## 1nldea2019002

## Source details

**Canonical URL:** [1nldea2019002](https://www.imf.org/-/media/files/publications/cr/2019/1nldea2019002.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2019/1nldea2019002.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2019/1nldea2019002.pdf.json)

---

### Corporate saving and the Dutch current account
- The Netherlands’ current account (CA) surplus:
  - Historical average: about 5 percent of GDP.
  - 2012: 10.7 percent of GDP.
  - 2015: 6.3 percent of GDP.
  - 2017: 10.5 percent of GDP.
- Major contributors and balances:
  - Trade surplus in goods averaged 10 percent of GDP over the last decade.
  - Household net saving averaged around ¼ percent of GDP over the past two decades.
  - Government net saving: about –5½ percent of GDP in 2009 to about 1 percent of GDP in 2017.
  - Dutch non-financial corporations (NFCs) net saving averaged around 6 percent of GDP since 2000.
- International investment positions:
  - Gross international assets and liabilities rose from close to 6 times of GDP in 2003 to more than 11 times of GDP in 2017.
  - About 60 percent of assets and 50 percent of liabilities are FDI.
  - Net IIP moved to 60 percent of GDP in 2017, explained by the increase in net FDI position.
  - Special financial institutions (SFIs) account for about half of Dutch foreign assets and liabilities.

### Sources and uses of Dutch corporate gross saving
- Definition:
  - NFC gross savings (retained earnings) = gross value added (GVA) less costs of labor, taxes, interest and rent, dividends, and other net transfers (including reinvested earnings from FDI).
- Trends and cross-country structure:
  - Saving rate: 26 percent in 1995 → 35 percent in 2012 → 29 percent in 2016.
  - Labor costs as percent of GDP stable; declining labor share in GVA.
  - Cross-country comparisons: Dutch NFC cost structure similar to Denmark; Dutch firms had higher dividend payout than Denmark; German NFCs had very low costs but paid much higher dividends; compared with Swiss NFCs, Dutch firms had much lower labor costs.
- Investment and use of gross saving (1995–2017 averages unless noted):
  - Dutch firms invested about 10 percent of GDP per year in capital formation (1995–2017 average).
  - Dutch gross saving averaged 17 percent of GDP; only 63 percent of savings spent on domestic business investment (investment-saving ratio 0.62).
  - Comparative investment-saving ratios (1995–2016 averages):
    - Denmark: 0.78
    - Sweden: 0.96
    - France: 1.21
    - Italy: 1.25
    - Luxembourg: 1.27
- Uses of NFC net saving (1995–2017 averages):
  - Net saving averaged 7 percent of GDP.
  - Issuance of equity averaged 4 percent of GDP; debt/loans 6 percent of GDP.
  - Purchases of equity averaged 8 percent of GDP; debt/loans 5 percent of GDP; current assets 2 percent of GDP.
  - Net basis: most Dutch NFC net saving went to equity purchases, with co-movement indicating large extent used for foreign equity purchases.

### Roles of MNCs in corporate saving and investment allocation
- Concentration of gross savings:
  - Large firms contributed, on average, half of NFC gross savings during 2010–17.
  - The top 24 firms contributed about 40 percent of total NFC gross savings.
  - Royal Dutch Shell plc (Shell) reported gross savings around 2.8 percent of Dutch GDP in 2017 (about 15 percent of total NFC gross savings).
- Investment behavior of top firms:
  - Top 24 firms spent around 84 percent of their gross savings on investment either in the Netherlands or abroad (2010–2017 average); national average domestic investment rate was 54 percent.
  - Excluding Shell, the two investment ratios are similar.
  - Shell invested on average about 1½ times its gross savings between 2010 and 2017, financed by debt or equity issuance—an amount close to 2.8 percent of Dutch GDP in 2016.
  - 60 percent of Shell’s expenditure was for upstream investment; only 17 percent of that was invested in Europe.
  - If all European investment were in the Netherlands, Shell would have contributed 1.8 percent of GDP to the Netherlands CA surplus in 2016.
- Sectoral variation in investment intensity (2010–2017 averages; columns: Share in total gross saving; Investment-saving ratio; Stock buyback in percent of gross saving; Intangibles in percent of total assets):
  - Energy: Share 32.8; Investment-saving ratio 0.99; Stock buyback 5.9; Intangibles 11.6
  - Consumer Staples: Share 17.1; Investment-saving ratio 0.80; Stock buyback 17.1; Intangibles 25.2
  - Industrials: Share 15.7; Investment-saving ratio 0.39; Stock buyback 16.5; Intangibles 40.2
  - Communication Services: Share 12.4; Investment-saving ratio 0.46; Stock buyback 67.0; Intangibles 46.8
  - Information Technology: Share 8.0; Investment-saving ratio 0.14; Stock buyback 36.0; Intangibles 35.6
  - Materials: Share 6.1; Investment-saving ratio 0.62; Stock buyback 5.9; Intangibles 19.8
  - Health Care: Share 6.0; Investment-saving ratio 0.16; Stock buyback 20.1; Intangibles 36.2
  - Utilities: Share 1.9; Investment-saving ratio 3.21; Stock buyback 13.4; Intangibles 0.9
- Correlations and financing behavior:
  - Intangibles and stock buybacks correlate with low domestic investment.
  - High-investment sectors (energy, utilities) often issue debt to finance investment.

### Attribution and international ownership complexities
- Ownership and attribution challenges:
  - Dutch MNCs are largely owned by foreigners, but ownership often routed through Dutch holding companies.
  - Example ownership disclosures for Shell: Nederlands Centraal Instituut Voor Giraal Effectenverkeer BV holding 41.2% of shares; BlackRock Investment Management (UK) Ltd. holding 4.7%.
  - Eggelte et al. (2014) indicate around 90 percent of Shell shares are owned by foreigners.
  - Complex holding structures make allocation of retained earnings to ultimate shareholders difficult and complicate attribution of retained earnings in the CA.

### From corporate saving to the current account
- Trade and re-exports:
  - Major exports: manufactured goods (machinery and transport equipment), food and fuels; main destinations: Germany, France, United Kingdom, Belgium.
  - More than 40 percent of Dutch exports are re-exports; Rotterdam’s port plays a major role.
  - OECD (2017) shows foreign-owned firms in the Netherlands are more export intensive than domestic-owned firms.
- Primary income balance:
  - Dividends received and paid by Dutch firms are in the range of around 30 to 40 percent of GDP, derived from even larger stock positions.
  - Primary income balance is small, implying large parts of dividends are paid to foreigners and that net flows may not align with rising stock positions because of intra-MNC distributions.

### Conclusion on corporate saving and CA implications
- High Dutch CA surplus is driven by high corporate saving dominated by a few MNCs.
- Dutch NFCs, particularly MNCs headquartered in the Netherlands, receive substantial dividends and interest from foreign subsidiaries, often exempt from domestic corporate taxes under the “double participation exemption.”
- MNCs’ savings are used for global investment through foreign equity purchases, benefiting from exemptions on capital gains on foreign equity; consequently, the Dutch BoP records high corporate net saving.
- Retained earnings in principle should be attributed to foreign shareholders (leading to CA outflows), but complex holding structures may prevent attribution and potentially lead to an overestimate of the CA surplus.
- Large trade surpluses and small primary income balances are consistent with the dominance of MNCs in the Netherlands’ external positions.

*Source: Kingdom of the Netherlands — Netherlands (International Monetary Fund), January 23, 2019.*

---

### Separating MNCs’ activities from the Dutch current account (external sector assessment)
- Rationale for separation:
  - Rising corporate saving is a global phenomenon and could be related to rising market power or profit shifting, possibly associated with rising intangible asset intensity.
  - Granular information on MNCs’ activities in the Netherlands is needed to disentangle global factors from domestic policy distortions.
  - Separating MNCs’ activities from both internal and external accounts would help identify domestic policy distortions and address imbalances effectively, and help identify imbalances of other economic sectors.
- Observed domestic distortions and distributional issues:
  - SMEs are stagnant and remain financially constrained.
  - Small household net saving hides that households are still highly leveraged and consumption constrained by stagnating disposable income.
- Policy recommendation:
  - Improve statistics and separate MNCs’ activities from both internal and external accounts to better identify domestic policy distortions and address imbalances effectively.

### Self-employment and support for Dutch pension reform — context and objectives
- System context:
  - The Dutch pension system ranked first in a group of 34 countries in the Melbourne Mercer Global Pension Index 2018.
  - Pressure points: population aging, low long-term interest rates, changes in labor market structure.
  - Timeline: consultations initiated in 2014; coalition agreement of 2017 proposed reform route; agreement with social partners expected in early 2018 with legislation to be passed in 2020 (agreement on final package delayed).
- Paper objective and key finding:
  - The chapter analyzes pension system preferences and self-employment.
  - Finding: after controlling for demographic characteristics and individual psychological traits, individuals who prefer pension plans with more freedom of choice and more investment risk are significantly more likely to choose self-employment than regular employment.
  - Policy implication: a pension reform giving more freedom of choice to regular employees should reduce incentives to go into self-employment and include individuals who do not enjoy pension coverage into personal savings schemes.

### Dutch pension system design and reform features
- Three pillars:
  - 1st pillar: public, pay-as-you-go, minimum flat pension (AOW) subject to age and residency requirements.
  - 2nd pillar: mandatory occupational schemes covering over 90 percent of all employees; supplementary benefits based on lifetime wages.
  - 3rd pillar: voluntary, tax-exempt contracts; contributes about 10 percent of pension systems’ assets.
- 2nd pillar characteristics:
  - Contributions quasi-mandatory, derived from collective labor agreements; levied at a uniform rate, independent of age; can be increased if the pension fund falls below the target solvency ratio but have de facto been fixed.
  - Benefits accrued annually at a maximum rate of 1.875 percent, providing 75 percent of average lifetime pay after 40 years of contributions.
  - Dutch DB contracts are deferred variable annuities with annuities indexed to wages or inflation but can be frozen or cut in nominal terms conditional on funding rate.
  - Funding ratio: 108.4 percent in mid-2018, against statutory minimum of 104.2 percent (DNB, June 2018).
- Proposed reform (coalition agreement):
  - Replace variable annuities with personal entitlements in the form of financial assets (akin to DC).
  - New system to feature insurance contracts hedging longevity risk and possibly collective buffers allowing risk sharing across generations: “personal pensions with risk-sharing and collective buffers” (PPR-CB).
  - PPR-CB aims to produce a more predictable income stream via collective investment and decumulation, tailor instruments to individual risk preferences, and shift pension-related risks from employers to participants.
  - Reform would improve transparency (at least the PPR part) and shift risks to participants.

### Political economy, stakeholder positions, and transition costs
- Stakeholder positions:
  - Employers: aligned with government; prefer shifting risks off balance sheets; regressive premiums could lower hiring costs for younger workers and possibly crowd some self-employed into the pension system.
  - Unions: favor redistribution and more intergenerational risk sharing; insist on mandatory participation in collective buffers; favor allowing financial buffers to turn negative in downturns to avoid curtailing benefits (opposed by central bank); insist on compensation for older generations.
  - Estimated cost of compensation for older generations: 55 billion euros.
- Implementation considerations:
  - Complexity and increased transparency raise demand for individuals to take a more active role; learning process may take time.
  - Maintaining trust in institutions, currently high, is crucial for consensus and transition.

### Key pension system statistics and parameters (2018 figures and related)
- Pensionable age: increasing to 67 by 2021, then linked to life expectancy
- Pensionable age (reference): 68 in 2018
- Accrual rate, annual: 1.875
- Maximum (in eur, 2017): 103,317
- Average (in eur, 2017): 14,737
- Contributions (gross, billions): 31.2
- Contributions: employers 17, workers 7
- Contributions as Percent of GDP: 4.4
- Size of fund (in 2017) Billions eur: 1620
- Size of fund (in 2017) Percent of GDP: 220
- Non-indexed funding ratio (2017): 103/105 (regulatory min)
- Indexation post-retirement: minimum wages / CPI

### Self-employment trends and characteristics
- Trends:
  - Self-employment share of total employment increased by more than 4 percentage points between 2005 and 2017; 15.7 percent in 2017.
  - “Own-account workers” accounted for 11.6 percent of employed in 2017.
  - About 40 percent of self-employed belong to age group 50–64.
  - Slightly more than one in three self-employed are women.
  - Half of self-employed in 2017 active in construction, trade, social work and professional, scientific and technical activities.
  - The share of self-employed with tertiary education expanded in recent years.
- Motives and concerns:
  - Motivations: autonomy; lower individual tax burden (self-employed on average pay 20 percent lower taxes on equivalent gross income compared to employees).
  - Self-employed do not contribute to sickness and disability insurance and tax deductions available for entrepreneurship.
  - Concerns: undermines level playing field, revenue loss for government, exposure of self-employed to unstable contractual relationships and income risk.
  - With few exceptions, self-employed are not covered by sectoral or occupational 2nd pillar arrangements.

### Evidence from the Dutch National Bank household survey (DHS)
- Survey scope and timing:
  - Administered annually to about two thousand Dutch households (five thousand individuals).
  - Latest wave (25th): April - October 2017.
- Pension coverage among self-employed:
  - Only a fifth of self-employed report current/last job entitles them to a retirement pension (aside from AOW) vs. 84 percent for employed.
  - Only 11 percent of self-employed report having made other arrangements for their pension apart from employer-built mechanisms.
  - About 45 percent of self-employed opt for managing their savings independently; about 65 percent of employees delegate to a fund.
  - Two thirds of workers claim to have received an overview of pension rights in 2016; about 40 percent of self-employed report the same.
- Risk and choice preferences:
  - Self-employed express a higher attitude toward risk taking on average and stronger preference for freedom of choice in pension management.

### Empirical model and main regression results
- Data and model:
  - Panel: answers from 2014–17; about 20,000 observations; main estimation N obs = 1,847; Pseudo R2 = 0.130.
  - PROBIT model: dependent variable SE (1 if self-employed); key independent variables DC1 (risk-taking: DNB95=2), DC2 (freedom of choice: DNB96=1); X includes demographic and psychological characteristics.
- Key coefficients (coefficient (standard error)):
  - DC1 (risk taking): 0.019 * (0.010)
  - DC2 (freedom of choice): 0.036 *** (0.010)
- Selected other coefficients:
  - Age 46-55: 0.029 ** (0.013)
  - Age 56-67: 0.041 *** (0.013)
  - Gender (F): 0.015 (0.010)
  - Tax burden: -0.020 ** (0.008)
  - Vocational_interm.: 0.063 *** (0.019)
  - Vocational_high: 0.042 ** (0.019)
  - Pre-university: 0.046 ** (0.022)
  - University: 0.062 *** (0.020)
  - Controlling: 0.015 ** (0.007)
  - Mindful: -0.030 *** (0.011)
- Interpretation:
  - DC1 and DC2 are significant and positively associated with probability of being self-employed.
  - Psychological traits: lower “mindful” score and higher “controlling” associated with higher self-employment probability.
  - Time dummies, gender, having children, civil status and gross income are not significant in main estimation except where shown.

### Implications for pension reform and labor-market outcomes
- Main empirical implication:
  - Individuals preferring greater freedom of choice and higher investment risk are more likely to be self-employed.
- Potential responses to reform:
  - A reform that gives regular employees more freedom of choice should reduce incentives to enter self-employment and could include previously uncovered individuals into personal savings schemes.
  - If reform decreases perceived pension wealth, some individuals could increase personal savings or change labor relationships.
  - Historical precedent: the 2006 reform that abolished preferential tax treatment of early retirement lowered willingness to use liquid assets for entrepreneurship.
- Policy considerations:
  - Design of pension reform interacts with tax treatment, information provision, and availability of individually managed yet inclusive pension products.
  - Consider information deficits: self-employed report lower receipt of pension overviews and feel less informed.

### Complexity, consensus, and implementation priorities for PPR-CB
- Implementation challenges:
  - PPR-CB increases transparency and complexity, requiring more active participant engagement; learning may take time even among those preferring choice.
  - Maintaining trust in institutions is crucial for consensus and successful transition.
- Informational gaps to address:
  - Quantify redistribution costs of the transition.
  - Specify modalities for financing transition costs.
  - Identify losers and winners and design compensation options within a broader reform package.
- Policy recommendations:
  - Build and preserve public trust to enable consensus and smooth transition.
  - Provide greater transparency and targeted information addressing the “missing arguments.”
  - Use political economy tools (elimination, compensation, information building) to reduce reform resistance.

*Source: IMF staff analysis and Kingdom of the Netherlands — Netherlands (International Monetary Fund), content unit 1nldea2019002.*

### References _______________________________________________________________________________ 11

### CORPORATE SAVING IN THE NETHERLANDS

### Introduction
- The Netherlands’ current account (CA) surplus has a historical average of about 5 percent of GDP.
- The CA surplus reached 10.7 percent of GDP in 2012, fell to 6.3 percent of GDP in 2015, and rose to 10.5 percent of GDP in 2017.
- The trade surplus in goods averaged 10 percent of GDP over the last decade and is the biggest contributor to the CA surplus.
- Household net saving has averaged around ¼ percent of GDP over the past two decades.
- Government net saving improved from about –5½ percent of GDP in 2009 to about 1 percent of GDP in 2017.
- Dutch non-financial corporations (NFCs) have averaged net saving around 6 percent of GDP since 2000; high NFC net saving reflects high gross saving and low corporate investment.
- Gross international assets and liabilities rose from close to 6 times of GDP in 2003 to more than 11 times of GDP in 2017; about 60 percent of assets and 50 percent of liabilities are FDI.
- Net IIP moved to 60 percent of GDP in 2017, explained by the increase in net FDI position.
- Special financial institutions (SFIs) account for about half of Dutch foreign assets and liabilities.

### Sources and Uses of Dutch Corporate Gross Saving
- NFC gross savings (retained earnings) definition: gross value added (GVA) less costs of labor, taxes, interest and rent, dividends, and other net transfers (including reinvested earnings from FDI).
- From 1995 to 2012, NFC gross saving increased faster than GVA due to lower costs and lower net dividend payments; saving rate rose from 26 percent in 1995 to 35 percent in 2012 and dropped to 29 percent in 2016.
- Labor costs as a percent of GDP have been quite stable; declining labor share in GVA observed.
- Cross-country comparisons:
  - Dutch NFC cost structure similar to Denmark; Dutch saving rate slightly lower with higher dividend payout.
  - German NFCs had very low costs but paid much higher dividends.
  - Compared with Swiss NFCs, Dutch firms had much lower labor costs.
- Dutch firms spent the lowest share of their gross saving on domestic investment among peers:
  - Dutch firms invested about 10 percent of GDP per year in capital formation (1995–2017 average).
  - Dutch gross saving averaged 17 percent of GDP; only 63 percent of savings spent on domestic business investment (investment-saving ratio 0.62).
  - Comparatives (1995–2016 averages):
    - Denmark: investment-saving ratio 0.78
    - Sweden: 0.96
    - France: 1.21
    - Italy: 1.25
    - Luxembourg: 1.27
- Uses of NFC net saving (1995–2017 averages):
  - Net saving averaged 7 percent of GDP.
  - Issuance of equity averaged 4 percent of GDP; debt/loans 6 percent of GDP.
  - Purchases of equity averaged 8 percent of GDP; debt/loans 5 percent of GDP; current assets 2 percent of GDP.
  - On net basis, most Dutch NFC net saving went to equity purchases, with co-movement indicating large extent used for foreign equity purchases.

### Roles of MNCs
- NFC gross savings are highly concentrated:
  - Large firms contributed, on average, half of NFC gross savings during 2010–17.
  - The top 24 firms contributed about 40 percent of total NFC gross savings.
  - Royal Dutch Shell plc (Shell) reported gross savings around 2.8 percent of Dutch GDP in 2017, about 15 percent of total NFC gross savings.
- MNCs allocate savings to global investment:
  - Top 24 firms spent around 84 percent of their gross savings on investment either in the Netherlands or abroad (2010–2017 average); national average domestic investment rate was 54 percent.
  - Excluding Shell, the two investment ratios are similar.
  - Shell invested on average about 1½ times its gross savings between 2010 and 2017, financed by debt or equity issuance—an amount close to 2.8 percent of Dutch GDP in 2016.
  - 60 percent of Shell’s expenditure was for upstream investment; only 17 percent of that was invested in Europe.
  - If all European investment were in the Netherlands, Shell would have contributed 1.8 percent of GDP to the Netherlands CA surplus in 2016.
- Sectoral variation in MNC investment intensity (2010–2017 averages):
  - Energy: Share in total gross saving 32.8; Investment-saving ratio 0.99; Stock buyback in percent of gross saving 5.9; Intangibles in percent of total assets 11.6
  - Consumer Staples: Share 17.1; Investment-saving ratio 0.80; Stock buyback 17.1; Intangibles 25.2
  - Industrials: Share 15.7; Investment-saving ratio 0.39; Stock buyback 16.5; Intangibles 40.2
  - Communication Services: Share 12.4; Investment-saving ratio 0.46; Stock buyback 67.0; Intangibles 46.8
  - Information Technology: Share 8.0; Investment-saving ratio 0.14; Stock buyback 36.0; Intangibles 35.6
  - Materials: Share 6.1; Investment-saving ratio 0.62; Stock buyback 5.9; Intangibles 19.8
  - Health Care: Share 6.0; Investment-saving ratio 0.16; Stock buyback 20.1; Intangibles 36.2
  - Utilities: Share 1.9; Investment-saving ratio 3.21; Stock buyback 13.4; Intangibles 0.9
- Intangibles and stock buybacks correlate with low domestic investment; high-investment sectors (energy, utilities) often issue debt to finance investment.
- Attribution of retained earnings to ultimate shareholders is practically difficult:
  - Dutch MNCs are largely owned by foreigners, but ownership often routed through Dutch holding companies.
  - Example: Shell’s largest shareholder listed as Nederlands Centraal Instituut Voor Giraal Effectenverkeer BV holding 41.2% of shares; second largest BlackRock Investment Management (UK) Ltd. holding 4.7%.
  - Eggelte et al. (2014) indicate around 90 percent of Shell shares are owned by foreigners.
  - Complex holding structures make allocation of retained earnings to ultimate shareholders difficult.

### From the Saving–Investment Balance to the Current Account
- Large exports and re-exports reflect MNC activities:
  - Major exports: manufactured goods (machinery and transport equipment), food and fuels; main destinations are European countries (Germany, France, United Kingdom, Belgium).
  - More than 40 percent of Dutch exports are re-exports; Rotterdam’s port plays a major role.
  - OECD (2017) shows foreign-owned firms in the Netherlands are more export intensive than domestic-owned firms.
  - Tangible movable assets (e.g., transportation equipment) may be used for tax planning via assignment of legal ownership to leasing companies in low-tax jurisdictions.
- Primary income balance is small, implying large parts of dividends are paid to foreigners:
  - Dividends received and paid by Dutch firms are in the range of around 30 to 40 percent of GDP, derived from even larger stock positions.
  - When primary income flows are driven by MNCs, net balances may not match the rising stock positions depending on intra-MNC distributions.

### Conclusion and Implications
- The high Dutch CA surplus is driven by high corporate saving dominated by a few MNCs.
- Dutch NFCs, particularly MNCs headquartered in the Netherlands, receive substantial dividends and interest payments from foreign subsidiaries, often exempt from domestic corporate taxes under the “double participation exemption.”
- MNCs’ savings are used for global investment through foreign equity purchases, benefiting from exemptions on capital gains on foreign equity; consequently, the Dutch BoP records high corporate net saving.
- Retained earnings in principle should be attributed to foreign shareholders (leading to CA outflows), but in practice complicated holding structures may prevent this attribution and potentially lead to an overestimate of the CA surplus.
- Large trade surpluses and small primary income balances are consistent with the dominance of MNCs in the Netherlands’ external positions.

*Source: Kingdom of the Netherlands — Netherlands (International Monetary Fund), January 23, 2019.*

### 16.      Separating MNCs’ activities from the Dutch CA for the external sector assessment

### 16.      Separating MNCs’ activities from the Dutch CA for the external sector assessment

### Rationale for separating MNCs’ activities
- Rising corporate saving is a global phenomenon and could be related to rising market power or profit shifting, possibly associated with rising intangible asset intensity.
- Granular information on MNCs’ activities in the Netherlands is needed to disentangle global factors from domestic policy distortions.
- Separating MNCs’ activities from both internal and external accounts would help identify domestic policy distortions and address imbalances effectively.
- Separating MNCs’ activities would help identify imbalances of other economic sectors.

### Observed domestic distortions and distributional issues
- Small and medium enterprises (SMEs) are stagnant and remain financially constrained.
- Small household net saving hides the fact that households are still highly leveraged, and their consumption constrained by a stagnating disposable income.

### Policy implication / recommendation
- Improve statistics and separate MNCs’ activities from both internal and external accounts to better identify domestic policy distortions and address imbalances effectively.

*References cited in the unit include Chen, P., Karabarbounis, L., and Neiman, B., 2017; CPB, 2015; DNB, 2004; Eggelte, J., Hillebrand, R., Kooiman, T., and Schotten, G., 2014; OECD, 2017.*

---

### SELF-EMPLOYMENT AND SUPPORT FOR THE DUTCH PENSION REFORM

### Introduction and context
- The Netherlands’ pension system has been regarded as one of the world’s strongest and most sophisticated.
- Melbourne Mercer Global Pension Index 2018: the Dutch pension system ranked first in a group of 34 countries across 40 indicators.
- Pressure points: population aging, low long-term interest rates, and changes in the labor market structure have strained redistribution and brought intergenerational tensions to the fore.
- Timeline noted: government initiated consultations in 2014; coalition agreement of 2017 proposed a route for reform; agreement with social partners expected in early 2018 with legislation to be passed in 2020 (agreement on final package delayed).

### Paper objectives and key findings
- The chapter analyzes pension system preferences and self-employment.
- Finding: after controlling for demographic characteristics and individual psychological traits, individuals who prefer pension plans with more freedom of choice and more investment risk are significantly more likely to choose self-employment than regular employment under the current pension system.
- Policy implication: a pension reform giving more freedom of choice to regular employees should reduce incentives to go into self-employment and include individuals who do not enjoy pension coverage into personal savings schemes.

### Structure of the rest of the chapter
- Section B: background on the pension reform.
- Section C: description of the self-employed population based on household survey data.
- Section D: model and estimation results.
- Section E: conclusions.

### Background on the pension system (three pillars)
- 1st pillar: public, pay-as-you-go, minimum flat pension subject to age and residency requirements.
- 2nd pillar: mandatory occupational schemes covering over 90 percent of all employees; supplementary benefits based on lifetime wages.
- 3rd pillar: voluntary, tax-exempt contracts; contributes about 10 percent of pension systems’ assets, but growing.

### Characteristics of the 2nd pillar (Dutch occupational schemes)
- Contributions are quasi-mandatory and derived from collective labor agreements; levied at a uniform rate, independent of age; rates can be increased if the pension fund falls below the target solvency ratio but have de facto been fix.
- Benefits accrued annually at a maximum rate of 1.875 percent, providing 75 percent of average lifetime pay after 40 years of contributions.
- Dutch DB contracts are structured as deferred variable annuities with an investment strategy targeting a stable lifetime income; annuities are indexed to wages or inflation but can be frozen or cut in nominal terms conditional on the funding rate.
- 2nd pillar funds are managed by boards consisting of employers, unions, retirees, and independent specialists who act as fiduciary trustees.

### Advantages and risks of the current design
- Advantages: high pension coverage, risk pooling and risk sharing, and long-term sustainability over a 10-year horizon given fund mechanisms to address funding shortfalls through changes in revenues (contributions), accrual of benefits, and payment of earned annuities.
- Funding ratio noted: 108.4 percent in mid-2018, against the statutory minimum of 104.2 percent (DNB, June 2018).
- Risks and disadvantages: increasing complexity and opacity of transfers; uniform accrual rates imply intergenerational transfer from the young to the old and from lower educated to higher educated; indirect link between changes in contributions and benefits increases uncertainty; allocation of losses and gains across individuals and generations needs greater transparency.
- Pension funds have used overly optimistic assumptions about expected returns which allowed them to circumvent nominal cuts (Wills Tower Watson, 2018; and DNBulletin 2017). Pension funds' financial positions have been improving recently as favorable developments in the equity and foreign exchange markets have been offsetting the negative consequences of the decline in interest rates.

### Proposed reform features (coalition agreement)
- Replace variable annuities (characteristic of the Dutch “hybrid DBs”) with personal entitlements in the form of financial assets, akin to a DC.
- New system would feature insurance contracts hedging participants against longevity risk, and possibly collective buffers that allow sharing systemic risks across generations.
- Concept described as “third space” or “personal pensions with risk-sharing and collective buffers” (PPR-CB).
- PPR-CB aims to produce a more predictable income stream via collective investment and decumulation, while tailoring instruments to individual risk preferences and shifting virtually all pension-related risks that may currently be borne by employers to participants.
- Reform would improve transparency (at least the PPR part) and shift risks to participants.

### Political economy and stakeholder positions
- Public priorities: pension reform not top priority in a 2018 European Commission survey; health, social security and housing prioritized. More individuals still see pension income as an important personal preoccupation, albeit less than in 2016.
- Government rationale: reform necessary because of intergenerational risk transfers and system inadequacy for changing labor markets; contributions histories are sometimes erratic and incomplete.
- Employers’ position: aligned with government; prefer shifting risks off their balance sheets; regressive premiums (varying by age) could lower hiring costs for younger workers and possibly crowd some self-employed into the pension system.
- Unions’ position: favor redistribution and more risk sharing across generations; insist on mandatory participation in collective buffers at industry level and that self-employed should be forced to participate; they favor allowing financial buffers of funds to turn negative in downturns to avoid curtailing benefits (opposed by the central bank); insist on a stable (higher) ultimate forward rate by funds in solvency calculation; insist on compensation for older generations affected negatively by the transition.
- Estimated cost of compensation for older generations: 55 billion euros.

### Self-employment trends and characteristics
- Self-employment in the Netherlands has increased over the past decade, contrary to most other OECD countries.
- The share of self-employment in total employment picked up by more than 4 percentage points between 2005 and 2017, standing at 15.7 percent in 2017.
- The largest increase was in “own-account workers” which accounted for 11.6 percent of employed in 2017.
- Older workers are more represented among the self-employed: about 40 percent belong to the age group 50–64.
- Slightly more than one in three self-employed workers are women.
- Half of self-employed in 2017 were active in four main categories: construction, trade, social work and professional, scientific and technical activities.
- In recent years, the share of self-employed workers with tertiary education has expanded.

### Motives, advantages, and policy concerns related to self-employment
- Self-employment is often motivated by prospects of lowering the individual tax burden.
- Self-employed do not contribute to the sickness and disability insurance and tax deductions are available for incentivizing entrepreneurship (Dekker and others, 2018).
- Self-employed on average pay 20 percent lower taxes on equivalent gross income compared to employees (Bosch and others, 2015).
- The cap on tax deduction for contributions in 3rd pillar DC schemes increases with age.
- Self-employment offers flexibility and can contribute to job satisfaction; primary motive often autonomy.
- Concerns: undermines level playing field in the labor market, exposes the government to revenue loss, and subjects self-employed to risks such as unstable contractual relationships and possible income loss.
- With few exceptions, the self-employed are not covered by sectoral or occupational 2nd pillar arrangements (Mastrogiacomo and others, 2016).

*Source: 1nldea2019002 - 16. Separating MNCs’ activities from the Dutch CA for the external sector assessment (IMF).*

### 17.      Are self-employed risk attitudes and preference for freedom of choice at the core of

### 17. Are self-employed risk attitudes and preference for freedom of choice at the core of their employment status?

### Evidence from the DHS (Dutch National Bank household survey)
- Survey scope and timing:
  - Administered annually to about two thousand Dutch households (five thousand individuals).
  - The latest wave (the 25th wave) was conducted over the period April - October 2017.
- Survey content:
  - Six questionnaires covering health status, work patterns, income (including from retirement and benefits), wealth data, demographic, economic and psychological concepts.
  - Questions gauge risk sharing preferences, preference for choosing the investment profile of savings individually, and options for meeting retirement income shortfalls and pension system sustainability risks.
- Self-employment composition and demographics:
  - Two thirds of the approximately 2,000 respondents broadly falling into the labor force category has a paid job.
  - The share of self-employed with a paid job in total self-employed population is 63 percent.
  - “Self-employed and freelancers” constitute the overwhelming majority; individuals owning a family business account for less than 9 percent of total.
  - Age and gender group compositions in the survey align with the total labor force, with slightly fewer female workers represented in the self-employed.
- Education and income differences:
  - Majority of self-employed completed vocational education at high and intermediate level; self-employed are proportionally more represented at university and high vocational education compared with total population.
  - About 35 percent of self-employed fall in the first quintile on the gross income distribution, above the 20 percent recorded for the total working population.
  - The top quintile comprises a similar share across self-employed and total population.
  - On average, both groups plan to retire around age 66.
- Pension coverage and arrangements:
  - Only a fifth of self-employed answered affirmatively that their current or last job entitles them to a retirement pension (aside from the 1st pillar, AOW) compared with 84 percent for the total surveyed employed population.
  - Roughly the same share reports participating in an employer-offered plan while only 11 percent of self-employed report having made other arrangements for their pension apart from employer-built pension mechanisms.
  - Conclusion: the majority of the self-employed are not covered by any formal pension arrangement though they may still accumulate wealth in other ways.
- Information about pensions:
  - While two thirds of workers claim to have received an overview of pension rights from their fund in 2016, only about 40 percent of self-employed report the same.
  - Overall, in the self-employed group fewer individuals feel adequately informed and a higher share feels not adequately informed and uninterested in the pension system.
- Preferences on pension management and risk:
  - About 45 percent of self-employed opt for managing their savings independently.
  - About 65 percent of employees gladly delegate pension decisions to a fund.
  - When asked whether they would rather pay more premium for a guaranteed pension or a lower premium for a pension whose final payment can vary with investment risk, self-employed expressed a somewhat higher attitude toward risk taking on average.

### Model estimations and stylized facts
- Panel and estimation:
  - Panel comprises answers collected during the years 2014-17 for a total of about 20,000 observations.
  - Model estimated by PROBIT: dependent variable SE (dummy = 1 if self-employed); independent variables DC1 (risk-taking: DNB95=2), DC2 (freedom of choice: DNB96=1); X is a vector of demographic and psychological characteristics.
  - N obs = 1,847; Pseudo R2 = 0.130.
- Stylized determinants of self-employment (literature and expectations cited):
  - Gender: women less likely to enter self-employment (intrinsically more risk averse).
  - Age: positive influence on self-selection into self-employment (accumulated capital, preference for flexibility).
  - Children: may increase likelihood via time flexibility needs but can also increase risk aversion.
  - Education: relationship ambiguous; may be U-shaped depending on industry and managerial ability.
  - Income and wealth: higher income and higher tax burden may increase likelihood to be self-employed (contributions lower); liquidity eases business start-up.
  - Personality traits: risk attitudes, patience, control, mindfulness affect occupational choice; lower risk aversion and higher “control” expected to increase probability of self-employment, “mindful” reduces it.
- Psychological trait survey results (2017):
  - Average score on “risk aversion” is 5 on a range of 1 to 7 based on 6 questions.
  - Self-employed show slightly lower average score on “risk aversion” and higher on “patience” compared with total population.
  - Expectations: lower risk aversion, higher patience and control associated with higher self-employment probability; mindfulness associated with lower probability.

### Regression results (selected coefficients and significance)
- Key reform-preference variables:
  - DC1 (risk taking): 0.019 * (0.010)
  - DC2 (freedom of choice): 0.036 *** (0.010)
- Age groups (reference implied; coefficients shown):
  - <35: -0.020 (0.016)
  - 35-45: (listed as Age group (35-45) header)
  - 46-55: 0.029 ** (0.013)
  - 56-67: 0.041 *** (0.013)
- Gender:
  - Gender (F): 0.015 (0.010)
- Income and tax:
  - Gross income: 0.000 (0.000)
  - Tax burden: -0.020 ** (0.008)
- Children:
  - Children (Y): -0.012 (0.010)
- Education (base = primary and below):
  - Vocational_interm.: 0.063 *** (0.019)
  - Vocational_high: 0.042 ** (0.019)
  - Pre-university: 0.046 ** (0.022)
  - University: 0.062 *** (0.020)
- Information:
  - Informed: 0.002 (0.004) (not significant)
- Psychological traits:
  - Risk averse: -0.004 (0.005) (not significant)
  - Patient: -0.001 (0.006) (not significant)
  - Controlling: 0.015 ** (0.007)
  - Mindful: -0.030 *** (0.011)
- Additional notes:
  - Time dummies, gender, having children, civil status (not reported) and gross income are not significant in the main estimation except where coefficients are shown above.
  - DC1 and DC2 (preferences for greater freedom of choice and risk-taking) are significant and bear the expected positive sign on the probability of being self-employed.

### Interpretation and implications for pension reform and labor choices
- Main empirical result:
  - Individuals who prefer pension plans with more freedom of choice and more investment risk are significantly more likely to choose self-employment than regular employment.
- Potential labor-market responses to pension reforms:
  - If reform decreases pension wealth, or perceived pension wealth, some individuals could increase savings or shift labor relationships (for example, spend more time in contractual employee relationships).
  - Historical precedent: the 2006 reform that abolished preferential tax treatment of early retirement lowered the share of people willing to use liquid assets for entrepreneurship and thus self-employment (Mastrogiacomo and others, 2016).
  - A reform that introduces individual contracts allowing more flexibility should incentivize participation in the pension system and discourage self-employment.
- Policy-relevant observations:
  - Self-employed participate only marginally in occupational pension schemes; about a fifth report job entitlement to a retirement pension versus 84 percent among employees.
  - Self-employed express higher preference for freedom of choice and greater willingness to tolerate risk on final pension incomes.
  - The link between self-employment and support for greater individualization of savings plans appears strong after controlling for observable and unobservable individual characteristics.
  - Perceived information about pension conditions is not a significant predictor of self-employment in the regression, but self-employed report lower receipt of pension overviews (about 40 percent) compared with two thirds of workers.

### Conclusions
- Findings summarized:
  - Preferences for greater freedom of choice and for taking investment risk in pension savings are positively and significantly associated with being self-employed.
  - Self-employment in the Netherlands has increased over the past decade, notably among older workers and those with higher education levels.
  - Self-employed participation in occupational pension schemes is marginal; many self-employed prefer individual management and tolerate higher pension income risk.
  - The current pension reform proposal is broadly comparable to a transition toward a “plain” DC framework but is complex; mapping survey answers into reform support is not unequivocal.
- Policy takeaway:
  - The pension system architecture can influence individual employment choices; reforms that alter pension wealth, perceived pension wealth, or the flexibility/individualization of pension contracts may affect the incidence of self-employment and labor-market composition.
  - Consideration of how pension reform interacts with tax treatment, information provision, and the availability of individually managed yet inclusive pension products is important for labor-market and pension coverage outcomes.

*Source: IMF staff analysis of the Dutch National Bank household survey (DHS) and related text in content unit 1nldea2019002.*

### 30.      Because of the complexity involved in managing new pension options, consensus and

### 1nldea2019002 - 30.      Because of the complexity involved in managing new pension options, consensus and

### Complexity, consensus, and implementation timing
- The PPR-CB may combine the most attractive features of both DB and DC plans while moving a step forward towards greater individualization, but:
  - The reform brings the system closer to individual preferences and increases transparency, which also increases complexity.
  - The new framework will require people to take more active interest in the management of retirement savings, which the majority has not done until now.
  - The learning process may take time; even individuals who prefer freedom of choice may not be ready to exercise it (van Dalen and Henkens, 2017).
  - Maintaining trust in institutions, which is currently at its highest in recent years, will be crucial for reaching a consensus and navigating the transition successfully (Figure 1, Appendix I).

### Political economy and information gaps
- Greater transparency on the “missing arguments” would help build consensus and speed up reform implementation.
- The political economy literature offers approaches for overcoming reform resistance through elimination, compensation and information building (Tsebelis, 2000; Tsebelis and Hahm, 2014; Tompson and Price, 2009; James and Brooks, 2001).
- Specific informational gaps to address:
  - The redistribution cost in the transition to the new contract type should be made explicit.
  - The modalities of financing the transition costs should be specified.
  - The reform should clearly identify the losers and the winners, highlighting options for compensation in a wider reform package.
- Answering these open questions is presented as key to reaching a reform consensus in an inclusive and equitable consultative society.

### Key statistics and parameters (Table 1, 2018 figures and related)
- Pensionable age: increasing to 67 by 2021, then linked to life expectancy
- Pensionable age (reference): 68 in 2018
- Accrual rate, annual: 1.875
- Maximum (in eur, 2017): 103,317
- Average (in eur, 2017): 14,737
- Contributions (gross, billions): 31.2
- Contributions: employers 17, workers 7
- Contributions as Percent of GDP: 4.4
- Size of fund (in 2017) Billions eur: 1620
- Size of fund (in 2017) Percent of GDP: 220
- Non-indexed funding ratio (2017): 103/105 (regulatory min)
- Indexation post-retirement: minimum wages / CPI
- Source cited for table: OECD Aging Report

### Trust in institutions (Figure 1, 2000–18)
- The text emphasizes that trust in institutions is at its highest in recent years and that consumer confidence in the banking sector and employers is also high and rising (Banking Confidence Monitor 2017, The Dutch Banking Association; Statista 2018, The Edelman Trust Barometer, 2018).

### Policy recommendations and implementation priorities
- Build and preserve public trust in institutions to enable consensus and smooth transition.
- Provide greater transparency and targeted information to address the “missing arguments”:
  - Quantify redistribution costs of the transition.
  - Specify financing modalities for transition costs.
  - Identify losers and winners and design compensation options within a broader reform package.
- Use political economy tools (elimination, compensation, information building) to reduce reform resistance.

### Appendix I & II — Data sources and survey instruments (high-level)
- Appendix I: Table 1 (Key Parameters of the Dutch Pension System, 2018) and Figure 1 (Trust in Institutions, 2000–18) are referenced as data inputs.
- Appendix II: The DNB Households Survey (DHS) — key variables described include:
  - Dependent variables:
    - DC1 = 1 if individual willing to take a risk related to pension income in exchange for paying a lower premium (from dnb95).
    - DC2 = 1 if respondent expressed a preference for managing pension savings individually (from dnb96).
  - Demographic and socioeconomic variables:
    - age (constructed from birth year; individuals older than 67 dropped)
    - educational dummies (including “low”)
    - Female (gender dummy)
    - Children (dummy)
    - Married (registered partnership dummy)
    - Health (step dummy: 1 = excellent, 2 = good, 3 = fair, 4 = not so good, 5 = poor)
    - Informed (step dummy: 1 to 6; lower value = better informed)
    - tax (implied overall tax burden = (gross - net) / net)
    - income measures: total gross (btot) and total net (ntot)
  - Employment status dummies:
    - Employee, Self-employed, Unemployed, Other work
  - Psychological traits variables (survey scales and construction notes):
    - Risk averse: mean of values (between 1 and 7) on 6 questions
    - Patient: mean of values (between 1 and 7) on 12 questions
    - Controlling: mean of values (between 1 and 7) on 13 questions
    - Mindful: mean of values (between 1 and 5) on 18 questions
  - DHS question examples provided for dnb94, dnb95, dnb96, dnb207a, dnb207b, dnb116, dnb210 (text of response options preserved in the source).

*Source: 1nldea2019002 - 30. Because of the complexity involved in managing new pension options, consensus and (IMF PDF content).*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2019/1nldea2019002.pdf_