Introduction

The provision of parental leave is a crucial aspect of social policy that significantly impacts both family well-being and economic outcomes. This analysis examines parental leave policies across OECD countries, focusing on both the duration and compensation levels of leave available to mothers and fathers. This topic is particularly relevant for the academic and public debate as it intersects with important policy issues such as gender equality in the workplace, child development, and work-life balance.

Research Question

How do parental leave policies differ between OECD countries, and what do these differences mean for economic policy?

Key observations that frame this question:

  • All but one OECD country offer paid maternity and paternity leave around childbirth at the national level, but countries differ markedly in the duration and payment conditions
  • Some have long paid-leave entitlements, but provide skimpy benefits for a large part of them. Others offer shorter leave periods but at full pay

This analysis aims to understand these variations and their implications for economic policy by examining both the duration and compensation levels of parental leave across OECD countries.

Data

# Load required libraries
library(ggplot2)
library(dplyr)
library(tidyr)

# Read the CSV files
mother_data <- read.csv("PF2_1_Parental_leave_systems mother.csv", sep=";", check.names=TRUE, stringsAsFactors=FALSE)
father_data <- read.csv("PF2_1_Parental_leave_systems father.csv", sep=";", check.names=TRUE, stringsAsFactors=FALSE)

# Clean and prepare the data
names(mother_data)[1:3] <- c("country", "Length_weeks", "Full_rate_equivalent_weeks")
names(father_data)[1:3] <- c("country", "Length_weeks", "Full_rate_equivalent_weeks")

# Remove unwanted countries
remove_countries <- c("Costa Rica", "Malta", "Colombia", "Israel", "Chile", "United States")
mother_data <- mother_data %>%
  filter(!(country %in% remove_countries)) %>%
  mutate(across(c(Length_weeks, Full_rate_equivalent_weeks), 
                ~as.numeric(gsub(",", ".", .)))) %>%
  arrange(Full_rate_equivalent_weeks)

father_data <- father_data %>%
  filter(!(country %in% remove_countries)) %>%
  mutate(across(c(Length_weeks, Full_rate_equivalent_weeks), 
                ~as.numeric(gsub(",", ".", .))))

# Calculate summary statistics
mother_summary <- mother_data %>%
  summarise(
    avg_total_leave = mean(Length_weeks, na.rm = TRUE),
    median_total_leave = median(Length_weeks, na.rm = TRUE),
    avg_paid_leave = mean(Full_rate_equivalent_weeks, na.rm = TRUE),
    median_paid_leave = median(Full_rate_equivalent_weeks, na.rm = TRUE)
  )

father_summary <- father_data %>%
  summarise(
    avg_total_leave = mean(Length_weeks, na.rm = TRUE),
    median_total_leave = median(Length_weeks, na.rm = TRUE),
    avg_paid_leave = mean(Full_rate_equivalent_weeks, na.rm = TRUE),
    median_paid_leave = median(Full_rate_equivalent_weeks, na.rm = TRUE)
  )

he analysis uses data from the OECD’s Family Database, specifically the PF2.1 indicators on parental leave systems. This indicator provides an overview of parental leave systems across OECD and EU countries. Metrics used in research:

  1. Maternity leave (or pregnancy leave):
    • Employment-protected leave for employed women around childbirth or adoption
    • Minimum 14 weeks according to ILO convention
    • Most countries allow combining pre- and post-birth leave
    • Some countries have compulsory pre-birth and post-birth periods
    • Almost all OECD countries provide public income support
    • Some countries (Australia, Iceland, New Zealand, Norway, Sweden) integrate it into parental leave
  2. Paternity leave:
    • Employment-protected leave for employed fathers shortly after childbirth
    • Generally much shorter than maternity leave
    • Often includes full wage payments due to short duration
    • Not stipulated by international convention
    • In some countries (e.g., Iceland), integrated into parental leave scheme

Key statistics:

For mothers:

  • Average total paid leave: 59 weeks

  • Median total paid leave: 52 weeks

  • Average full-rate equivalent: 34.6 weeks

  • Median full-rate equivalent: 34.4 weeks

For fathers:

  • Average total paid leave: 13.6 weeks

  • Median total paid leave: 11 weeks

  • Average full-rate equivalent: 8.1 weeks

  • Median full-rate equivalent: 6.8 weeks

Result

# Identify EU countries
eu_countries <- c("Austria", "Belgium", "Bulgaria", "Croatia", "Cyprus", "Czechia", 
                  "Denmark", "Estonia", "Finland", "France", "Germany", "Greece", 
                  "Hungary", "Ireland", "Italy", "Latvia", "Lithuania", "Luxembourg", 
                  "Netherlands", "Poland", "Portugal", "Romania", "Slovakia", 
                  "Slovenia", "Spain", "Sweden")

# Calculate averages for Total Paid Leave
avg_mothers_EU <- mean(mother_data$Length_weeks[mother_data$country %in% eu_countries], na.rm = TRUE)
avg_fathers_EU <- mean(father_data$Length_weeks[father_data$country %in% eu_countries], na.rm = TRUE)

# Create ordered factor for countries based on Full-paid equivalent for mothers
country_order <- mother_data$country

# Prepare data in long format
mother_data_long <- bind_rows(
  mother_data %>%
    mutate(country = factor(country, levels = country_order),
           type = "Total paid leave", weeks = Length_weeks, parent = "Mothers") %>%
    select(country, type, weeks, parent),
  mother_data %>%
    mutate(country = factor(country, levels = country_order),
           type = "Full-paid equivalent", weeks = Full_rate_equivalent_weeks, parent = "Mothers") %>%
    select(country, type, weeks, parent)
)

father_data_long <- bind_rows(
  father_data %>%
    mutate(country = factor(country, levels = country_order),
           type = "Total paid leave", weeks = -Length_weeks, parent = "Fathers") %>%
    select(country, type, weeks, parent),
  father_data %>%
    mutate(country = factor(country, levels = country_order),
           type = "Full-paid equivalent", weeks = -Full_rate_equivalent_weeks, parent = "Fathers") %>%
    select(country, type, weeks, parent)
)

# Combine data and reorder the legend
all_data <- bind_rows(mother_data_long, father_data_long) %>%
  mutate(fill_group = factor(paste(parent, type),
                             levels = c("Mothers Total paid leave",
                                        "Mothers Full-paid equivalent",
                                        "Fathers Total paid leave",
                                        "Fathers Full-paid equivalent")))

# Create the plot
ggplot(all_data, aes(x = country, y = weeks, fill = fill_group)) +
  geom_col(data = subset(all_data, type == "Total paid leave"), width = 0.7, alpha = 0.5) +
  geom_col(data = subset(all_data, type == "Full-paid equivalent"), width = 0.7) +
  geom_hline(yintercept = avg_mothers_EU, linetype = "dashed", color = "gray25", linewidth = 1) +
  geom_hline(yintercept = -avg_fathers_EU, linetype = "dashed", color = "gray25", linewidth = 1) +
  coord_flip() +
  scale_y_continuous(
    breaks = seq(-100, 200, 50),
    labels = abs,
    limits = c(-100, 200)
  ) +
  scale_fill_manual(
    name = NULL,
    values = c(
      "Mothers Total paid leave" = "#FFCCCC",
      "Mothers Full-paid equivalent" = "#E6550D",
      "Fathers Total paid leave" = "#B3D9FF",
      "Fathers Full-paid equivalent" = "#3182BD"
    )
  ) +
  labs(
    title = "Which countries lead in supporting new parents?",
    subtitle = "Parental-leave packages across the OECD, in weeks",
    x = "",
    y = "Weeks",
    caption = "Source: OECD. Figure: Tatiana Zubareva"
  ) +
  theme_minimal() +
  theme(
    plot.title = element_text(size = 24, face = "bold", hjust = 0),
    axis.title.x = element_text(size = 18),
    plot.subtitle = element_text(size = 20, hjust = 0),
    axis.text.y = element_text(size = 18),
    axis.text.x = element_text(size = 18),
    legend.text = element_text(size = 18),
    legend.position = c(0.8, 0.25),
    legend.background = element_rect(fill = "white", color = NA),
    panel.grid.major.y = element_blank(),
    panel.grid.minor.y = element_blank(),
    plot.caption = element_text(size = 14, hjust = 0)
  )

The visualization reveals several key patterns in parental leave policies across OECD countries:

  1. Leading Countries: Eastern European countries, particularly Romania, Estonia, and Hungary, offer the most generous parental leave packages when measured by full-rate equivalent weeks.

  2. Gender Disparities: There is a significant disparity between leave available to mothers versus fathers across most countries, with mothers typically having access to longer periods of paid leave.

  3. Regional Patterns:

    • Eastern European countries tend to have more generous policies
    • Anglo-Saxon countries (UK, Ireland, Australia) generally offer less generous packages
    • Nordic countries (Sweden, Norway, Finland) show more balanced policies between mothers and fathers

  4. Policy Approaches: Some countries offer longer periods of leave at lower pay rates (shown by the difference between total paid leave and full-rate equivalent), while others provide shorter but better-compensated leave periods.

Conclusion

The analysis reveals several key patterns in parental leave policies across OECD countries:

  1. Regional Leadership in Central and Eastern Europe: Countries in Central and Eastern Europe offer some of the most generous parental leave packages, driven by demographic concerns over rapidly shrinking populations. For example, in Estonia and Hungary, new mothers can stay at home for three years, receiving total pay equivalent to about a year and a half of the average salary. These generous policies appear to be part of broader government strategies to encourage higher birth rates.

  2. Variation Among Wealthy Nations: There is significant disparity among wealthy OECD countries. While some offer extensive leave periods, others like Australia, Britain, Ireland, and Switzerland provide relatively modest benefits, offering mothers leave equivalent to three months or less of full pay. This suggests that economic prosperity doesn’t necessarily correlate with more generous leave policies.

  3. Paternity Leave Evolution: While most OECD countries now provide some paid leave reserved for fathers, only a minority offer longer than two weeks. Japan stands out as particularly generous to fathers, offering them a full year at a rate equivalent to 34 full-pay weeks. However, this case also illustrates the importance of cultural factors - despite the generous policy, only about 2% of new fathers in Japan take advantage of it due to cultural stigma.

  4. Policy Implementation Challenges: The Japanese example highlights how policy effectiveness depends not just on generous provisions but also on cultural and workplace norms. This suggests that successful parental leave policies must address both financial and social barriers to their uptake.

  5. Policy Balance Trade-offs: Countries take different approaches to balancing leave duration and compensation levels. Some offer extended periods with lower payment rates, while others provide shorter but better-compensated leaves, highlighting different philosophies in supporting new parents.

These variations in parental leave policies reflect different national priorities, demographic challenges, and approaches to work-life balance. The data suggests that policy makers must consider not just the length of leave offered, but also compensation levels, cultural factors, and the balance between maternal and paternal leave when designing effective family support systems.