## _wp0692

## Source details

**Canonical URL:** [_wp0692](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0692.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0692.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0692.pdf.json)

---

### Key findings on female labor force participation (Canada and cross-country)
- After a long period of stability, the Canadian labor participation rate has risen rapidly over the last decade, driven mainly by an increase in female participation (Figure 1).
- Female participation in Canada: "At 73 percent, female participation in Canada is described as 'now only just short of the rate in the Nordic countries (Figure 2).'"
- Econometric analysis suggests reforms in the Canadian tax and benefit system since 1995 can explain 30–80 percent of the rise in female participation rates through 2001.
- Aggregate participation rate in Canada has surpassed that in the United States and converged to the high levels of the Nordic countries.
- Male participation rate has fallen modestly since the early 1990s, while female participation has risen significantly.
- Actual increase in female participation between 1995 and 2001: 3.2 percentage points.

### Quantitative snapshot (Select OECD Countries: Labor Force Participation, Ages 15–64, by Gender, 1995 and 2004)
- Table 1 (1995 2004 1995 2004):
  - Canada 74.6 78.2 67.7 73.5
  - Nordics 77.2 78.0 73.1 75.1
  - EU15 67.6 70.5 57.0 62.6
  - G10 72.8 74.1 63.4 66.6
  - OECD 69.5 70.0 58.0 59.9
  - United States 76.9 75.4 69.7 69.2
- (Source: OECD, Labor Force Statistics Database, 2005.)

### Theoretical framework and policy variables examined
- Marginal female worker profile: married with children and typically a secondary earner; participation depends on after-tax household income relative to costs (opportunity cost of lost leisure time, child care).
- Key policy/incentive variables:
  - Tax wedges (higher tax wedges raise the opportunity cost of working and discourage work-effort).
  - Childcare subsidies and formal childcare coverage (benefits that lower the opportunity cost of work and encourage participation).
  - Net tax-benefit wedge (combined effect of taxes and benefits on incentives).
- Other determinants and institutional variables included:
  - Unemployment rate (captures labor market pressures and a "discouraged worker" effect).
  - Female wage rate (significant and of the expected sign).
  - Male-female wage gap.
  - Wage compression (lower wage compression raises participation by lowering the relative market price of hiring a nanny).
  - Union density (seems to encourage female participation).
  - Employment protection regulation (seems to encourage female participation).
  - Preferences for education and child-bearing.
  - Proportion of parliamentary seats possessed by women (captures cultural/social norms).
  - Country-specific time trends (partly intended to capture demographics).
  - Parental leave provision (tends to have a small but negative effect on participation).

### Data, methods, and scope
- Econometric analysis: panel of 10 large industrial countries over 1980–2001.
- Countries: Australia, Canada, Denmark, Finland, France, Germany, Great Britain, Spain, Sweden, United States.
- Main objective: explain importance of changes in the tax and benefit system in raising Canadian female participation, controlling for determinants above.
- Policy variables: tax and benefit wedges (and squares) used to capture nonlinearities.
- Country-specific time trends included in some specifications.
- Study period limited to 1980–2001 due to data limitations for post–2001 benefit wedge construction.

### Econometric framework (specification and variables)
- Empirical participation equation (notation preserved):
  - .
    itiiititit
    tcZXPRεγβ++++=                                                (1)
- Dependent variable: female labor participation rate (aged 15–64), for country i and period t.
- X_it: policy variables (tax and benefit wedges, and their squares); sometimes primary earner’s tax wedge and its square included.
- Z_it: nonpolicy variables (labor market structure, preferences for education and child-bearing, social norms).
- Country fixed effects c_i included; time trends t introduced in some specifications.
- All variables expressed in their natural logarithm; sample is unbalanced.
- Z_it variables and a priori expected signs:
  - Union density: unclear a priori.
  - Employment protection: expected to encourage participation.
  - Wage compression (median / 10th percentile ratio): expected to hinder participation.
  - Proportion of population below 15 years: expected negative sign.
  - Male unemployment rate: expected negative sign.
  - Paid parental leave (maximum weeks for mother): theoretical effect unknown.
  - Proportion of parliamentary seats held by women: expected positive sign.
  - Education enrollment (years of study for women aged 25+): ambiguous sign.
  - Wage gap: expected negative sign.
  - Female wage rate: expected positive sign.

### Main empirical results and magnitudes
- General conclusions:
  - Tax and benefit wedges are significant determinants of female labor force participation.
  - Nonlinear impact of tax wedge: negative level term and positive squared term imply tax wedge lowers participation as wedge rises for tax levels below 35 percent of the APW wage.
- Elasticities and marginal impacts:
  - A 1-percentage-point cut in the current tax wedge could raise Canadian female participation by 0.17–0.19 percentage points (tax elasticity around −0.24), with the largest impact when country-specific time trends are included.
  - Benefit wedge positive impact with time trends: a 1-percentage-point increase in the benefit wedge raises female participation rate by 0.14 percentage points. Without time trends, estimated effect larger (sign switches at a 34 percent wedge).
- Model predictions vs actual for Canada (1995–2001):
  - Predicted increases in Canadian female participation: 4.3 and 3.8 percentage points (two specifications), actual increase around 3.2 percentage points.
  - Models overpredict contribution of nonpolicy variables.
- Counterfactual contribution of Canadian reforms (model with time trends):
  - Reforms in the tax and benefit system accounted for 1.2 percentage points of the rise in female participation, equally divided among taxes and benefits.
- Summary quantitative statement:
  - The decrease in the secondary earner’s tax wedge, coupled with the increase in childcare benefits, can explain at least one-third (or 1.2–4 percentage points) of the increase in Canadian women’s participation rate between 1995 and 2001.

### Table 4 — Country-specific impacts of a 1-percentage-point change (preserved exactly)
- Australia: Tax Wedge −0.28; Benefit Wedge 1.17; With Time Trends: −0.31; 0.15
- Canada: Tax Wedge −0.17; Benefit Wedge 0.89; With Time Trends: −0.19; 0.14
- Finland: Tax Wedge −0.59; Benefit Wedge 0.19; With Time Trends: −0.67; 0.12
- France: Tax Wedge −0.63; Benefit Wedge 0.19; With Time Trends: −0.71; 0.12
- Germany: Tax Wedge 0.60; Benefit Wedge 0.65; With Time Trends: 0.71; 0.14
- Spain: Tax Wedge −0.80; Benefit Wedge 1.06; With Time Trends: −0.92; 0.15
- Sweden: Tax Wedge −0.31; Benefit Wedge −0.11; With Time Trends: −0.35; 0.12
- United Kingdom: Tax Wedge −0.36; Benefit Wedge 1.42; With Time Trends: −0.40; 0.16
- United States: Tax Wedge −0.29; Benefit Wedge 0.83; With Time Trends: −0.32; 0.14

### Additional empirical findings (by specification)
- Without country-specific time trends:
  - Female wage rate positive and significant.
  - Higher wage compression lowers female participation.
  - Higher union density raises female participation.
  - Proportion of parliamentary seats held by women positively associated with participation.
  - Parental leave has a small negative effect.
  - Wage differentials and education enrollment not important in this specification.
- With country-specific time trends:
  - Some institutional factors drop out (wage compression, wage gap, women’s parliamentary seats, unemployment rate, parental leave).
  - Education enrollment acts as a substitute to labor participation.
  - Union density and employment protection induce higher female participation.

### Policy-relevant implications and recommendations
- Reductions in secondary earner tax wedges can raise female participation; marginal impact sizable at current tax levels (1-percentage-point tax cut → 0.17–0.19 percentage-point rise).
- Increases in public childcare provision and benefits encourage female labor market entry; a 1-percentage-point increase can raise participation by about 0.14 percentage points (with time trends).
- “Family-friendly” policies (childcare expansion, parental leave design, targeted tax credits) interacted with tax reforms contributed materially to Canada’s rising female participation in the 1990s.
- Policy design should consider nonlinearities:
  - Tax wedge effects differ by level (impact changes around 35 percent of APW wage).
  - Benefit wedge may exhibit diminishing returns (effect switches sign at a 34 percent wedge).
- Additional targeted reforms suggested:
  - Adopt policies to boost participation among older workers (e.g., aged 55 and above) where Canada’s participation is only average by international standards.
  - Strengthen participation among low-income workers, where participation is very weak.
  - Consider complementary reforms (e.g., lower tax wedges, reforms in the pension system) to sustain and extend gains from “family-friendly” policies.
- Long-term considerations:
  - Policies that induce particularly young women to enter the workforce could raise life-time participation through hysteresis.
  - Given population aging and strains on welfare, pension, and health systems, raising participation remains a policy challenge even for Canada.

### Canadian tax and benefit system developments (1990s–early 2000s)
- Tax wedge trends:
  - Average tax wedge on primary workers stayed relatively stable since 1996 at around 23 percent.
  - Tax wedge for secondary earners fell from 35 to 31 percent.
- Major federal tax changes:
  - 3 percent general surtax initially eliminated for low- and middle-income families in 1998; rescinded for other income groups in 1999.
  - 5 percent deficit-reduction surtax for incomes up to C$85,000 eliminated for other income groups.
  - Middle-income personal income tax rate lowered from 26 percent to 24 percent in mid-2000.
  - 1997: Canada Child Tax Benefit announced with enriched supplements initially for low-income families; in 2000 also available for modest- and middle-income families.
  - Five-year tax reduction plan initiated after 2000.
- Benefit and childcare policy timeline:
  - 1997: National Children’s Agenda agreement; National Child Benefit introduced (federal).
  - 2000: Early Childhood Development (ECD) Agreement to provide federal funds for childcare and services for children under 6.
  - January 2002: maternity and parental leave expanded from 35 to 50 weeks; eligibility hours reduced from 700 to 600.
  - March 2003: Multilateral Framework on Early Learning and Childcare signed — federal funds to provinces/territories must be used to “improve access to affordable, quality, provincially/territorially-regulated early learning and childcare programs”.
  - 2004 budget: federal spending of C$5 billion in the next five years to help develop further childcare provision.
  - Québec (since 1996/97): comprehensive family support policy including full-day universal childcare for C$5 per day.
  - Ontario’s 2004 Best Start Strategy: universal expansion of childcare to all children above 2½ years old within the next 10–15 years.
- Quantitative measures and trends:
  - Benefit wedge (proxy based on public spending on childcare) has tripled in the last decade, with the largest increase in the second half of the 1990s.
  - In 2001, Canadian benefit wedge was close to the average for large industrial countries.
  - Combining tax cuts and increased benefits, Canadian net tax wedge on secondary earners (tax wedge minus benefit wedge) has fallen by 30 percent between 1995 and 2001.

### Childcare, quality, and province-specific notes (highlights from appendices)
- Early Childhood Education and Care (ECEC) institutional setup:
  - Kindergarten: public education for children five years old and above; regarded as public and universal good.
  - Childcare: provincial/territorial jurisdiction; not an entitlement or mandated service.
- Financing and fees:
  - Childcare fees mostly paid by parents; these fees, excluding in Québec, fluctuate between 34–82 percent of the cost.
  - Example: parents in Ontario pay around one-fifth of their APW’s wage in childcare fees.
  - Federal contributions include Childcare Expense Deduction and direct funding for targeted programs; transfers via ECD Agreement and Multilateral Framework.
- Quality and availability:
  - Access: Almost one-third of mothers in paid employment have access to center-based childcare, up from less than one in five a decade ago.
  - Staffing: since 1998, five provinces improved childcare staff’s salaries by earmarking funds.
  - Regulations: staff-to-child ratios usually range from 1:3 to 1:5 for infant care.
  - Qualifications: many provinces require at least two-thirds of staff to have a university degree in early childhood education; in Canada, more than four out of five childcare workers have at least one year of postsecondary education in early childhood education (OECD, 2004b).
- Québec specifics:
  - Québec policy introduced in 1997; objectives include fighting poverty, increasing mothers’ labor participation, enhancing child development.
  - Québec public funding for universal childcare: ¾ percent of GDP in 2001, up from ¼ percent in 1995.
  - Parental contribution initially C$5 per day (now C$7) for up to 10 hours.
  - Childcare spaces (age 0–5): around 80,000 in 1997 to 180,000 in March 2004 (expected to reach 200,000 by 2006 for 370,000 children).
  - Proportion of children (aged 0–12) in formal childcare increased from 19 percent to 45 percent.
  - Québec accounts for 40 percent of Canadian childcare capacity while its population of young children is only 22 percent of the Canadian total.
  - Maternal employment (mothers with youngest child aged below 3 years): 61 percent in Québec by 2001; increase of almost 7 percent following reforms.
  - Full-time dual-earner families with dependent children: Québec moved from less than 40 percent in 1996 to about one out of two families by 2001; comparative figures: 45 percent in Canada as a whole, 40 percent in Sweden, around 30 percent in the United Kingdom.

### Constructing wedges and data series (methodological highlights)
- Tax wedge definitions:
  - Primary breadwinner tax wedge: combined burden of income taxes plus employee and employer social security contributions as a percentage of labor costs, for a one-earner married couple with two children earning APW’s wage.
  - Secondary breadwinner tax wedge: share of her earnings that goes into paying additional household taxes; A = one-earner married couple earning 100 percent of APW; B = two breadwinners earning 100 and 67 percent of APW.
  - Secondary earner assumed to have gross earnings equal to two-thirds of the primary worker’s earnings.
- Benefit wedge (scope and definition):
  - Includes only publicly provided benefits in kind related to family care (child and youth); cash benefits ignored.
  - Benefit wedge on the secondary breadwinner: share of the secondary earner’s gross income that is obtained through public childcare provision, conditional on female labor force participation; formula uses A and B as defined above.
  - Benefit wedge proxies public provision of childcare and is interpreted as mixture of price and income effects.
- Data and key series:
  - Time period: 1980–2001; sample countries listed above.
  - Female labor force participation rate: OECD Labor Force Statistics database.
  - Primary earner’s tax wedge: OECD Taxing Wages Statistics-Historical Tax Rates.
  - Secondary earner’s tax wedge: Jaumotte (2003) updated with OECD Taxing Wages Database.
  - Benefit wedge: constructed using OECD Social Expenditure Database (2004c) and OECD Labor Force Statistics Database; missing data interpolated with Jaumotte (2003).
  - Average production wage (APW) values (US dollars, PPP) used in calculations:
    - Australia 35,867
    - Canada 32,488
    - Denmark 36,142
    - Finland 27,988
    - France 23,771
    - Germany 34,260
    - Netherlands 32,747
    - Norway 32,412
    - Spain 20,946
    - Sweden 24,259
    - United Kingdom 29,701
    - United States 32,360
  - Other inputs: union density (OECD), wage compression (median/10th percentile ratio), female wage rate (average hourly wages in manufacturing from Gauthier, 2003), education enrollment (Jaumotte, 2003), employment protection index (Allard, 2003), parental leave (Jaumotte, 2003), proportion young population (OECD), male unemployment rate (OECD), women parliamentary seats (Huber and others, 2004), wage gap (Gauthier, 2003).

*Source content unit: _wp0692 - REFERENCES (PDF chapter/section).*

### References..............................................................................................................

### _wp0692 - References

### Key findings on female labor force participation (Canada and cross-country)
- After a long period of stability, the Canadian labor participation rate has risen rapidly over the last decade, driven mainly by an increase in female participation (Figure 1).
- At 73 percent, female participation in Canada is described as "now only just short of the rate in the Nordic countries (Figure 2)."
- The econometric analysis suggests that reforms in the Canadian tax and benefit system since 1995 can explain 30–80 percent of the rise in female participation rates through 2001.
- The aggregate participation rate in Canada has surpassed that in the United States and converged to the high levels of the Nordic countries.
- Male participation rate has fallen modestly since the early 1990s, while female participation has risen significantly.

### Quantitative snapshot from Table 1 (Select OECD Countries: Labor Force Participation, Ages 15–64, by Gender, 1995 and 2004)
- 1995 2004 1995 2004
- Canada 74.6 78.2 67.7 73.5
- Nordics 77.2 78.0 73.1 75.1
- EU15 67.6 70.5 57.0 62.6
- G10 72.8 74.1 63.4 66.6
- OECD 69.5 70.0 58.0 59.9
- United States 76.9 75.4 69.7 69.2
- (Source: OECD, Labor Force Statistics Database, 2005.)

### Theoretical framework and policy variables examined
- The analysis follows a framework where the marginal female worker is married with children and typically a secondary earner. Women's participation decision depends on the additional household after-tax income relative to costs associated with work (opportunity cost of lost leisure time, child care).
- Key policy/incentive variables:
  - Tax wedges (higher tax wedges raise the opportunity cost of working and discourage work-effort).
  - Childcare subsidies and formal childcare coverage (benefits that lower the opportunity cost of work and encourage participation).
  - Net tax-benefit wedge (combined effect of taxes and benefits on incentives).
- Other determinants and institutional variables included:
  - Unemployment rate (captures labor market pressures and a "discouraged worker" effect).
  - Female wage rate (significant and of the expected sign).
  - Male-female wage gap.
  - Wage compression (lower wage compression raises participation by lowering the relative market price of hiring a nanny).
  - Union density (seems to encourage female participation).
  - Employment protection regulation (seems to encourage female participation).
  - Preferences for education and child-bearing.
  - Proportion of parliamentary seats possessed by women (used to capture cultural/social norms).
  - Country-specific time trends (partly intended to capture demographics).
  - Parental leave provision (tends to have a small but negative effect on participation).

### Data, methods, and scope
- Econometric analysis uses a panel of 10 large industrial countries over the period 1980–2001 to explain the importance of changes in the tax and benefit system in raising Canadian female participation, after controlling for the determinants noted above.
- The analysis examines both tax and benefit wedges to gauge the effect of policies aimed at helping women combine work and family life.
- Country-specific time trends are included in some specifications to capture demographics and other unobserved, slowly moving factors.
- The study is limited to the period 1980–2001 because data limitations hinder the construction of the benefit wedge for the post–2001 period.

### Empirical implications and interpretation
- Policy initiatives (tax and benefit reforms) can be pivotal in boosting labor supply, particularly in view of the problem of population aging.
- Nonpolicy variables also play important roles; examples highlighted include:
  - Unemployment rate: poor job prospects lead women to leave the labor market.
  - Female wage rate: higher wages increase participation likelihood.
  - Institutional features (union density, employment protection) increase willingness to enter the labor market by enhancing job security.
  - Greater parental leave provision tends to slightly depress measured participation.

### Data visuals and appendices referenced (topics only)
- Figures referenced: Canada labor force participation by gender 1990–2004; Select OECD female labor force participation 2004; formal childcare coverage and female participation 2001; tax wedge series for Canada and select OECD countries; benefit wedge series; predicted vs actual labor market developments; contributions of policy variables to participation developments.
- Appendices address: Constructing a Tax Wedge; Early Childhood Education and Care in Canada; Quality and Availability of Canadian Child Care; Childcare Provision in Québec; Constructing a Benefit Wedge; Data Analysis.
- Appendix tables and figures include public spending on childcare services, parental fees, average production workers’ wage, childcare fees, regulated childcare spaces, maternal employment, and full-time dual earner families.

*Source: _wp0692 - References (IMF working paper PDF content provided).*

### Section III analyzes the main developments in the tax and benefit system in Canada in the

### _wp0692 - Section III analyzes the main developments in the tax and benefit system in Canada in the

### Literature on childcare and taxation
- Child care findings (non-Canadian studies):
  - Elasticities for married mothers’ employment range from zero to −1.6 (Blau and Robins, Connelly, Kimmel).
  - Single mothers’ elasticities reported between −4.54 and +1.38 (Kimmel, 1998).
  - More recent studies (Blau and Currie, Blau, Choné et al.) find a rather weak link between childcare prices and labor supply.
- Taxation findings (non-Canadian studies):
  - Wide range of estimates for income and substitution effects; little consensus on labor supply elasticities (Pencavel; Killingsworth; Fuchs et al.).
  - Conflicting conclusions on importance of taxes for labor supply (Prescott; Ueberfeldt; Blanchard).
- Canadian evidence:
  - Child care: Cleveland, Gunderson, and Hyatt (1996) — a 10 percent decrease in prices increases maternal employment by almost 4 percent; Powell (1997) — childcare price elasticity of −0.38 for participation and −0.32 for hours worked; Lefebvre and Merrigan (2005) — Québec low-fee daycare (1996) had substantial labor supply effects.
  - Taxation: OECD (2005) — 1988 tax reforms incentivized women with high-earning husbands to enter labor market; Jeon (2004) — participation rates for women with high-earning husbands increased by 7.3 percent and working hours increased by 200 hours compared to women with low-earning husbands.
- Cross-country panel studies:
  - Jaumotte (2003): female participation boosted by public childcare spending and parental leave.
  - Genre, Salvador, and Lamo (2005): confirms parental leave effect.
  - Immervoll and Barber (2005): childcare costs must be analyzed with wider financial incentives.
  - Findings on tax wedges mixed: Smith et al. (2003) find large effects; Jaumotte (2003) finds negative effect of secondary earner’s tax wedge; Tsounta and Bonato (2005) and Genre et al. find statistically insignificant effects using alternative measures.

### The Canadian tax system (1990s–early 2000s)
- Trends and magnitudes:
  - Average tax wedge on primary workers has stayed relatively stable since 1996 at around 23 percent.
  - Tax wedge for secondary earners fell from 35 to 31 percent (Appendix I; Figures 4 and 5).
- Federal policy changes:
  - 3 percent general surtax initially eliminated for low- and middle-income families in 1998; rescinded for other income groups in 1999.
  - 5 percent deficit-reduction surtax for incomes up to C$85,000 eliminated for other income groups.
  - Middle-income personal income tax rate lowered from 26 percent to 24 percent in mid-2000.
  - 1997: Canada Child Tax Benefit announced with enriched supplements initially for low-income families; in 2000 also available for modest- and middle-income families.
  - Five-year tax reduction plan initiated after 2000.
- Provincial measures and interactions:
  - Provincial income tax cuts occurred, but increases in Ontario’s surtax to address heath care costs may have increased the primary earner’s tax wedge, offsetting federal cuts.
- International comparisons:
  - Canada’s secondary earner’s average tax wedge is much lower than Germany and Denmark, but higher than United States, Sweden, Switzerland, and Australia (Figures 6 and 7).
- Methodological note:
  - Figures calculated for Ontario; tax wedge for secondary earners excludes social security contributions; secondary earner assumed to have gross earnings equal to two-thirds of the primary worker’s earnings.

### The Canadian benefit system and childcare policies
- Major policy actions and dates:
  - 1997: National Children’s Agenda agreement; National Child Benefit introduced (federal) — increased incentives for single parents and freed provincial funds for family support projects.
  - 2000: Early Childhood Development (ECD) Agreement to provide federal funds for childcare and services for children under 6.
  - January 2002: maternity and parental leave expanded from 35 to 50 weeks; eligibility hours reduced from 700 to 600.
  - March 2003: Multilateral Framework on Early Learning and Childcare signed — federal funds to provinces/territories must be used to “improve access to affordable, quality, provincially/territorially-regulated early learning and childcare programs”.
  - 2004 budget: federal spending of C$5 billion in the next five years to help develop further childcare provision.
  - Québec (since 1996/97): comprehensive family support policy including full-day universal childcare for C$5 per day.
  - Ontario’s 2004 Best Start Strategy: universal expansion of childcare to all children above 2½ years old within the next 10–15 years.
- Quantitative measures and trends:
  - Benefit wedge (proxy based on public spending on childcare) has tripled in the last decade, with the largest increase in the second half of the 1990s (Figure 8).
  - In 2001, Canadian benefit wedge was close to the average for large industrial countries (Figure 9).
  - Combining tax cuts and increased benefits, Canadian net tax wedge on secondary earners (tax wedge minus benefit wedge) has fallen by 30 percent between 1995 and 2001 (Figure 10).
- Methodological note:
  - Benefit wedge quantifies the proportion of labor costs of a potential secondary earner lost if market work is forgone to raise children; benefits in kind relating to childcare included; calculation assumes benefit depends on the female’s participation in the labor market.

### Econometric framework
- Sample and period:
  - The sample covers 10 large industrial countries over the period 1980–2001.
- Empirical participation equation (notation preserved):
  - .
    itiiititit
    tcZXPRεγβ++++=                                                (1)
  - Dependent variable: female labor participation rate (aged 15–64), for country i and period t.
  - X_it: policy variables (tax and benefit wedges, and their squares); in some specifications the primary earner’s tax wedge and its square included.
  - Z_it: nonpolicy variables (labor market structure, preferences for education and child-bearing, social norms).
  - Country fixed effects c_i included; time trends t introduced in some specifications.
  - All variables expressed in their natural logarithm; sample is unbalanced.
- Variables in Z_it and a priori expected signs:
  - Union density: unclear a priori.
  - Employment protection: expected to encourage participation.
  - Wage compression (median / 10th percentile ratio): expected to hinder participation.
  - Proportion of population below 15 years: expected negative sign.
  - Male unemployment rate: expected negative sign.
  - Paid parental leave (maximum weeks for mother): theoretical effect unknown; in many countries women on parental leave count as out of labor force.
  - Proportion of parliamentary seats held by women: expected positive sign.
  - Education enrollment (years of study for women aged 25+): ambiguous sign.
  - Wage gap: expected negative sign.
  - Female wage rate: expected positive sign.

### Results — main findings and magnitudes
- General conclusions:
  - Tax and benefit wedges are significant determinants of female labor force participation.
  - Nonlinear impact of tax wedge: negative level term and positive squared term imply tax wedge lowers participation as wedge rises for tax levels below 35 percent of the APW wage.
- Elasticities and marginal impacts:
  - A 1-percentage-point cut in the current tax wedge could raise Canadian female participation by 0.17–0.19 percentage points (i.e., tax elasticity around −0.24), with the largest impact when country-specific time trends are included.
  - The benefit wedge has a positive impact; with time trends included, a 1-percentage-point increase in the benefit wedge raises female participation rate by 0.14 percentage points. Without time trends, the estimated effect is larger (Table 4; sign switches at a 34 percent wedge).
- Table 4 country-specific impacts of a 1-percentage-point change (preserved exactly as in source):
  - Australia: Tax Wedge −0.28; Benefit Wedge 1.17; With Time Trends: −0.31; 0.15
  - Canada: Tax Wedge −0.17; Benefit Wedge 0.89; With Time Trends: −0.19; 0.14
  - Finland: Tax Wedge −0.59; Benefit Wedge 0.19; With Time Trends: −0.67; 0.12
  - France: Tax Wedge −0.63; Benefit Wedge 0.19; With Time Trends: −0.71; 0.12
  - Germany: Tax Wedge 0.60; Benefit Wedge 0.65; With Time Trends: 0.71; 0.14
  - Spain: Tax Wedge −0.80; Benefit Wedge 1.06; With Time Trends: −0.92; 0.15
  - Sweden: Tax Wedge −0.31; Benefit Wedge −0.11; With Time Trends: −0.35; 0.12
  - United Kingdom: Tax Wedge −0.36; Benefit Wedge 1.42; With Time Trends: −0.40; 0.16
  - United States: Tax Wedge −0.29; Benefit Wedge 0.83; With Time Trends: −0.32; 0.14
- Additional empirical findings:
  - Without country-specific time trends:
    - Female wage rate positive and significant.
    - Higher wage compression lowers female participation.
    - Higher union density raises female participation.
    - Proportion of parliamentary seats held by women positively associated with participation.
    - Parental leave has a small negative effect.
    - Wage differentials and education enrollment not important in this specification.
  - With country-specific time trends:
    - Some institutional factors drop out (wage compression, wage gap, women’s parliamentary seats, unemployment rate, parental leave).
    - Education enrollment acts as a substitute to labor participation.
    - Union density and employment protection induce higher female participation.
- Model predictions vs actual for Canada:
  - Predicted increases in Canadian female participation: 4.3 and 3.8 percentage points (two specifications), actual increase around 3.2 percentage points.
  - The models overpredict contribution of nonpolicy variables.
- Counterfactual contribution of reforms in Canada:
  - In the model with time trends, reforms in the tax and benefit system accounted for 1.2 percentage points of the rise in female participation, equally divided among taxes and benefits.

### Policy-relevant implications (as described in the source)
- Reductions in secondary earner tax wedges can raise female participation; marginal impact sizable at current tax levels (1-percentage-point tax cut → 0.17–0.19 percentage-point rise).
- Increases in public childcare provision and benefits (benefit wedge) encourage female labor market entry; a 1-percentage-point increase can raise participation by about 0.14 percentage points (with time trends).
- “Family-friendly” policies (childcare expansion, parental leave design, and targeted tax credits) interacted with tax reforms contributed materially to Canada’s rising female participation in the 1990s.
- Policy design should consider nonlinearities: tax wedge effects differ by level (impact changes around 35 percent of APW wage) and benefit wedge may exhibit diminishing returns (effect switches sign at a 34 percent wedge).

*Source: _wp0692 - Section III analyzes the main developments in the tax and benefit system in Canada in the (IMF staff calculations and analysis as presented in the provided content).*

### 3.2 percentage points actual increase in female participation between 1995 and 2001

### 3.2 percentage points actual increase in female participation between 1995 and 2001

### Data, scope, and methodology
- Annual panel data of 10 large industrial countries over the period 1980–2001 were used.
- The analysis compares predicted and actual changes in female participation between 1995 and 2001 (Figures referenced: Figure 11, Figure 13, Figure 14).
- Models are estimated with and without time trends; effects are larger in models that exclude time trends because of a much larger benefit elasticity.

### Main quantitative findings on Canada (1995–2001)
- The decrease in the secondary earner’s tax wedge, coupled with the increase in childcare benefits, can explain at least one-third (or 1.2–4 percentage points) of the increase in Canadian women’s participation rate between 1995 and 2001.
- Estimated policy impacts:
  - A 1–percentage-point tax cut would raise female participation rate by 0.17–0.19 percentage points.
  - An equivalent increase in childcare benefits has an impact similar or larger, with estimates ranging from 0.14–0.99 percentage points.
- Nonlinearities:
  - The negative impact of tax wedges on female participation is only valid for tax wedges below 35 percent.
  - Some results suggest benefit wedge effects may be nonlinear: benefits above 35 percent of APW’s wage possibly discourage female participation.

### Role of nonpolicy variables
- Unemployment rate:
  - Dampens female participation, possibly due to the “discouraged worker” effect.
- Labor market regulation:
  - Degree of regulation (e.g., union density, employment protection) has a positive impact on female participation in all specifications — incentives to enter the labor market are higher when risks of unemployment are smaller.
- Wage compression:
  - High wage compression discourages female participation since it raises the relative price of childcare, as child-minders typically belong to the lower end of the wage distribution.
- Preferences and lifecycle factors:
  - Years of education enrollment (decision to study) and the decision to have children both act as disincentives to enter the labor market.
  - Parental leave has an ambiguous effect: in some specifications it raises participation (not statistically significant); in others it discourages participation.
- Cultural and institutional characteristics:
  - Role of women in politics and country fixed effects, reflecting institutional characteristics, preferences, and social norms not explicitly modeled, are important in understanding female participation.

### Policy implications and recommendations
- Policies that reduce the secondary earner’s tax wedge and increase childcare benefits (as implemented in Canada) can significantly boost female participation and could be effective in other countries with low female participation.
- Policies that induce particularly young women to enter the workforce could have positive long-term implications through hysteresis by raising life-time participation.
- Given population aging and increasing strains on welfare, pension, and health systems, raising participation remains a policy challenge even for Canada.
- Additional targeted reforms suggested:
  - Adopt policies to boost participation among older workers (e.g., aged 55 and above) where Canada’s participation is only average by international standards.
  - Strengthen participation among low-income workers, where participation is very weak.
  - Consider complementary reforms (e.g., lower tax wedges, reforms in the pension system) to sustain and extend the gains from “family-friendly” policies.

*Source: IMF staff calculations and analysis from the referenced study (annual panel of 10 large industrial countries, 1980–2001).*

### REFERENCES

### _wp0692 - REFERENCES

### Constructing a Tax Wedge (Appendix I)
- Choice and rationale:
  - Uses average tax wedges (readily available) and their square terms to incorporate non-linearities.
  - Marginal tax rates influence working-hours decisions; average rates influence labor market participation (Koskela, 2001).
  - De Haan, Sturm, and Volkerink (2003) and Nickell and Layard (1999) find strong correlation between measures of tax burden; any measure used provides reliable results.
- Tax wedge definitions:
  - Tax wedge on the primary breadwinner: combined burden of income taxes plus employee and employer social security contributions as a percentage of labor costs (gross earnings plus employer social security contributions), for a one-earner married couple that has two children and earnings equal to the APW’s wage.
  - Tax wedge on the secondary breadwinner (with two children), following Jaumotte (2003), defined as the share of her earnings that goes into paying additional household taxes, calculated as:
    AB
    AB
    Income) Gross Household(Income) Gross (Household
    Income)Net  Household(Income)Net  (Household
    1Tax wedge
    −
    −
    −=
    (Equation as presented in source)
  - A denotes one-earner married couple earning 100 percent of APW’s wage; B denotes two breadwinners who earn 100 and 67 percent of APW’s wage, respectively.
- Data note:
  - Only gross earnings are considered since labor costs data for the income group under consideration are unavailable prior to 1997.

### Constructing a Benefit Wedge (Appendix V)
- Scope and definition:
  - Includes only publicly provided benefits in kind related to family care (child and youth); cash benefits are ignored because they are included in OECD tax wedge calculations.
  - Benefit wedge on the secondary breadwinner (with two children) is defined symmetrically to the tax wedge: the share of the secondary earner’s gross income that is obtained through public childcare provision, conditional on female labor force participation.
  - Formula given in source:
    AB
    Income Gross Household-Income Gross Household
    years 15 below Population0.5
    familyfor  kindin  Benefits Public
    *Rateion Participat Female
    dgeBenefit We
    ×
    =
    (Equation as presented in source; A and B as defined above.)
- Assumptions and interpretation:
  - Assumes benefit wedge targets female labor participation and thus affects secondary breadwinners only.
  - Due to data limitations, cannot distinguish benefit wedge for primary and secondary earners; author argues this limitation does not hinder analysis because family social benefits tend to affect women more.
  - Public provision of childcare creates both:
    - Substitution effect: raises opportunity cost of home work.
    - Income effect: subsidized childcare reduces additional income needed to pay for childcare.
  - Resulting estimates are mixtures of price and income elasticities.

### Early Childhood Education and Care in Canada (Appendix II)
- Institutional setup:
  - Main ECEC services: kindergarten (primary purpose child development) and childcare (primary purpose to encourage parents’ labor market participation).
  - Jurisdiction: provincial/territorial. Each province/territory has separate childcare and kindergarten programs, including regulation, funding arrangements, and policy.
  - Kindergarten is part of public education for children five years old and above and is regarded as a public and universal good. Childcare is neither an entitlement nor a mandated service.
- Financing structure:
  - Cost shared by federal government, provincial/territorial government, and parents. Allocation varies by province/territory and service type.
  - Key bullet findings:
    - Childcare fees are mostly paid by parents; these fees, excluding in Québec, fluctuate between 34–82 percent of the cost.
    - Example: parents in Ontario pay around one-fifth of their APW’s wage in childcare fees.
    - Provincial/territorial governments pay fee subsidies directly to programs on behalf of low-income parents; in most provinces (e.g., Ontario and British Colombia) fee subsidization is conditional on social and economic criteria (parents must be in the labor force or engaged in training or education and have a low income).
    - Federal government contributions:
      - Childcare Expense Deduction: an income tax deduction for working families that pay for childcare.
      - Direct funding for programs targeting Aboriginal children, military families, and new immigrants.
      - Transfers to provinces/territories via the Early Childhood Development Agreement and the Multilateral Framework on Early Learning and Childcare.
- Data/figure notes (as presented):
  - Figure 1 (source Immervoll and Barber, 2005): Center-based full-time childcare fees, 2001, in percent of the average production worker's wage; data refer to fees parents pay after deducting government subsidies. Data refer to 2001 or latest available year. Canada data refer to Ontario (1998). USA data refer to CCDF day care in Michigan.

### Quality and Availability of Canadian Child Care (Appendix III)
- Overall trends:
  - Quality and availability of publicly provided childcare improved in recent years.
  - Access: Almost one-third of mothers in paid employment have access to center-based childcare, up from less than one in five a decade ago.
- Quality findings from studies:
  - Johnson and Mathien (1998): seventy percent of kindergartens obtained ratings in the “acceptable to good range.”
  - Studies of regulated childcare providers: most centers provide a healthy and safe environment, but offer few activities that stimulate children’s development.
  - Over recent years, provinces/territories introduced initiatives to enhance quality (higher staff salaries, more training).
- Staffing and standards:
  - Since 1998, five provinces improved childcare staff’s salaries by earmarking funds.
  - All provinces adopted regulation on quality and staff quantity; staff-to-child ratios usually range from 1:3 to 1:5 for infant care.
  - Many provinces (e.g., Québec, Ontario, Manitoba) require at least two-thirds of staff to have a university degree in early childhood education.
  - Comparative notes:
    - In Sweden and Finland about half and one-third of preschool staff hold higher education degrees, respectively (NAER, 2004).
    - In Canada, more than four out of five childcare workers have at least one year of postsecondary education in early childhood education (OECD, 2004b).
- Public spending trend:
  - Figure 1: Canada: Public Spending for Regulated Child Care Per Child, Ages 0–12, 1992–2004 (1992=100). Source: Friendly and Beach, 2005.

### Childcare Provision in Québec (Appendix IV)
- Policy introduction and objectives:
  - Québec introduced comprehensive family support policy in 1997, including public funding for universal childcare.
  - Policy objectives: fight poverty, increase mothers’ labor participation, enhance child development (MESSF, 2003, 2004).
- Fiscal magnitude:
  - Québec public funding for universal childcare comprised ¾ percent of GDP in 2001, up from ¼ percent in 1995 (Appendix IV, Table 1).
- Fees and access:
  - Parental contribution initially set at C$5 per day per child (now C$7) for up to 10 hours; access to childcare spaces increased significantly.
  - With regulated childcare at a cost of C$7 per day for up-to 10 hours, Quebec residents enjoy cheap childcare by international standards (Appendix IV, Table 2).
  - International fee comparisons cited: maximum childcare fee in Finland around EUR 200 per month (covering around 16 percent of total cost); Sweden around EUR 140 (covering around 11 percent of total cost). Quebec’s childcare fees cover around one-fifth of total cost and account for only one-fourth of the amount that the rest of Canadian parents pay.
  - Example: in rest of Canada in 2001, a single-earner family with a child in childcare would have a monthly gross earning of around C$3,000 and pay C$400–600 as a childcare fee if no subsidy is offered.
- Capacity and coverage:
  - Childcare spaces for age group 0–5 more than doubled: around 80,000 in 1997 to 180,000 in March 2004 (expected to reach 200,000 by 2006 for 370,000 children).
  - Proportion of children (aged 0–12) in formal childcare increased from 19 percent to 45 percent (OECD, 2005, p. 118).
  - Québec accounts for 40 percent of Canadian childcare capacity while its population of young children is only 22 percent of the Canadian total.
- Labor market outcomes:
  - Maternal employment (mothers with youngest child aged below 3 years):
    - By 2001 at 61 percent in Québec; an increase of almost 7 percent following reforms.
  - Full-time dual-earner families with dependent children:
    - Québec moved from less than 40 percent in 1996 to about one out of two families being full-time dual earners by 2001.
    - Comparative figures: proportion of dual-earner full-time families is 45 percent in Canada as a whole, 40 percent in Sweden, and around 30 percent in the United Kingdom.
- Supporting tables and figures referenced:
  - Table 1: Public Spending on Childcare Services (selected countries): Finland, Quebec, Sweden, United Kingdom comparisons (values presented in Appendix IV).
  - Table 2: Parental Fees (In percent of APW and net earnings by income group and family type).
  - Figures: Québec childcare coverage and regulated childcare spaces trends.

### Data Analysis (Appendix VI) — Methodology and Key Data Series
- Time period and countries:
  - Time period: 1980–2001.
  - Countries: Australia, Canada, Denmark, Finland, France, Germany, Great Britain, Spain, Sweden, and the United States.
- Key variables, definitions, and data sources:
  - Female labor force participation rate: OECD Labor Force Statistics database. Data interpolated when unavailable.
  - Primary earner’s tax wedge: OECD Taxing Wages Statistics-Historical Tax Rates. Defined as tax wedge of a married household with two children, sole breadwinner earning 100 percent of APW. Data interpolated when missing.
  - Secondary earner’s tax wedge: Jaumotte (2003) updated with OECD Taxing Wages Database. Defined as proportion of additional household earnings that goes into increased household taxes; formula as in Constructing a Tax Wedge (A = single-earner married couple earning 100 percent of APW; B = two breadwinners earning 100 and 67 percent of APW).
  - Benefit wedge: constructed using OECD Social Expenditure Database (2004c) and OECD Labor Force Statistics Database; formula matches Appendix V definition. Missing data interpolated using Jaumotte’s (2003) data on public childcare spending per child.
  - Average production wage (APW): values provided for select OECD countries (US dollars with equal purchasing power). Source: OECD (2002), Taxing Wages. APW list:
    - Australia 35,867
    - Canada 32,488
    - Denmark 36,142
    - Finland 27,988
    - France 23,771
    - Germany 34,260
    - Netherlands 32,747
    - Norway 32,412
    - Spain 20,946
    - Sweden 24,259
    - United Kingdom 29,701
    - United States 32,360
  - Other series and data sources:
    - Union density: OECD Labor Force Statistics database.
    - Wage compression: proxied by ratio of median to 10th percentile of gross earnings for all employed (OECD Labor Force Statistics).
    - Female wage rate: average hourly wages in manufacturing (in PPP) from Gauthier (2003); interpolated and expressed in real terms using CPI.
    - Education enrollment: average years of education of women aged 25 and over (Jaumotte, 2003).
    - Employment protection: index from Allard (2003), covering permanent and temporary contracts.
    - Parental leave: data from Jaumotte (2003) using Gauthier and Bortnik (2001).
    - Proportion of young population: share aged less than 15 years from OECD Labor Force Statistics.
    - Male unemployment rate: OECD Labor Force Statistics.
    - Women parliamentary seats: percentage of seats held by women from Huber and others (2004).
    - Wage gap: ratio of male to female wage rate from Gauthier (2003).

*Italic: Source content unit: _wp0692 - REFERENCES (PDF chapter/section).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0692.pdf_