[GDP] The North-South Italy gap: is it shrinking or growing? (2000–2024)

[Gabrymi93]

Claim

Italy's North-South GDP per capita gap has persisted for decades — but is it getting better or worse?

Data

Dataset: eurostat_gdp_nuts3 — GDP at current market prices by NUTS 3 region.
Unit: EUR_HAB (euro per inhabitant) at NUTS2 level (regions).

Trend: North vs South average GDP per capita

Year	North avg (€)	South avg (€)	Gap (€)	Gap (%)
2000	26,463	15,263	11,200	73.4%
2005	30,513	18,000	12,513	69.5%
2010	32,250	19,000	13,250	69.7%
2015	33,588	19,375	14,213	73.4%
2020	34,300	19,463	14,838	76.2%
2024	45,275	26,250	19,025	72.5%

North = Piemonte, Lombardia, Liguria, Veneto, Emilia-Romagna, Friuli-Venezia Giulia, Trentino, Bolzano, Valle d'Aosta
South = Abruzzo, Molise, Campania, Puglia, Basilicata, Calabria, Sicilia, Sardegna

What the data says

• The absolute gap has grown from €11,200 to €19,025 (+70% in 24 years)
• The relative gap fluctuates around 70-76% — no clear convergence
• Both North and South grew, but the North grew faster in absolute terms
• The COVID dip (2020) was shallower in the South, but the recovery (2024) widened the gap again

For context: EU average GDP per capita (NUTS2)

Year	EU avg (€)	Italy North	Italy South
2000	15,713	168% of EU avg	97% of EU avg
2024	34,605	131% of EU avg	76% of EU avg

The North is converging downward toward the EU average, while the South is diverging downward.

Query to reproduce

SELECT 
  year,
  ROUND(AVG(value) FILTER (WHERE area = 'North'), 0) AS north_avg,
  ROUND(AVG(value) FILTER (WHERE area = 'South'), 0) AS south_avg,
  ROUND(AVG(value) FILTER (WHERE area = 'North') - AVG(value) FILTER (WHERE area = 'South'), 0) AS gap_abs,
  ROUND((AVG(value) FILTER (WHERE area = 'North') - AVG(value) FILTER (WHERE area = 'South')) 
    / AVG(value) FILTER (WHERE area = 'South') * 100, 1) AS gap_pct
FROM (
  SELECT year, geo_label_en, value,
    CASE 
      WHEN geo_label_en IN ('Piemonte','Lombardia','Liguria','Veneto','Emilia-Romagna',
           'Friuli-Venezia Giulia','Provincia Autonoma di Trento',
           'Provincia Autonoma di Bolzano/Bozen','Valle d''Aosta/Vallée d''Aoste') THEN 'North'
      WHEN geo_label_en IN ('Abruzzo','Molise','Campania','Puglia','Basilicata',
           'Calabria','Sicilia','Sardegna') THEN 'South'
      ELSE 'Center'
    END AS area
  FROM data
  WHERE geo LIKE 'IT%' AND unit = 'EUR_HAB' AND nuts_level = 'NUTS2' 
    AND value IS NOT NULL AND year IN (2000,2005,2010,2015,2020,2024)
) 
GROUP BY year ORDER BY year;

Caveats

• NUTS2 level (regions), not NUTS3 (provinces) — EUR_HAB at province level is mostly null
• Current prices, not PPP-adjusted — inflation affects the absolute values
• 2024 data is labelled "provisional" in some cases (check flag_desc_en)

What do you think?

• Why does the gap persist despite decades of EU cohesion funds?
• Which Southern region has the best chance of converging?
• Is NUTS2 the right granularity, or should we look at NUTS3 with other variables?

Let's dig in. 👇