## _wp1261

## Source details

**Canonical URL:** [_wp1261](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp1261.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp1261.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp1261.pdf.json)

---

### Abstract and central question
- Examines the contribution of “structural factors” to current account imbalances for a panel of countries over 1975–2009.
- Uses standard empirical current account models augmented with structural policy variables.
- Principal finding: commonly recommended structural factors cannot explain the widening of imbalances prior to the 2008–09 crisis, though structural factors explain some long-standing cross-country differences in current account levels.

### Data, method, and scope
- Panel: 106 advanced, emerging and developing countries; OECD sample included 27 countries.
- Time frame: 1975–2009, using five-year non-overlapping averages (1975–79, 1980–84, 1985–89, 1990–94, 1995–99, 2000–04, 2005–09).
- Baseline controls: income growth and level, population age structure (old/young dependency ratios and change in old dependency), fiscal balance, initial net foreign asset position, financial integration (sum of foreign assets and liabilities as percent of GDP), trade openness, contemporaneous oil price for oil producers.
- Structural policy indicators examined: credit market regulation (four components: degree of public ownership of the banking system, control of interest rates, percentage of credit extended to private sector, competition from foreign banks), corporate income tax, doing-business paying-taxes rank, labor tax wedge, ratio of minimum wage to mean wage, unemployment gross replacement rate, employment protection, product/services market regulation, regulation in professional services/retail/energy/transport/communication.
- Estimation: random effects model (primary) and OLS with cluster robust standard errors (robustness). Many explanatory variables entered as deviations from the PPP-weighted sample average; several variables timed as of the year preceding the five-year period.

### Key empirical findings (aggregate)
- Fundamentals and timing:
  - Standard fundamentals (income, demographics, fiscal balance, initial net foreign assets, financial integration) did not evolve significantly during the final years before the crisis and therefore cannot explain the emergence of global imbalances in that period.
  - Residuals from the baseline current account regression largely mirror the imbalances that emerged in the mid-2000s, implying global/cyclical forces were important.
- Structural policy time-series behavior:
  - Structural policy variables changed little over time or moved similarly in surplus and deficit countries; unlikely that structural factors on their own explain the emergence of imbalances prior to the crisis.
  - Notable exception: China — substantial credit market deregulation in 2005 and substantial increase in ratio of minimum-to-mean wage in 2002 (both expected to reduce China’s current account surplus).
- Cross-country correlations (long-standing differences):
  - Stricter credit market regulation tends to be associated with higher current account balances (index interpretation noted: Fraser Institute index ranges 0–10; higher = less regulation).
  - Higher corporate income tax associated with higher current account balances.
  - Higher unemployment gross replacement rate tends to be associated with higher current account balances.
  - Lower ratio of minimum wage to mean wage is associated with higher current account balances; ratio of minimum-to-mean wage typically has a negative coefficient on current account (i.e., higher ratio → lower current account).
  - Less strict employment protection and lower minimum wage are associated with larger current account surpluses in several specifications.
- Robustness and sample differences:
  - Many structural-variable correlations are not robust across samples and time periods; coefficients often insignificant in the OECD-only sample.
  - Overall model explains about 35 percent of variation in current account balances (between R-square around 0.5 for cross-country variation).

### Interaction effects and amplification role of structural policies
- Structural policies may act as absorbers or amplifiers of shocks, but evidence suggests they account for only a small fraction of the imbalances even in this moderating role.
- Selected interaction results:
  - Ratio of minimum-to-mean wage * Previous period growth: negative and significant in many specifications (e.g., -0.0119**; -0.0109**; -0.0110**; -0.0136***; -0.0199** in Table 6 interaction terms), implying higher minimum wage in fast-growing countries may raise labor costs, encourage capital-for-labor substitution, increase investment, and lower the current account.
  - Gross replacement rate * Previous period growth: negative in several specifications (e.g., -0.0175**; -0.0166**; -0.0257**), indicating the positive effect of unemployment replacement on current account is moderated by rapid prior growth.
  - Financial integration * Previous period growth: consistently negative and often significant (examples: -0.0016***; -0.0015***; -0.0014***; -0.0017***), indicating financial integration amplifies the effect of prior growth on lowering current account balances.
- Diagnostic: Residuals from regressions that include interaction terms remain aligned with observed imbalances (though closer to zero than baseline for most countries), reinforcing that structural policies cannot fully account for the emergence of imbalances.

### Specific documented coefficient highlights (selected statistically significant estimates reported in tables)
- Log of GDP per capita: coefficients reported across tables include 0.0247***; 0.0236***; 0.0265***; 0.0211***; 0.0176** and multiple other positive, often highly significant estimates.
- Fiscal balance to GDP: examples include 0.3853***; 0.3782***; 0.3790***; 0.1882**; 0.2325*** and other positive, often highly significant estimates.
- Net foreign assets to GDP: examples include 0.0146**; 0.0197***; 0.0194***; 0.0253***; 0.0248** and other positive significant estimates (Table 1–3).
- Old dependency ratio: strongly negative in many specifications, e.g., -0.3226***; -0.3400***; -0.3744***; -0.3810*** and larger negative values in other tables.
- Increase in old dependency ratio over 5 years: positive and often significant, e.g., 0.7330***; 0.6511**; 0.6117**; 0.5964**; 1.0855***.
- Contemporaneous oil price * Oil producer: small positive coefficients reported (e.g., 0.0005**; 0.0004**; 0.0005**; 0.0006***).
- Financial integration * Previous period growth: negative and often significant (examples: -0.0010**; -0.0013***; -0.0011***; -0.0011***).
- Credit Market Regulation (where significant): negative coefficients reported (e.g., -0.0047**; -0.0081***; -0.0080***; -0.0059**; -0.0059***).
- Gross replacement rate (where significant): positive coefficients reported (e.g., 0.0572**; 0.0971***; 0.0789**; 0.1130***; 0.0955***).
- Ratio of minimum wage to mean wage (where significant): negative coefficients reported (e.g., -0.0400**; -0.0399**; -0.0348**; -0.0344**).
- Employment protection index (where significant): negative coefficients in some specifications (e.g., -0.0125**; -0.0076**).

### Interpretation: cyclical vs structural drivers
- A significant fraction of the imbalances in the run up to the crisis reflected the global cycle rather than changes in structural policies or fundamentals.
- Example cited: Germany’s surge in current account surplus to 7½ percent of GDP in 2007 partly attributed to global demand for technology-intensive goods and windfall export profits that did not immediately translate into higher domestic investment.
- Empirical diagnostics show residuals from regressions (with and without structural variables) largely mirror mid-2000s imbalances, reinforcing a prominent role for cyclical/global demand factors and asset/sectoral booms (housing, stock markets).

### Policy implications (general)
- Structural reforms are not uniformly expected to reduce current account surpluses; effects are selective and context-dependent.
- Policies that empirical correlations indicate could reduce a current account surplus include:
  - Lower business taxation (corporate income tax).
  - Reduced credit market regulation (notably smaller public share in the banking system).
  - Reduction in gross unemployment replacement rate.
- Policies that empirical correlations suggest could increase the surplus (trade-offs exist):
  - Lower minimum wage and lower employment protection (labor-market liberalizing measures) are associated in several specifications with larger current account surpluses.
- Recommendation: structural policy measures should be selective and tailored to country-specific circumstances; a broad universal package is unlikely to be optimal.
- Trade-offs: policy choices should weigh other objectives (inequality, fiscal sustainability, labor-market outcomes); some measures that reduce a surplus may conflict with other policy goals.

### Implications for Germany (selected results and numerical estimates)
- Germany’s current account surplus reached 7½ percent of GDP in 2007 (historical high).
- Drivers noted: expansion of global trade benefiting Germany’s comparative advantage in specialized product varieties; corporate sector large contributor to surplus; corporate profits did not translate into increased domestic investment.
- Estimated effect of a policy package for Germany (reported changes in current account, percent of GDP):
  - Credit Market Deregulation to OECD average: -0.5
  - Reduction in taxes and simplification of tax procedures to US rank: -0.3
  - Reduction in gross unemployment replacement ratio to OECD average: -0.4
  - Total: -1.2
- Paper text: suggests a package could reduce the surplus by about 1¼ percent of GDP (noting that "the impact of these structural policies on the surplus will likely be modest").
- Practical caveats for Germany:
  - Lower taxes on businesses, further reduction in gross unemployment replacement rate, and a smaller public share in the banking system could reduce the surplus but the aggregate impact is likely modest.
  - Raising minimum wage and strengthening employment protection would tend to reduce the surplus but may raise unemployment, making such measures potentially undesirable.
  - Reductions in taxes and unemployment benefits should not jeopardize long-term fiscal sustainability goals.

### Literature context and comparisons (selected)
- Confirms large literature finding fundamentals (relative per capita income, fiscal stance, demographics, oil prices, initial net foreign assets, financial integration) are important determinants of current account.
- Notes mixed and scarce evidence on direct impact of structural policies:
  - Kennedy and Sløk (2005): no robust link in pooled regressions for 14 OECD countries; some evidence that more open product and financial markets associated with weaker current accounts.
  - Kerdain, Koske and Wanner (2010): structural policies influence savings and investment via productivity and public finances; social spending associated with lower savings and current account; various mixed effects.
  - Ju and Ii (2007): rigid labor markets reduce speed of current account adjustment to equilibrium.
  - Jaumotte and Sodsriwiboon (2010): financial liberalization and higher minimum wage lower the current account; no direct link for level of employment protection or unemployment benefits.

### Data notes and measurement highlights
- Structural/regulatory indicators: credit market regulation (Fraser Institute index, 0–10 higher = less regulation), gross unemployment replacement rate (Aleksynska and Schindler), ratio of minimum wage to mean wage (Aleksynska and Schindler), employment protection (OECD for OECD countries; out-of-sample forecast for others), corporate income tax (OECD / IMF Fiscal Affairs), doing-business paying-taxes rank (World Bank), labor tax wedge (OECD), product market regulation and sectoral regulation indicators (OECD, assigned to five-year periods).
- Dependent variable: Current Account to GDP (5-year average) from AMECO or IMF WEO where AMECO not available.
- Many variables averaged over five-year non-overlapping periods or measured as of the year before the period start; many entered as deviations from PPP-weighted sample average.

*Source: IMF Working Paper WP/12/61 (sections 1–5 of _wp1261).*

### Section 1

### Current Account Imbalances: Can Structural Policies Make a Difference? — Section 1

### Abstract and central question
- Examines the contribution of “structural factors” to current account imbalances for a panel of countries over 1975–2009.
- Uses standard empirical current account models augmented with structural policy variables.
- Principal finding: commonly recommended structural factors cannot explain the widening of imbalances prior to the 2008–09 crisis, though structural factors explain some long-standing cross-country differences in current account levels.

### Data, method, and scope
- Panel: 106 advanced, emerging and developing countries.
- Time frame: 1975–2009, using five-year non-overlapping averages.
- Baseline controls include: income growth and level, population age structure, fiscal balance, initial net foreign asset position, and degree of financial integration.
- Structural policy indicators examined include: credit market regulation (four components: degree of public ownership of the banking system, control of interest rates, percentage of credit extended to private sector, competition from foreign banks), taxes on business, minimum wage, unemployment benefits, employment protection, product/services market liberalization, and other labor market rigidities.

### Key empirical findings
- Standard fundamentals (income, demographics, fiscal balance, initial net foreign assets, financial integration) did not evolve significantly during the final years before the crisis and therefore cannot explain the emergence of global imbalances in that period.
- Structural policy variables changed little over time or moved similarly in surplus and deficit countries, so they also explain very little of the emergence of imbalances prior to the crisis.
- Some structural factors correlate with long-standing differences in current account balances:
  - Countries with stricter credit market regulation tend to have higher current account balances.
  - Higher taxes on businesses are associated with higher current account balances.
  - Lower minimum wage is associated with higher current account balances (especially in slow growing economies).
  - More generous unemployment benefits tend to be associated with higher current account balances.
  - Less strict employment protection and lower minimum wage are associated with larger current account surpluses (consistent with some earlier studies).
- Even where correlations exist, results are often not robust across country samples and time periods; some recommended policies increase and some reduce current account balances.

### Interpretation and role of cyclical factors
- A significant fraction of the imbalances in the run up to the crisis reflected the global cycle rather than changes in structural policies or fundamentals.
- Example: Germany’s surge in current account surplus to 7½ percent of GDP in 2007 is interpreted in part as benefiting from global demand for technology-intensive goods where Germany has comparative advantage; windfall export profits did not translate immediately into higher domestic investment as firms viewed the boom as temporary and Germany’s perceived growth potential remained low.

### Interaction effects and amplification role of structural policies
- Structural policies may act as absorbers or amplifiers of changes in fundamentals, but evidence suggests they account for only a small fraction of the imbalances even in this moderating role.
- The impact of structural measures appears country-specific; a tailored approach is preferable to a broad universal package.

### Policy implications (general and for Germany)
- Policies that could reduce a current account surplus (per the empirical correlations found) include:
  - Lower business taxation.
  - Reduced credit market regulation (notably smaller public share in the banking system).
  - Further reduction in the gross unemployment replacement rate.
- For Germany specifically:
  - Lower taxes on businesses, further reduction in the gross unemployment replacement rate, and a smaller public share in the banking system could reduce the surplus, but the aggregate impact is likely modest.
  - Raising the minimum wage and strengthening employment protection would tend to reduce the surplus but may raise unemployment, making such measures potentially undesirable.
- For some European peripheral economies, reducing minimum wage and lowering employment protection could contribute to a reduction in current account deficits.

### Literature context and comparisons
- Confirms the large literature finding that fundamentals (relative per capita income, fiscal stance, demographics, oil prices, initial net foreign assets, financial integration) are important determinants of the current account.
- Notes mixed and scarce evidence on the direct impact of structural policies on the current account; prior studies show:
  - Kennedy and Sløk (2005): no robust link between specific structural policies and the current account in pooled regressions for 14 OECD countries; some evidence that more open product and financial markets are associated with weaker current accounts.
  - Kerdain, Koske and Wanner (2010): structural policies influence savings and investment via productivity and public finances; social spending (notably healthcare) associated with lower savings and lower current account; stricter employment protection linked to lower savings when unemployment benefits are low and higher investment (leading to lower current account); product market liberalization temporarily boosts investment; financial deregulation may lower savings in less developed countries; direct impact on current account often not detectable.
  - Ju and Ii (2007): rigid labor markets reduce the speed of adjustment of the current account to long-run equilibrium (two-step approach; large economies excluded).
  - Jaumotte and Sodsriwiboon (2010): financial liberalization and higher minimum wage lower the current account; no direct link detected for level of employment protection or level of unemployment benefits.

### Overall conclusion from Section 1
- Structural policies are not the primary driver of the widening global imbalances before the 2008–09 crisis; cyclical/global demand factors played a larger role.
- Still, certain structural factors are correlated with persistent cross-country differences in current account balances; policy interventions should be selective and tailored to country circumstances rather than broad packages.

*Prepared by Anna Ivanova; IMF Working Paper WP/12/61 — Section 1*

### Section 2

### _wp1261 - Section 2

### Literature and background findings
- Bayoumi et al (2010) — econometric study covering 100 advanced, emerging and developing countries for the period 2001–09 (annual data) — find that countries with more (less) credit market regulations have higher (lower) current account balances while controlling for traditional fundamentals.
- Berger and Nitsch (2010) — sample of 18 European countries over 1948 through 2008 — find that countries with less flexible labor and product markets exhibit systematically lower bilateral trade surpluses.
- Studies identifying “excessive” pre-crisis imbalances:
  - Barnes, Lawson, and Radziwill (2010) — sample of 25 OECD countries.
  - Lane and Milessi-Ferretti (2011) — sample of 65 advanced and emerging economies; conclude countries with the largest excesses before the 2008–09 crisis experienced the largest corrections thereafter, with adjustment in deficit countries achieved primarily through demand compression rather than expenditure switching; high output costs of rapid corrections support research into distortions not internalizing sudden-stop risk.
  - Chinn, Eichengreen, and Ito (2011) — sample of 109 industrial and developing countries.
- Barnes, Lawson, and Radziwill (2010) and Chinn, Eichengreen, and Ito (2011) find that housing investment, real housing appreciation and stock market performance partly explain excesses, but large residuals remain, notably for the U.S. and China.
- Theoretical note (Vogel 2011): supply-side structural policies can regain competitiveness short run but income effects can offset this as imports rise, implying lasting long-term rebalancing also requires correction of demand imbalances.

### Contribution of the paper
- Three dimensions of contribution:
  - Investigates the role of structural policies in the emergence of imbalances in the run up to the 2008–09 financial crises.
  - Assesses the direct impact of a commonly recommended package of structural policies on the current account across a large sample of advanced, emerging and developing countries while controlling for traditional macroeconomic fundamentals.
  - Assesses potential size of current account reduction due to these policies in Germany.
- Overall findings: support some earlier results but highlight lack of robustness of many determinants of the current account; emphasize muted role of structural factors in causing the growth of imbalances just prior to the recent crisis; for Germany, offers policy directions but cautions quantitative effects may be small.

### Baseline model (Section III)
- Estimation approach:
  - Random effects model on a sample of 106 advanced, emerging and developing countries.
  - OLS with cluster robust standard errors used as robustness check (yields similar results).
  - Current account averaged over five-year non-overlapping periods spanning the period of 1975–2009 to capture a medium-term “structural current account.”
  - Many explanatory variables entered as deviations from the PPP-weighted sample average in a given period.
- Main baseline results:
  - Higher relative income per capita, fiscal balance, initial net foreign assets position, and higher oil prices for oil producers are associated with higher current account balances.
  - Countries with relatively high current dependency ratios have lower current account balances; countries with higher expected increase in the dependency ratio (speed of aging) are found to have higher current account surpluses.
  - Financial integration (sum of foreign assets and liabilities in percent of GDP) reduces the current account balance in countries with higher previous-period growth; link works more robustly through growth than income level.
- OECD-specific differences:
  - Relationship between income-per-capita, fiscal balance, net foreign assets/GDP, old dependency ratio, and interaction of financial integration and past growth with the current account broadly unchanged, though some coefficients become insignificant.
  - Young dependency ratio coefficient becomes positive and significant in OECD sample (consistent with Kerdrain, Koske, and Wanner (2010) and Barnes, Lawson and Radziwill (2010)).
  - Trade openness matters more in OECD sample: higher trade openness associated with higher current account surplus.
- Model fit:
  - Explains about 35 percent of the variation in current account balances in the sample.
  - Between R-square of 0.5, indicating better cross-country variation explanation.
- Important diagnostic:
  - Residuals from the current account regression largely mirror the imbalances that emerged in mid-2000s, implying fundamentals did not evolve to generate the imbalances and suggesting global forces were at work.

### Structural policies and the current account (Section IV)
- Structural policy variables examined: regulation, taxation, minimum wage level, unemployment benefits (estimated as five-year averages).
- Possible roles of structural policies:
  - Direct impact on current accounts (e.g., high business taxation → reduced investment → lower investment and higher current account balances).
  - Shaping the response of the current account to macro fundamentals and shocks (acting as shock absorbers or amplifiers).
  - Explaining cross-country levels of current accounts even if not driving imbalances over time.
- Time variation and trends:
  - Many structural indicators did not change substantially over time during the emergence of global imbalances; when they did, changes often occurred in same direction for surplus and deficit countries.
  - China notable exception: substantial credit market deregulation in 2005 and substantial increase in ratio of minimum to mean wage in 2002 (both expected to reduce China’s current account surplus).
  - Conclusion: unlikely that structural factors on their own explain the emergence of imbalances.
- Regression results when augmenting baseline with structural indicators (full sample and OECD sample):
  - Generally no robust relationship between current account and structural policies.
  - In some full-sample specifications:
    - Unemployment gross replacement ratio coefficient is positive and significant.
    - Ratio of minimum-to-mean wage coefficient is negative and significant.
    - Employment protection indicator coefficient is negative and significant.
  - No significant association found for OECD countries (sample small).
- Interpretations of specific associations:
  - Positive association: gross unemployment replacement rate → higher current account possibly via higher precautionary savings due to higher unemployment risk and rates (Bassani and Duval 2006, Vandenberg 2010). Countervailing effect could be higher income during unemployment raising consumption, but historical evidence suggests the precautionary saving effect dominated.
  - Negative association: higher ratio of minimum-to-mean wage → higher labor costs → lower competitiveness, which can reduce corporate profitability and savings but may also encourage capital substitution for labor.
  - Negative association: higher employment protection → lower current account, consistent with literature that higher employment protection reduces savings and increases investment via higher implicit/explicit labor costs.
- Diagnostic:
  - Residuals from regression including three structural variables (unemployment gross replacement rate, ratio of minimum-to-mean wage, employment protection index) continue to mirror the imbalances; structural factors on their own did not evolve to generate imbalances.
- Note on interactions:
  - Hypothesis that structural factors shaped response to macro shocks tested in Section VI via interactions with dynamic fundamentals (introduced here but analyzed later).

### Long-standing structural differences and the current account (Section V)
- Approach:
  - Constructed country-level long-term structural indicators by averaging structural variables over all available years to capture long-standing characteristics; regressions control for fundamentals and enter structural variables as deviations from PPP GDP-weighted sample mean to capture relative standing.
- Key robust findings:
  - Ratio of minimum-to-mean wage: negative sign and significant in almost all specifications.
  - Unemployment gross replacement rate: positive impact fairly robust.
- Other structural indicators:
  - Impacts not robust across global and OECD samples.
  - Some commonly recommended policy changes have mixed predicted effects on current account:
    - Lower business taxation, lower credit market regulation, and lower unemployment benefits can reduce current account surplus.
    - Lower minimum wage and lower employment protection, often recommended to make labor markets more flexible, are associated with larger current account surpluses.
- Policy implication highlighted:
  - Structural reforms are not uniformly expected to reduce current account surpluses; some labor-market liberalizing measures (lower minimum wage, lower employment protection) could, according to these results, increase surpluses, while other measures (lower business taxation, deregulation of credit markets, lower unemployment benefits) could reduce surpluses.

*Source: _wp1261 - Section 2 (PDF).*

### Section 3

### _wp1261 - Section 3

### Major empirical findings on structural policies and the current account
- Little evidence that the commonly recommended package of structural policies (financial regulation, tax policy, labor market flexibility) contributed substantially to the emergence of global imbalances; large imbalances likely reflected mainly a booming world economy.
- Structural factors help explain long-standing cross-country differences in current account levels, but effects are selective rather than broad and diffuse.
- Evidence that stricter credit market regulation (index includes public ownership of the banking system, interest rate controls, percentage of credit extended to private sector, and competition from foreign banks) is associated with higher current account balances.
- Countries with higher taxes on businesses, generous unemployment benefits, lower minimum wage, and less strict employment protection tend to have higher current account balances.
- The relationship between structural variables and the current account is not always robust across samples; results for OECD sample are generally not significant and should be interpreted with caution due to a relatively small number of observations.
- Interaction results:
  - The negative impact of the minimum-to-mean wage ratio on the current account is stronger in countries that experience rapid income growth; suggests higher minimum wage in fast-growing countries may raise labor costs, encourage capital-for-labor substitution, increase investment, and lower the current account.
  - The positive impact of the gross unemployment replacement rate on the current account is moderated by previous period per capita income growth (reduced in rapidly growing countries) and strengthened by a higher initial net foreign assets position.
  - Most interaction findings are not robust across specifications; indirect impact of structural policies on the current account is inconclusive.
- Residuals from the regression with interaction terms track imbalances and, though closer to zero than in the baseline model for all countries except Japan, indicate structural policies cannot fully account for the emergence of imbalances.
- Cyclical factors likely played a role in the emergence of imbalances; correlation exists between “excess imbalances” and housing investment/housing real price and stock market performance.

### Specific results on finance and regulatory measures
- Two indicators that became significant in the overall sample when structural variables enter as averages over time: corporate income tax rate/indicator of doing business paying taxes and credit market regulation (higher value = less regulation).
- Countries with a long-standing tradition of relatively high business taxes are on average found to have higher current account balances; possible mechanism: higher corporate taxation reduces investment incentives, raising the current account balance.
- Credit market regulation index results suggest stricter credit market "regulation" raises the current account, potentially via constrained investment and/or higher household and corporate savings; result may reflect broader banking sector effectiveness/efficiency rather than just credit to private sector.
- Inclusion of financial development (ratio of private credit to GDP at start of period) and capital account openness (Chinn and Ito measure) does not significantly alter conclusions for most structural variables; credit market regulation coefficient becomes insignificant when capital account openness included, suggesting overlap in captured aspects of credit availability.

### Implications and policy trade-offs
- Some commonly proposed structural policies would reduce the current account, while others would increase it; policy choice should consider other objectives (e.g., inequality, fiscal sustainability).
- Structural policy recommendations should be selective and tailored to country circumstances rather than applied as a broad package.

### Implications for Germany
- Germany’s current account surplus reached a historical high of 7½ percent in 2007; improvement driven by trade balance on goods and expansion of global trade.
- Germany’s export competitiveness stems from a comparative advantage in a large number of specialized product varieties; exports grew mainly due to world trade expansion rather than increasing market share.
- Imports increasingly tilted toward products produced most cost-effectively by China; German imports from China grew rapidly from a low base.
- German corporate sector was the largest contributor to the increased current account surplus; corporate sector did not match a substantial increase in profits with increased domestic investment despite consistently low investment rates relative to European peers.
- Possible explanations for low corporate investment include long-standing low returns to investment and lack of confidence that the boom would last; estimated potential growth in Germany remained low (close to 1 percent) in those years.
- Estimated effect of a policy package for Germany (reported changes in current account, percent of GDP):
  - Credit Market Deregulation to OECD average: -0.5
  - Reduction in taxes and simplification of tax procedures to US rank: -0.3
  - Reduction in gross unemployment replacement ratio to OECD average: -0.4
  - Total: -1.2
- The paper suggests a package including scaling down public provisioning of banking services (allow larger private sector participation), reduction in unemployment benefits towards the OECD average, and reduction and simplification in business taxes to move Germany to the U.S. rank in the doing business survey could reduce the surplus by about 1¼ percent of GDP; reductions in taxes and unemployment benefits should not jeopardize long-term fiscal sustainability goals.

### Conclusions
- Structural policies only partially account for the emergence of imbalances and mainly help explain persistent cross-country differences in current account levels.
- Policy prescriptions to address imbalances should focus on select structural measures tailored to country circumstances and weigh trade-offs with other policy objectives.

*Source: _wp1261 - Section 3*

### Section 4

### _wp1261 - Section 4

### Summary statement
- "However, altogether, the impact of these structural policies on the surplus will likely be modest and a broader strategy for raising potential growth and raising domestic consumption and investment in the medium term would be essential."

### Key empirical findings (selected statistically significant coefficients from tables)
- Log of GDP per capita:
  - 0.0247***; 0.0236***; 0.0265***; 0.0211***; 0.0176** (Table 1)
  - 0.0264***; 0.0205***; 0.0206***; 0.0157***; 0.0158***; 0.0187***; 0.0191***; 0.0130; 0.0263*** (Table 2)
  - 0.0215***; 0.0149***; 0.0143**; 0.0113**; 0.0111**; 0.0146***; 0.0153***; 0.0113; 0.0202** (Table 3)
- Fiscal Balance to GDP:
  - 0.3853***; 0.3782***; 0.3790***; 0.1882**; 0.2325*** (Table 1)
  - 0.3941***; 0.3123***; 0.3369***; 0.3202**; 0.3201**; 0.2954**; 0.2925**; 0.1979**; 0.2683 (Table 2)
  - 0.3643***; 0.3084***; 0.3418***; 0.3219**; 0.3222**; 0.2965**; 0.2910**; 0.1589*; 0.2666 (Table 3)
- Net foreign assets to GDP:
  - 0.0146**; 0.0197***; 0.0194***; 0.0253***; 0.0248**; 0.0229**; 0.0282**; 0.0229**; 0.0196 (Table 1)
  - 0.0211***; 0.0194**; 0.0193**; 0.0228**; 0.0226**; 0.0277***; 0.0283***; 0.0209; 0.0611*** (Table 2)
  - 0.0339***; 0.0347***; 0.0354***; 0.0383***; 0.0386***; 0.0431***; 0.0429***; 0.0338***; 0.0661*** (Table 3)
- Old dependency ratio:
  - -0.3226***; -0.3400***; -0.3744***; -0.3810*** (Table 1)
  - -0.3481***; -0.3966***; -0.4024***; -0.4652***; -0.4657***; -0.4644***; -0.4671***; -0.0167; -0.4961*** (Table 2)
  - -0.3061***; -0.3373***; -0.3240***; -0.4374***; -0.4378***; -0.4487***; -0.4504***; -0.0533; -0.6214*** (Table 3)
- Increase in the old dependency ratio over 5 years:
  - 0.7330***; 0.6511**; 0.6117**; 0.5964**; 0.2593; 0.6761***; 0.8173**; 0.6761***; 1.0855*** (Table 1)
  - 0.6131**; 0.6208**; 0.6741**; 1.0845***; 1.0905***; 0.9556***; 0.9197***; 0.6809*; 1.6603*** (Table 2)
  - 0.3301; 0.4243; 0.4938; 0.8303**; 0.8340**; 0.7158**; 0.6772*; 0.4677; 1.5275*** (Table 3)
- Contemporaneous oil price * Oil producer:
  - 0.0005**; 0.0004**; 0.0005**; 0.0006*** (Table 1)
  - 0.0004**; 0.0005**; 0.0005**; 0.0005**; 0.0005**; 0.0005**; 0.0005**; 0.0000; 0.0004* (Table 2)
  - 0.0005**; 0.0006***; 0.0006***; 0.0005**; 0.0005**; 0.0005**; 0.0005**; 0.0002; 0.0005** (Table 3)
- Financial integration and interaction with prior growth:
  - Financial integration: examples include 0.0034; 0.0035*; 0.0031; 0.0032* (Table 1)
  - Financial integration * Previous period growth: -0.0010**; -0.0013***; -0.0011***; -0.0011*** (Table 1)
  - Financial integration: 0.0034; 0.0032; 0.0030; 0.0016; 0.0015; 0.0023; 0.0027; 0.0098; -0.0002 (Table 2)
  - Financial integration * Previous period growth: -0.0011***; -0.0013***; -0.0014***; -0.0013**; -0.0013**; -0.0013***; -0.0014***; 0.0012; -0.0006 (Table 2)
  - Financial integration: 0.0061***; 0.0047***; 0.0048***; 0.0044*; 0.0043*; 0.0050*; 0.0054**; 0.0122; 0.0027 (Table 3)
  - Financial integration * Previous period growth: -0.0016***; -0.0015***; -0.0016***; -0.0017***; -0.0017***; -0.0018***; -0.0018***; 0.0009; -0.0014** (Table 3)
- Credit Market Regulation (where significant):
  - -0.0047**; -0.0081***; -0.0080***; -0.0059**; -0.0059***; -0.0070***; -0.0067*** (Table 2)
  - -0.0039**; -0.0069***; -0.0066**; -0.0050**; -0.0052**; -0.0064***; -0.0061*** (Table 3)
- Gross replacement rate (where significant):
  - 0.0572**; 0.0421; 0.0314; 0.0816* (Table 1)
  - 0.0971***; 0.0998***; 0.0929**; 0.0945**; 0.1130***; 0.1007**; 0.0511; 0.1128* (Table 2)
  - 0.0789**; 0.0760**; 0.0768**; 0.0798**; 0.0955***; 0.0837*; 0.0492; 0.1009* (Table 3)
- Ratio of minimum wage to mean wage (where significant):
  - -0.0169; -0.0271*; -0.0169; -0.0194 (Table 1)
  - -0.0400**; -0.0399**; -0.0325*; -0.0328*; -0.0122; -0.0669** (Table 2)
  - -0.0348**; -0.0344**; -0.0287*; -0.0293*; -0.0089; -0.0684** (Table 3)
- Employment protection index (where significant):
  - -0.0125** (Table 1)
  - -0.0004; -0.0048 (Table 2)
  - -0.0010; -0.0043 (Table 3)

### Cross-cutting interpretations implicit in the tables
- Positive associations with current account to GDP:
  - Higher log of GDP per capita (multiple models).
  - Stronger fiscal balance (Fiscal Balance to GDP coefficients consistently positive and often highly significant).
  - Higher net foreign assets to GDP.
  - In several specifications, gross replacement rate is positively associated with higher current account balances.
  - Contemporaneous oil price for oil producers has a small but positive coefficient (e.g., 0.0005**, 0.0006***).
- Negative associations with current account to GDP:
  - Higher old dependency ratio (large and consistently negative coefficients).
  - Stricter credit market regulation in many pooled specifications (negative and significant coefficients).
  - Higher ratio of minimum wage to mean wage in many specifications (negative and often significant).
  - Interaction term: Financial integration * Previous period growth is consistently negative and often significant (e.g., -0.0016***, -0.0015***), indicating financial integration amplifies the effect of prior growth on lowering current account balances.
- Demographic dynamics:
  - Increase in the old dependency ratio over 5 years often has a positive coefficient (e.g., 0.7330***; 1.0855***), indicating transitional demographic changes affect current account balances differently than level of old dependency ratio.

### Data scope and modeling notes (as reported in tables)
- Dependent variable: Current Account to GDP (5-year average).
- Sample periods and models include: 1975-2009, subperiods 1975-1994 and 1995-2009, total sample, OECD sample.
- Estimation approaches include random effects models with robust standard errors and OLS with cluster robust standard errors.
- Observations and country counts reported in tables vary by specification, e.g.:
  - Observations: 548; 532; 426; 400; 323; 349; 160; 157; 115; 160 (various tables).
  - Number of countries: 106; 101; 77; 60; 65; 34; 27; 26; 19 (various tables).
- Variable measurement notes (from table footnotes):
  - 1/ Deviation from US level in a given year
  - 2/ At the beginning of the period, for example for a 5-year period covering 2005-2009, 2004 value was used.
  - 3/ 5-year period average
  - 4/ Deviation from a PPP GDP-weighted sample average
  - 5/ Financial Integration is one year before the beginning of a given 5-year period; growth is the average over the previous 5-year period
  - 6/ Structural variable are country averages over all available years in a given period
  - 7/ Gross replacement rate is the average over 2 years of unemployment
  - 8/ For OECD countries OECD employment protection index was used. For a broader sample an index was constructed as an out-of-sample forecast from the regression of the employment protection index on advance notice period and severance pay after 9 months.

*Source: _wp1261 - Section 4*

### Section 5

### _wp1261 - Section 5

### Regression findings: Determinants of Current Account to GDP (5-year average)
- Dependent variable: Current Account to GDP (5-year average).
- Sample periods reported: 1975-2009 (three specifications), 1975-1994, 1995-2009.
- Key estimated coefficients (coefficient [t-statistic]) from Table 6 (random effects model with robust standard errors; structural variables are averages over the whole period; total sample):
  - Log of GDP per capita 1/2/: 
    - 0.0130** [2.53]; 0.0102* [1.95]; 0.0134*** [2.74]; 0.0111 [1.45]; 0.0200** [2.07]
  - Previous period growth 3/4/: 
    - 0.0011 [0.60]; 0.0008 [0.59]; 0.0016 [1.24]; 0.0012 [0.49]; -0.0037 [-1.60]
  - Fiscal Balance to GDP 3/4/: 
    - 0.3104** [2.43]; 0.2932*** [2.67]; 0.2846** [2.55]; 0.1732* [1.81]; 0.2735* [1.69]
  - Net foreign assets to GDP 2/: 
    - 0.0237** [2.41]; 0.0297*** [2.91]; 0.0424*** [5.44]; 0.0268** [2.19]; 0.0769*** [5.03]
  - Old dependency ratio 2/4/: 
    - -0.4416*** [-4.86]; -0.3271*** [-3.83]; -0.4353*** [-5.21]; -0.0229 [-0.14]; -0.4896*** [-3.48]
  - Young dependency ratio 2/4/: 
    - -0.0238 [-0.89]; -0.0237 [-1.03]; -0.0265 [-1.04]; 0.0463 [1.22]; 0.0566 [0.80]
  - Trade openness 2/: 
    - 0.0083 [0.73]; 0.0222** [1.97]; 0.0050 [0.45]; -0.0171 [-1.00]; 0.0102 [0.51]
  - Increase in the old dependency ratio over 5 years: 
    - 1.0614*** [3.16]; 1.2362*** [3.47]; 1.0508*** [3.29]; 0.8079** [2.37]; 1.5315** [2.48]
  - Contemporaneous oil price*Oil producer 3/: 
    - 0.0005** [2.26]; 0.0005** [2.57]; 0.0005** [2.39]; -0.0002 [-0.63]; 0.0004* [1.69]
  - Financial integration 2/: 
    - 0.0020 [0.90]; -0.0005 [-0.24]; 0.0018 [0.96]; 0.0099 [1.16]; -0.0020 [-1.14]
  - Financial integration*Previous period growth 5/: 
    - -0.0013** [-2.16]; -0.0009* [-1.75]; -0.0014*** [-3.35]; 0.0001 [0.03]; -0.0001 [-0.21]
  - Crerdit Market Regulation 4/6/ (sic): 
    - -0.0053** [-2.34]; -0.0078*** [-3.32]; -0.0059*** [-3.11]; -0.0008 [-0.34]; -0.0105*** [-2.79]
  - Gross replacement rate 4/6/7/: 
    - 0.0843** [2.38]; 0.1031* [1.68]; 0.1275*** [4.41]; 0.0385 [0.81]; 0.1962*** [3.91]
  - Corporate income tax rate 4/6/: 
    - 0.0015*** [2.81]; 0.0018*** [2.82]; 0.0011** [2.10]; 0.0003 [0.39]; -0.0004 [-0.53]
  - Ratio of minimum wage to mean wage 4/6/: 
    - -0.0495*** [-3.37]; -0.0725*** [-3.22]; -0.0460*** [-3.43]; -0.0388** [-2.07]; -0.0565** [-1.98]
  - Employment protection index 4/6/8/: 
    - -0.0004 [-0.07]; -0.0076** [-2.01]; -0.0027 [-0.62]; 0.0018 [0.30]; -0.0002 [-0.03]
- Interaction terms and other reported coefficients:
  - Crerdit Market Regulation*Previous period growth 3/4/6/: -0.0002 [-0.30]
  - Gross replacement rate*Previous period growth 3/4/6/7/: 
    - -0.0145 [-1.43]; -0.0175** [-2.21]; -0.0166** [-2.17]; -0.0115 [-1.21]; -0.0257** [-2.05]
  - Corporate income tax rate*Previous period growth 3/4/6/: 
    - 0.0003 [1.26]; 0.0003 [1.60]; 0.0004** [1.96]; 0.0001 [0.57]; 0.0003 [0.72]
  - Ratio of minimum wage to mean wage*Previous period growth 3/4/6/: 
    - -0.0119** [-2.26]; -0.0109** [-2.11]; -0.0110** [-2.24]; -0.0136*** [-2.75]; -0.0199** [-2.01]
  - Employment protection index*Previous period growth 3/4/6/8/: -0.0003 [-0.19]
  - Ratio of minimum wage to mean wage*Financial integration 2/4/6/: 0.0034 [0.38]
  - Gross replacement rate*Trade openness 2/4/6/7/: 0.0901 [1.25]
  - Gross replacement rate*Net foreign assets to GDP 2/4/6/7/: 
    - 0.1560*** [2.96]; 0.2037*** [4.66]; 0.0632 [0.86]; 0.2415** [2.44]
- Sample sizes and coverage:
  - Observations: 323; 323; 323; 116; 118
  - Number of countries: 60; 60; 60; 44; 59

### Structural indicators — Germany in comparison (Table 7)
- Latest available year(s) and values (as presented):
  - Credit Market Regulation 1/2008: United States 7.7; France 8.2; Spain 9.3; Japan 8.9; Germany 8.2; OECD Average 9.0
  - Combined Corporate Income Tax Rate 2/2009: United States 39; France 34; Spain 34; Japan 30; Germany 40; OECD Average 30; (additional values: 26 shown in table row but alignment unclear in source)
  - Average Labor Tax Wedge (single earner w/o children at 100 percent of average wage) 2009: United States 29; France 49; Spain 38; Japan 29; Germany 51; OECD Average 36
  - Doing business rank on paying taxes 2009: United States 54; France 55; Spain 58; Japan 61; Germany 11; OECD Average 58; NA noted
  - Unemployment Gross Replacement Rate 3/2008: United States 55; France 66; Spain 61; Japan 54; Germany 60; OECD Average 56
  - Product Market regulation 2008: United States 0.8; France 1.5; Spain 1.0; Japan 1.1; Germany 1.3; OECD Average 1.4
  - Regulation in Professional Services 2008: United States 1.1; France 2.1; Spain 2.1; Japan 1.5; Germany 2.9; OECD Average 2.0
  - Regulation in Retail Trade 2008: United States 2.6; France 3.1; Spain 2.7; Japan 2.4; Germany 2.4; OECD Average 2.4
  - Regulation in Energy Transport and Communication 2007: United States 1.8; France 2.2; Spain 1.6; Japan 2.2; Germany 1.1; OECD Average 2.1
  - Employment Protection Regulation 2008: United States 0.2; France 2.5; Spain 2.5; Japan 1.9; Germany 3.0; OECD Average 2.2
  - Ratio of minimum wage to mean wage 4/2005: United States 0.3; France 0.5; Spain 0.2; Japan 0.4; Germany 0.0; OECD Average 0.3
- Notes from table:
  - 1/ Higher value means less regulated. For the United States the index declined from 8.9 to 7.7 between 2007 and 2008.
  - 2/ Regression included federal corporate income tax rate; table presents combined corporate income tax rate including sub-central tax rates.
  - 3/ Regression included a two-year average unemployment gross replacement ratio; table presents unemployment net replacement ratio for OECD countries up to 2008.
  - 4/ Germany has no minimum wage in most sectors; de facto floor may be higher via collective bargaining.

### Germany and EU: Savings and Investment by Sector (Panel 1)
- Graphical panels present time series (1999–2009) comparing European Union and Germany for:
  - Corporate Savings (Net, Percent of GDP)
  - Corporate Investment (Net, Percent of GDP)
  - Household Savings (Net, Percent of GDP)
  - Household Investment (Net, Percent of GDP)
  - Government Savings (Net, Percent of GDP)
  - Government Investment (Net, Percent of GDP)
- Figures plotted across 1999–2009 show separate series for European Union and Germany (no numeric table values provided in source excerpt).

### Data and methodology (Appendix: Data Description)
- Sample:
  - 106 advanced, emerging and developing countries with population exceeding one million.
  - OECD sample included 27 countries.
  - New EU member states included starting from 1994.
- Data sources and definitions (as provided):
  - Current account to GDP: Annual Macroeconomic Database (AMECO) where available; IMF WEO otherwise.
  - Income per capita: real PPP GDP per capita in 2005 constant prices with 1996 reference year from Penn World Tables 6.3 up to 2007; remaining years extrapolated using per capita real GDP growth from IMF WEO.
  - Fiscal balance to GDP: general government net lending/borrowing from IMF WEO where available; otherwise general government overall fiscal balance from WEO.
  - Net foreign assets to GDP and financial integration: External Wealth of Nations (1970–2007) database (foreign assets minus liabilities)/GDP; financial integration = sum of foreign assets and foreign liabilities divided by GDP.
  - Old/young dependency ratios: World Development Indicators (WDI); old = population above 64 relative to 15–64; young = population below 15 relative to 15–64.
  - Increase in the old dependency ratio computed over five-year period.
  - Trade openness: exports + imports divided by GDP from Penn World Tables 6.3 (‘openc’/100).
  - Oil price: IMF WEO.
  - Averaging and timing:
    - Several macro variables averaged over 5-year non-overlapping periods: 1975–79, 1980–84, 1985–89, 1990–94, 1995–99, 2000–04, 2005–09.
    - Other variables included as of the year preceding start of five-year period (e.g., 2004 for 2005–09).
    - Many variables included as deviations from PPP-weighted sample average.
    - Real GDP per capita computed as ratio to US real GDP per capita in a given year.
- Structural/regulatory indicators:
  - Credit market regulation: Fraser Institute index (ranges 0–10; higher = less regulation) based on public ownership of banking, interest rate controls, percent of credit to private sector, competition from foreign banks.
  - Gross unemployment replacement rate: Aleksynska and Schindler (2011); average over two years of unemployment.
  - Ratio of minimum wage to mean wage: Aleksynska and Schindler dataset.
  - Employment protection indicator: OECD for OECD countries; for other countries, constructed as out-of-sample forecast from regression of OECD employment protection index on advance notice period (months) and severance pay after nine months (months) (Aleksynska & Schindler, 2010).
  - Corporate income tax rates: OECD database for OECD countries (central government statutory rate, gross of sub-central deduction); IMF Fiscal Affairs Corporate Income Tax rate database for other countries.
  - Doing business rank on paying taxes: World Bank Doing Business indicator (rank among 183 countries).
  - Labor tax wedge (OECD): combined central and sub-central income tax plus employee and employer social security contributions, expressed as percent of labor costs; adjusted for cash transfers.
  - Product market regulation: OECD indicator available for 1998, 2003, 2008; assigned to corresponding five-year periods when used as time-varying.
  - ETCR (regulation in energy, transport and communications): annual data for 1975–2007 with gaps for some countries; attributed to five-year periods.
  - Regulation in professional services and retail trade: available for 1996, 2003, 2008; assigned to corresponding five-year periods in econometric analysis.

*Source: _wp1261 - Section 5 (selected tables, panels, and Appendix: Data Description).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp1261.pdf_