This notebook presents a summary dashboard drawing on the analyses from Notebooks 01-05. Rather than introducing new data, it synthesizes the key findings into a coherent narrative about US-Europe inequality differences.
import sys
sys.path.insert(0, '..')
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.ticker as mtick
import matplotlib.gridspec as gridspec
import seaborn as sns
from data.config import FOCUS_COUNTRIES, COUNTRY_NAMES
from data.plotting import (set_style, country_color, country_name,
save_figure, annotate_countries)
from data import swiid, wid, health_social
set_style()A summary comparison table of key inequality metrics for the focus countries.
summary_countries = FOCUS_COUNTRIES
target_year = 2020
# Disposable income Gini
gini_disp = swiid.get_gini(summary_countries, gini_type='disp',
start_year=2017, end_year=2022)
# Market income Gini
gini_mkt = swiid.get_gini(summary_countries, gini_type='mkt',
start_year=2017, end_year=2022)
# Top 10% pre-tax income share
top10_pretax = wid.get_income_shares(
summary_countries, income_concept='pretax',
percentiles=['p90p100'], start_year=2017, end_year=2022)
# Top 10% wealth share
top10_wealth = wid.get_wealth_shares(
summary_countries, percentiles=['p90p100'],
start_year=2017, end_year=2022)
# Build summary table
rows = []
for cc in summary_countries:
row = {'Country': country_name(cc)}
# Disposable Gini
if not gini_disp.empty:
d = gini_disp[gini_disp['country_code'] == cc]
if not d.empty:
row['Gini (disp.)'] = f"{d.iloc[-1]['gini']:.2f}"
# Market Gini
if not gini_mkt.empty:
m = gini_mkt[gini_mkt['country_code'] == cc]
if not m.empty:
row['Gini (mkt.)'] = f"{m.iloc[-1]['gini']:.2f}"
# Redistribution
if 'Gini (disp.)' in row and 'Gini (mkt.)' in row:
mkt_val = float(row['Gini (mkt.)'])
disp_val = float(row['Gini (disp.)'])
if mkt_val > 0:
row['Redistribution'] = f"{(mkt_val - disp_val)/mkt_val*100:.0f}%"
# Top 10% income share
if not top10_pretax.empty:
t = top10_pretax[top10_pretax['country'] == cc]
if not t.empty:
row['Top 10% income'] = f"{t.iloc[-1]['value']*100:.1f}%"
# Top 10% wealth share
if not top10_wealth.empty:
w = top10_wealth[top10_wealth['country'] == cc]
if not w.empty:
row['Top 10% wealth'] = f"{w.iloc[-1]['value']*100:.1f}%"
rows.append(row)
summary_df = pd.DataFrame(rows)
if not summary_df.empty:
display(summary_df.set_index('Country'))
else:
print('Could not build summary table.')| Gini (disp.) | Gini (mkt.) | Redistribution | Top 10% income | Top 10% wealth | |
|---|---|---|---|---|---|
| Country | |||||
| United States | 38.90 | 53.00 | 27% | 46.9% | 69.6% |
| United Kingdom | 32.00 | 52.80 | 39% | 35.8% | 57.1% |
| France | 29.80 | 51.20 | 42% | 34.4% | 59.6% |
| Germany | 30.40 | 52.70 | 42% | 37.5% | 58.5% |
| Sweden | 28.90 | 49.40 | 41% | 29.3% | 68.2% |
| Denmark | 28.80 | 49.10 | 41% | 32.8% | 50.7% |
| Norway | 26.10 | 47.40 | 45% | 29.9% | 53.0% |
A 2x2 grid of key visualizations from the analysis.
fig = plt.figure(figsize=(16, 12))
gs = gridspec.GridSpec(2, 2, hspace=0.35, wspace=0.3)
# Panel A: Disposable Gini bar chart
ax1 = fig.add_subplot(gs[0, 0])
if not gini_disp.empty:
latest = (gini_disp.sort_values('year', ascending=False)
.groupby('country_code').first().reset_index()
.sort_values('gini'))
colors = [country_color(c) for c in latest['country_code']]
ax1.barh([country_name(c) for c in latest['country_code']],
latest['gini'], color=colors, alpha=0.8)
ax1.set_xlabel('Gini coefficient')
ax1.set_title('A. Disposable Income Inequality', fontsize=12, fontweight='bold')
# Panel B: Pre-tax top 10% share time series
ax2 = fig.add_subplot(gs[0, 1])
ts_countries = ['US', 'FR', 'DE', 'SE', 'GB']
pretax_top10 = wid.get_top_shares_timeseries(
ts_countries, percentile='p90p100', concept='pretax', start_year=1980)
if not pretax_top10.empty:
for cc in ts_countries:
data = pretax_top10[pretax_top10['country'] == cc].sort_values('year')
if not data.empty:
ax2.plot(data['year'], data['value'],
color=country_color(cc), linewidth=2,
label=country_name(cc))
ax2.yaxis.set_major_formatter(mtick.PercentFormatter(1.0))
ax2.legend(fontsize=8)
ax2.set_title('B. Pre-Tax Top 10% Share (since 1980)', fontsize=12, fontweight='bold')
# Panel C: Redistribution — market vs disposable Gini
ax3 = fig.add_subplot(gs[1, 0])
redist = swiid.get_redistribution(summary_countries, start_year=2017, end_year=2022)
if not redist.empty:
latest_r = (redist.sort_values('year', ascending=False)
.groupby('country_code').first().reset_index())
latest_r = latest_r.dropna(subset=['gini_mkt', 'gini_disp'])
latest_r = latest_r[latest_r['country_code'] != ''].sort_values('gini_disp')
y_pos = np.arange(len(latest_r))
ax3.barh(y_pos, latest_r['gini_mkt'].values, 0.4,
label='Market Gini', color='#c75b5b', alpha=0.6)
ax3.barh(y_pos + 0.4, latest_r['gini_disp'].values, 0.4,
label='Disposable Gini', color='#4878a8', alpha=0.8)
ax3.set_yticks(y_pos + 0.2)
ax3.set_yticklabels([country_name(c) for c in latest_r['country_code']])
ax3.set_xlabel('Gini coefficient')
ax3.legend(fontsize=8)
ax3.set_title('C. Market vs. Disposable Inequality', fontsize=12, fontweight='bold')
# Panel D: Gini vs life expectancy
ax4 = fig.add_subplot(gs[1, 1])
spirit_data = health_social.get_spirit_level_data(year=2019)
if (not spirit_data.empty and 'gini_disp' in spirit_data.columns
and 'life_expectancy' in spirit_data.columns):
df = spirit_data.dropna(subset=['gini_disp', 'life_expectancy'])
colors = [country_color(c) if c in COUNTRY_NAMES else '#999'
for c in df['country_code']]
ax4.scatter(df['gini_disp'], df['life_expectancy'],
c=colors, s=80, zorder=5, edgecolors='white', linewidth=0.5)
annotate_countries(ax4, df['gini_disp'].values,
df['life_expectancy'].values,
df['country_code'].values)
ax4.set_xlabel('Gini (disposable)')
ax4.set_ylabel('Life expectancy')
ax4.set_title('D. Inequality vs. Life Expectancy', fontsize=12, fontweight='bold')
fig.suptitle('US-Europe Inequality: A Summary Dashboard', fontsize=16, fontweight='bold', y=1.01)
save_figure(fig, '06_synthesis_dashboard')
plt.show()
Each chart corresponds to one of the core questions from our analysis.
Different inequality measures can rank countries differently. The Gini coefficient is sensitive to the middle of the distribution, while top income shares capture concentration at the top. The US looks more unequal by top shares than by Gini, because its inequality is concentrated at the very top.
Takeaway: Always ask which measure when someone says “inequality is higher.”
Europe achieves lower inequality through both more equal market incomes (predistribution) and more aggressive tax-transfer systems (redistribution). The US has higher pre-tax inequality and less redistribution — a double disadvantage.
The Blanchet et al. (2022) finding that predistribution dominates depends on counting social insurance as pre-tax income. The conceptual debate is about where to draw the line between market and state.
Takeaway: The answer depends on how you define “the market” vs. “the state.”
The US has higher absolute income levels at the top (P90 and above), but the gap narrows or reverses at the bottom (P10). US income growth since 1980 has been heavily skewed toward the top, while European growth has been more evenly distributed.
PPP comparisons also miss the “social wage” — Europeans receive public services (healthcare, childcare, education) that don’t appear in disposable income figures.
Takeaway: “Americans are richer” is true at the top but misleading at the bottom.
Nordic countries have low income inequality but high wealth concentration. This paradox arises because universal public services and pensions reduce the need for private savings, paradoxically concentrating measured wealth among the wealthy.
Takeaway: Income and wealth inequality are distinct phenomena with different drivers.
More unequal rich countries tend to have worse health and social outcomes, but the relationship is fragile — sensitive to country selection, time period, and whether you control for GDP per capita. The US is an extreme outlier that drives many of the correlations.
Takeaway: The Spirit Level thesis is suggestive but far from proven.
1. What does “more unequal” actually mean?
It depends on the measure. The Gini coefficient, top income shares, the Palma ratio, and percentile ratios each capture different aspects of the distribution. They can rank countries differently, and the choice of measure is itself a normative decision about which part of the distribution matters most.
2. Relative vs. absolute: how does the picture change?
The US looks much better on absolute measures at the top and much worse at the bottom. Growth incidence curves since 1980 show that US growth has gone disproportionately to the top, while European growth has been more shared. PPP comparisons also miss the value of European public services.
3. Does inequality damage society?
The evidence is suggestive but contested. Cross-country correlations exist but are fragile. The causal mechanisms (psychosocial stress, material deprivation, political economy, social capital) are plausible but hard to disentangle. The strongest statement the evidence supports is that very high inequality is associated with worse outcomes, not that it causes them.
Inequality is not a single number. It’s a multi-dimensional phenomenon that looks different depending on: - Which measure you use (Gini, top shares, absolute levels) - Which income concept you use (market, disposable, including services) - Whether you focus on income or wealth - Whether you look at levels or changes over time - Where you draw the line between market outcomes and state intervention
Good analysis requires being explicit about these choices rather than treating “inequality” as a simple, self-evident concept.