draws)
  - GDP: Supply 0.87; Demand 0.49; Lower Bound 0.02; Upper Bound 0.77; Shock Lower Bound 0.03; Shock Upper Bound 0.02; Additional value 0.63
  - Productivity: Supply 0.74; Demand 0.37; Lower Bound 0.01; Upper Bound 0.58; Shock Lower Bound 0.62; Shock Upper Bound 0.19; Additional value 0.91
  - Unit labor cost of the manufacturing sector: Supply 0.68; Demand 0.66; Lower Bound 0.06; Upper Bound 0.93; Shock Lower Bound 0.33; Shock Upper Bound 0.05; Additional value 0.84
  - Exports total: Supply 0.84; Demand 0.36; Lower Bound 0.02; Upper Bound 0.65; Shock Lower Bound 0.02; Shock Upper Bound 0.01; Additional value 0.42
  - EXP SITC 3: Mineral fuels, lubricants and related materials: Supply 0.50; Demand 0.09; Lower Bound 0.05; Upper Bound 0.76; Shock Lower Bound 0.75; Shock Upper Bound 0.09; Additional value 0.87
  - EXP SITC Total: Supply 0.91; Demand 0.24; Lower Bound 0.01; Upper Bound 0.60; Shock Lower Bound 0.27; Shock Upper Bound 0.01; Additional value 0.56
- Table I-4: Forecast Error Variance of the Common Components of Germany Variables Explained by the Supply and Demand Shock to Unit Labor Costs in Manufacturing, 1981-2006 (forecast horizon 20 quarters; bootstrap 500 draws)
  - GDP: Supply 0.70; Demand 0.13; Lower Bound 0.01; Upper Bound 0.83; Shock Lower Bound 0.82; Shock Upper Bound 0.11; Additional value 0.86
  - Private investment: Supply 0.93; Demand 0.05; Lower Bound 0.02; Upper Bound 0.58; Shock Lower Bound 0.86; Shock Upper Bound 0.28; Additional value 0.90
  - Unit labor cost of the manufacturing sector: Supply 0.74; Demand 0.34; Lower Bound 0.00; Upper Bound 0.62; Shock Lower Bound 0.66; Shock Upper Bound 0.35; Additional value 0.96
  - Exports total: Supply 0.69; Demand 0.15; Lower Bound 0.01; Upper Bound 0.79; Shock Lower Bound 0.81; Shock Upper Bound 0.14; Additional value 0.87
  - EXP SITC Total: Supply 0.92; Demand 0.37; Lower Bound 0.05; Upper Bound 0.82; Shock Lower Bound 0.44; Shock Upper Bound 0.01; Additional value 0.46
- Table I-5: Forecast Error Variance of the Common Components of Netherlands Variables Explained by the Supply and Demand Shock to Terms of Trade, 1981-2006 (forecast horizon 20 quarters; bootstrap 500 draws)
  - GDP: Supply 0.70; Demand 0.46; Lower Bound 0.08; Upper Bound 0.92; Shock Lower Bound 0.26; Shock Upper Bound 0.01; Additional value 0.43
  - Consumer prices: Supply 0.92; Demand 0.04; Lower Bound 0.00; Upper Bound 0.30; Shock Lower Bound 0.93; Shock Upper Bound 0.64; Additional value 0.98
  - Terms of trade: Supply 0.42; Demand 0.11; Lower Bound 0.01; Upper Bound 0.43; Shock Lower Bound 0.88; Shock Upper Bound 0.56; Additional value 0.99
  - Exports total: Supply 0.69; Demand 0.54; Lower Bound 0.12; Upper Bound 0.91; Shock Lower Bound 0.33; Shock Upper Bound 0.01; Additional value 0.48
  - EXP SITC Total: Supply 0.92; Demand 0.05; Lower Bound 0.01; Upper Bound 0.51; Shock Lower Bound 0.89; Shock Upper Bound 0.36; Additional value 0.96
- Table I-6: Forecast Error Variance of the Common Components of Germany Variables Explained by the Supply and Demand Shock to Terms of Trade, 1981-2006 (forecast horizon 20 quarters; bootstrap 500 draws)
  - GDP: Supply 0.87; Demand 0.39; Lower Bound 0.05; Upper Bound 0.82; Shock Lower Bound 0.03; Shock Upper Bound 0.00; Additional value 0.33
  - Consumer prices: Supply 0.92; Demand 0.05; Lower Bound 0.01; Upper Bound 0.58; Shock Lower Bound 0.20; Shock Upper Bound 0.01; Additional value 0.65
  - Terms of trade: Supply 0.20; Demand 0.62; Lower Bound 0.00; Upper Bound 0.80; Shock Lower Bound 0.37; Shock Upper Bound 0.19; Additional value 0.91
  - EXP SITC Total: Supply 0.91; Demand 0.08; Lower Bound 0.01; Upper Bound 0.31; Shock Lower Bound 0.75; Shock Upper Bound 0.41; Additional value 0.94

### Technical appendix note (band-pass filtering and frequency-domain issues)
- Assume tX is an I(1) process with ΔtX = vt such that vt has a Wold representation. The spectral density of vt is fvv(λ) > 0, for all λ.
- Discrete Fourier transform expression shows a deterministic trend term in the frequency domain with a random coefficient (0 1/2 n XX n− ).
- Corbae and Ouliaris (2006) result cited: imposing ()()10nnXXXX−=− produces an estimate that:
  - will have no finite sampling error;
  - has superior endpoint properties;
  - has much lower mean-squared error than popular time-domain filters such as HP or B-K;
  - and, "in contrast to B-K, it is consistent."
- The paper uses this as "the ideal band-pass filter."

*Source: _cr08172 - 28.      It is difficult to overestimate the relevance of product and factor markets*

### 30.      This study uses a large dimensional approximate dynamic factor model in the

### _cr08172 - 30.      This study uses a large dimensional approximate dynamic factor model in the

### Model setup and assumptions
- The N-dimensional panel Yt = (y1t, ..., yNt)' with T observations is represented as the sum of a common component Xt = (x1t, ..., xNt)' and an idiosyncratic component Ξt = (ε1t, ..., εNt)' (equation (1)): Yt = Xt + Ξt.
- The common component Xt is driven by r common factors Ft = (f1t, ..., frt)' with factor loadings C (an N×r matrix, r << N). The static factor model treats lagged or dynamic factors tF as additional static factors (so common factors include both lagged and contemporaneous factors).
- The idiosyncratic component is weakly correlated cross-sectionally and may be serially correlated.
- Under the law of large numbers (as N → ∞, with T fixed or large), the idiosyncratic component vanishes and the common component can be consistently estimated by principal component analysis (PCA).
- PCA implementation: compute the first r eigenvalues and eigenvectors from cov(Yt) to form the r×N matrix V; factor loadings satisfy C = V, so Ft = V'Yt (equation (2)) and Xt = V V' Yt (equation (3)). The idiosyncratic component is Ξt = Yt − Xt (equation (4)).
- The number of static factors r is chosen using Bai and Ng (2002) information criteria.

### Dynamics and estimation of factor dynamics
- The common factors Ft are modeled by an autoregressive representation of order 1 (VAR(1)):
  - Ft = B Ft−1 + ut (equation (5)), where B is an r×r matrix and ut is an r×1 residual vector.
- Residuals ut were white noise in the study, motivating the choice of VAR(1) as a parsimonious dynamic representation.

### Identification of structural shocks (three-step scheme)
- Step 1 — Orthogonalization and impulse-response calculation:
  - Decompose forecast-error residuals ut into k mutually orthogonal innovations using Cholesky decomposition of Var(ut) so that vt = A^{-1} ut (with Avt = u t in text), and E(vt vt') = I (equation (6) context).
  - Impulse-response of yi,t for an identified shock at horizon k uses ABcR_k (equation (7)) with ci the ith row of factor loadings C and R_k variance-covariance aggregation (0k = Σ_j R_ij^k for k-step ahead prediction errors).
  - Identification focuses on chosen variables: unit labor costs in manufacturing (ULCM) and terms of trade (TOT).
- Step 2 — Linking orthogonal shocks to fundamentals:
  - Identified orthogonal shocks vt are assumed linearly related to a vector of fundamentals ωt via vt = Q ωt (equation (8)), where Q is r×r.
  - Select Q so that the first column q1 maximizes the explained k-step-ahead forecast-error variance of the chosen variable (e.g., Dutch ULCM). The variance to be maximized is σ_k = q1' S_k q1, where S_k is constructed as in the text (equations (9) and surrounding).
  - The maximization under constraint q1'q1 = 1 yields q1 as the first eigenvector of S_k (Lagrangian in equation (10)); Q is the matrix of eigenvectors of S.
  - Horizons considered: 0k = 0 to 19k = (five years), covering short- and medium-run dynamics.
- Step 3 — Rotation and Monte Carlo exploration:
  - Principal component orthogonal shocks are identified only up to orthogonal rotation. For r = 2, fundamentals ωt are multiplied by a 2×2 orthogonal rotation matrix P = [[cos θ, sin θ], [−sin θ, cos θ]], with θ ∈ (0, π/0?) (text: ), and θ varied on a grid.
  - The paper uses a grid of 30 degrees (footnote context) which implies 12 angles between 0 and π. The text reports "6,191,736,421x10 10 (12 10) rotations." The rotated factors Pw ωt still explain the total variation of the first two eigenvalues; rotations produce all possible pairings of principal components.
  - A sign-identification strategy (Peersman, 2005) imposes inequality sign restrictions on impulse responses based on a typical aggregate demand and aggregate supply framework. Only rotations producing structural meaning and plausible impulse responses/variance decompositions are accepted.

### Sign restrictions used for identification (contemporaneous and first year)
- Supply shock vs demand shock sign restrictions (contemporaneous and during first year):
  - ULCM: supply shock ≥ 0; demand shock ≤ 0
  - Output: supply shock ≤ 0; demand shock ≥ 0
  - Real wages: supply shock ≤ 0; demand shock ≥ 0
  - Terms of trade: supply shock ≥ 0; demand shock ≥ 0
  - Consumption: supply shock ≥ 0; demand shock ≤ 0
  - Current account: supply shock ≤ 0; demand shock ≥ 0
- The sign-identification strategy accepts rotations only when they produce plausible impulse-response functions and variance decompositions.

### Monte Carlo rotation grid and computational remarks
- Rotation grid: uses 30-degree increments, producing 12 angles between 0 and π.
- The text reports an extreme number of rotations in the explored grid: "6,191,736,421x10 10 (12 10) rotations." Rotations preserve the total variation explained by the first two eigenvalues while associating principal components with structural shocks via matrix P.

### Fiscal implications for the Netherlands (summary conclusions)
- International tax competition within the EU has led to a "race to the bottom" in corporate income tax (CIT) rates, with sharp declines in statutory CIT rates in the period 2000–06.
- Despite rate cuts, CIT revenue in the Netherlands (as a share of GDP and of total tax revenue) shows a slightly upward trend since the early 1980s—presenting a CIT rate-revenue puzzle.
- Main findings on revenue effects and mechanisms:
  - Base-broadening CIT reform (e.g., reduced investment tax credits, loss-offset rules, interest deductibility, fiscal depreciation) is unlikely to have offset the ex-ante revenue losses from rate reduction.
  - Part of ex-ante revenue losses was offset by a shift between personal and corporate income taxes: lower corporate tax rates may have induced entrepreneurs to incorporate, broadening the corporate tax base and eroding personal income tax revenue.
  - International profit shifting may have partially compensated ex-ante revenue losses for the Netherlands.
- Estimated medium-term fiscal implication:
  - Tax competition with the new EU member states (NMS) may result in a decline of CIT revenue by ½–1½ percentage point of GDP.
- Paper organization note: subsequent sections assess interaction between the Netherlands and other western European countries on CIT rate setting; analyze the CIT rate-revenue puzzle; assess potential fiscal and policy implications for the Netherlands.

*Prepared by Wendly Daal (excerpt).*

### 45.      Developments in the last two decades suggest that EU member states have been

### Developments in the last two decades suggest that EU member states have been interacting in setting CIT rates

### Interaction in CIT rate setting and empirical estimates
- Statutory CIT rates in the Netherlands and several western European countries declined substantially over the period 1982–2006, from about 50 percent in 1982 to about 30-35 percent in 2006.
- Two measures that account for base broadening—the effective average corporate tax rate (EATR) and the effective marginal corporate tax rate (EMTR)—also show downward trends.
- Empirical findings from the literature:
  - Devereux, Lockwood, and Redoano (2004) (industrialized OECD, 1982–99): a 1 percentage point reduction in the average statutory CIT rates of other countries tends to reduce the home country rate by 0.7–0.8 percentage points.
  - Redoano (2007) (western Europe): for various weightings (geographical distance, GDP, per capita GDP, uniform), the parameter estimate for the average CIT rate of neighboring countries is positive and significantly above 1—implying that if neighboring countries lower taxes by 1 percentage point, the home country reacts by lowering its CIT rate by more than 1 percentage point.
  - Redoano’s results suggest tax competition in western Europe occurs mainly with respect to big “leader” countries (GDP weight performed better).
- Implication for the Netherlands:
  - Cutting of CIT rates in the Netherlands is largely due to tax competition within western European countries.
  - A 1 percentage point reduction in the average CIT rate in neighboring western European countries could lead to a rate cut of about 0.8–1.0 percentage point in the Netherlands.
  - Rate cuts in large neighboring countries, such as Germany and France, could lead the largest reaction in the Netherlands.

### The CIT rate–revenue puzzle (Netherlands)
- Observed divergence: statutory tax rate declined strongly, yet CIT revenues (as share of GDP and as share of total tax revenues including social security contributions) have been volatile but growing over time.
- Despite contraction at the beginning of the 2000s, Dutch CIT revenues remained much higher than during the 1980s and the first half of the 1990s.
- Possible explanations examined:
  - Broadening of the “legal” CIT base (statutory changes).
  - Broadening of the “economic” CIT base (increased profitability, enlarged domestic corporate sector, income tax base shifting from PIT to CIT, international profit shifting, and FDI).
- Evidence on legal base broadening:
  - Effective CIT rates at all levels of profitability declined in line with the statutory rates, implying the legal base for tax collection remained practically unchanged (text figure).
  - An ex-ante revenue-neutral reform (rate-cutting with base-broadening) would show a larger decline in the EATR for more profitable investments and an increase in the EMTR; this pattern was not observed.
  - Caveat: operational proxies for the legal corporate tax base (here proxied by the present value of depreciation allowances) may underestimate broader statutory changes affecting deductibility (depreciation, investment tax credits, loss offsets, interest deductibility).
- Evidence on economic base broadening:
  - The share of pretax corporate profits to GDP expanded from about 11 percent of GDP in 1979 to 16 percent in 2006.
  - Corporate profitability (profits as share of stock of capital) fluctuated around a more or less constant long-term trend for nonfinancial firms; increased profitability in the financial sector could have contributed but is unlikely to fully explain the enlargement of the CIT base.
  - Growth of the corporate sector measured by investment did not contribute to expansion: share of nonfinancial corporate investment to GDP declined; financial sector investment did not compensate and weakened in the last five years.
- Income tax base shifting:
  - Literature (De Mooij and Nicodème (2008)) finds gap between personal and corporate tax rates increases degree of incorporation in the EU; estimated that about 12 percent of the CIT-to-GDP ratio in early 1990s and 21 percent in recent years could be attributed to income shifting in the EU.
  - For the Netherlands, applying De Mooij and Nicodème (2008) methodology: income tax base shifting accounts on average for about 35 percent of CIT revenue due to a widening gap between PIT and CIT.
    - Implication: an ex-ante reduction in the CIT rate equivalent to one euro will cost on average only 65 euro cents in terms of CIT revenue lost ex-post; 35 euro cents can be regained through income shifting from the personal to the corporate tax base.
    - The gain in CIT revenue from income shifting erodes PIT revenue.
  - Recent 2006 tax reform: kept differential between marginal effective CIT rate and marginal effective PIT rate unchanged for SMEs by reducing the tax base for these enterprises by 9 percent—should prevent further income shifting for SMEs but implies tax competition may reduce PIT revenue as a percent of GDP as well as CIT revenue.
- International profit shifting and FDI:
  - De Mooij (2005) calculations:
    - With an FDI elasticity of -2.4, a 1 percentage point reduction in statutory CIT rate broadens the economic CIT base by 0.8 percent through FDI inflows, implying about 25 percent of ex-ante revenue loss is regained through FDI.
    - With an international profit shifting elasticity of -2.0, 30 percent of ex-ante revenue loss from a 1 percentage point reduction would be regained through international profit shifting.
  - Empirical data do not provide clear evidence that net FDI flows and CIT rate differential explain broadening of the economic base from abroad.
  - The Netherlands may nevertheless attract multinational companies or profit shifting due to favorable corporate tax features (see below).
  - Net FDI flows show outflows from the Netherlands have outpaced inflows; net FDI outflows increased in the last decade—indicating the Netherlands has not benefited from corporate tax rate cuts by attracting more net FDI inflows.
- Corporate tax features attracting profit shifting:
  - Netherlands exempts from CIT dividends and capital gains received from foreign shares and has no withholding taxes on interest payments (per Fynatten (2007)).
  - Recent reforms (abolition of capital duty on capital contributions; reduction in general dividend withholding tax rate) reinforced attractiveness as a holding company location.
- Net conclusion on drivers of CIT revenue resilience in the Netherlands:
  - Income tax base shifting appears to have been an important contributor (about 35 percent of CIT revenue on average).
  - International factors (profit shifting, FDI) may have contributed but evidence is inconclusive; net FDI inflows do not indicate a positive contribution.

### Looking ahead: potential fiscal implications and scenarios
- Context:
  - Recent accession of NMS (new member states) has raised concerns about intensified tax competition.
  - CIT rates in the Netherlands declined from 35 percent to 25.5 percent over the last decade.
  - Average CIT rates for the NMS fell from 32 percent in 1995 to below 18 percent in 2007.
  - Decline in NMS rates likely to continue due to competition from CIS and SEE countries.
- Projected strategic interaction and CIT rate in Netherlands:
  - Parameter estimates imply a 1 percentage point reduction in average statutory CIT rates of other EU countries would tend to reduce the Netherlands CIT rate by about 0.8–1 percentage point.
  - Assuming full convergence in CIT rates in the EU to the average of the NMS, estimate: CIT rate in the Netherlands may decline to less than 20 percent in the medium term (consistent with Bettendorf, Gorter, and van der Horst (2006)).
- Fiscal impact scenarios (Staff estimates):
  - With unchanged tax base, reduced CIT rate will lower the ratio of CIT revenues to GDP from 3.7 percent in 2006 to about 2 percent of GDP in the medium term.
  - Allowing for base broadening (income tax base shifting, international profit shifting, and FDI), the fall in the CIT ratio could be limited to about 3 percent of GDP.
  - Text table (Netherlands: CIT-to-GDP Ratio, in percent):
    - 2006:
      - CIT ratio (unadjusted): 3.7
      - Gain from international profit shifting: 0.6
      - CIT ratio (adjusted for profit shifting): 3.1
      - Gain from income shifting: 0.6
      - CIT ratio (adjusted for income and profit shifting): 2.5
    - Medium-Term:
      - CIT ratio (unadjusted): 3.2
      - Gain from international profit shifting: 0.6
      - CIT ratio (adjusted for profit shifting): 2.6
      - Gain from income shifting: 0.6
      - CIT ratio (adjusted for income and profit shifting): 2.0
- Factors that could reduce revenue offsets from international profit shifting:
  - Ongoing CIT harmonization within the EU (proposed CCCTB directive in 2008) would consolidate tax bases and allocate the consolidated base across member states via a sharing mechanism; member states would still apply their own rates but harmonization (including holding company regimes) will limit ability to attract multinationals and gains from profit shifting.
  - If tax differentials with the NMS persist, the Netherlands may see diminished profit shifting gains because:
    - Large Dutch multinationals may relocate profit or investment to NMS.
    - Revenue gains previously obtained from profit shifting from higher-rate neighbors (e.g., Germany) may deteriorate.
  - Size of potential revenue loss from profit shifting and investment relocation is difficult to assess.

### Policy options (context and implications)
- Fiscal implications of intensified tax competition:
  - Prospect of either losing CIT revenue to NMS or engaging in a race to cut CIT rates.
  - Expected further decline in Netherlands CIT rate to potentially less than 20 percent in the medium term.
  - Potential medium-term CIT-to-GDP ratios (with and without base broadening) indicate significant fiscal risks if tax base adjustments do not offset rate cuts.
- Policy considerations highlighted by the analysis:
  - Income tax base shifting has materially offset rate-cut revenue losses; policies preserving or limiting further shifting (e.g., harmonizing PIT/CIT differentials for SMEs) can affect distribution of tax burdens between PIT and CIT.
  - Harmonization of corporate tax bases (CCCTB) will reduce scope for international profit shifting and erode some of Netherlands’ revenue offsets.
  - Competitive features that made the Netherlands attractive to multinationals (dividend and capital gains exemptions, absence of withholding taxes on interest, reduced dividend withholding tax, abolition of capital duty) are policy levers that influence FDI and profit-shifting patterns.
  - Anticipated policy trade-offs:
    - Continued rate reductions to remain competitive risk eroding CIT revenue share unless offset by legal base broadening or economic base expansion.
    - Measures to limit income-shifting can protect PIT revenue but may reduce CIT revenue gains; recent reforms reducing SME tax base by 9 percent preserved marginal differentials and should prevent further SME income shifting but imply tax competition may reduce PIT revenue as a percent of GDP as well.

*Source: IMF staff analysis in _cr08172 - 45. Developments in the last two decades suggest that EU member states have been interacting in setting CIT rates_*

### 66.      Formal CIT rate harmonization in the EU, a possible way to cope with

### 66.      Formal CIT rate harmonization in the EU, a possible way to cope with international tax competition, is difficult.

### Difficulty and implications of formal CIT rate harmonization
- Formal harmonization would require standardization of the system of corporate income taxation; "at present, this is not a realistic option."
- Brøchner and others (2006) argued that gains from CIT rate harmonization are likely to be small because many important forms of saving and investment outside the corporate sector would continue to be subjected to widely diverging national tax rules, just as personal tax rates on corporate source income continue to differ.
- Member states are unlikely to give up a substantial part of their residual tax sovereignty because it would limit governments' ability to use corporate taxation to implement social and economic policies.
- Practical consequence: extension of tax harmonization in the EU from VAT and excises to corporate taxation would mean that, on average, less than one-third of tax revenues would remain under the control of national governments.
- Additional implication: such harmonization would "inevitably require some harmonization of budgetary expenditures as well, because of the constraints imposed on the revenue side."

### Domestic policy context and the challenge of aging
- Demographic aging implies a shrinking working-age population and a fall in potential growth if policies remain unchanged.
- Productivity must become the predominant source of growth.
- A key challenge: introduce tax reforms to increase labor supply and guarantee long-term sustainability of public finances in the face of demographic changes.
- Central policy question: "whether the Netherlands should rely on labor taxes to compensate for the expected revenue loss from further CIT rate cuts."

### Domestic policy options to offset CIT revenue loss
- General assessment:
  - Raising the PIT (or social payroll contributions) is "not the optimal strategy."
  - The tax burden on labor in the Netherlands "is still high by international standards."
  - Further easing of the tax burden on labor and reducing labor market rigidities are priorities to increase labor supply and demand and to boost economic growth.
  - Consequently, expected CIT revenue loss (and desirable easing of labor taxes) may have to be paid for by cuts in primary expenditure, but scope is limited due to aging-related upward pressure on public spending.
  - Alternative: shift more of the burden onto other tax bases (consumption or property).
- Consumption taxes (VAT):
  - Advantages listed: (i) minimize economic efficiency losses; (ii) be relatively neutral toward saving and investment decisions; (iii) avoid discriminating between imports and locally produced goods; (iv) preserve external competitiveness; (v) provide symmetric treatment of labor and capital income, reducing disincentives to work and better meeting intergenerational equity criteria than income taxes.
  - With a VAT rate slightly below the EU average, the Netherlands could probably raise consumption taxes.
  - Quantified scenario: "Increasing the VAT by 1–2 percentage points, assuming constant elasticities, would deliver revenues of about ½–1½ percent of GDP, enough to compensate for the expected decline in CIT revenues over the medium term."
- Property and preferential treatment of saving instruments:
  - The Netherlands can relax tax privileges granted to specific saving instruments; retirement schemes and housing investment benefit from the most generous tax breaks (motivated partly by social/economic objectives).
  - On average, revenue from property taxes as a share of GDP in the Netherlands "has not increased in line with the values of properties over the period 2000-06."
  - Some tax concessions have recently been lowered, but "there is still scope for further action that could partially compensate the revenue loss from international tax competition."
  - Higher taxation of property would improve neutrality of the tax system toward various forms of wealth (real estate, financial assets, and human capital).
- Environmental taxes:
  - The Netherlands already relies relatively more on environmental taxes than most other OECD countries; motor fuel and vehicle taxes account for the bulk of these revenues.
  - The Netherlands has implemented comprehensive "green tax" reforms over the last decade, including automobile taxation based largely on emissions.
  - Further venues: tax products based on the pollution created by their fabrication process, in light of global warming concerns.

### Empirical findings on international tax competition and CIT revenue
- Evidence indicates strong international tax competition within western European countries, including the Netherlands: "parameter estimates for the strategic interaction are highly significant and positive."
- The study finds that part of the decline in CIT revenue due to rate-cutting has been compensated with a shift from PIT to CIT revenue, explaining the "CIT puzzle." Thus, international tax competition has been reducing PIT instead of CIT revenue.
- International profit shifting may have contributed to the growing CIT revenue historically.
- Future prospects:
  - Accession of the NMS will intensify tax competition in the EU and "this may erode CIT revenue in the Netherlands."
  - The size of the potential impact depends on income tax base shifting and international profit shifting.
  - International profit shifting is "the most critical factor for the Netherlands, as it determines whether total tax revenue will decline or not."

### Policy conclusions and recommended strategic direction
- Policy options must be evaluated against aging: population aging will raise fiscal spending while lowering the potential of labor-based taxes.
- "The best strategy to address the potential decline in revenue as a share of GDP in the medium term is to broaden the tax base."
- With increased mobility of capital and high-skilled workers within the EU, broadening the base is best accomplished by shifting away from mobile tax bases toward greater reliance on consumption and property taxes.
  - Consumption taxes will likely reduce deadweight losses.
  - Property taxes will improve neutrality of the tax system toward different sources of wealth.
- Environmental taxes could also play a larger role.

*Source: _cr08172 - 66.      Formal CIT rate harmonization in the EU, a possible way to cope with international tax competition.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2008/_cr08172.pdf_