This paper discusses the classical approach to convergence analysis, focusing on σ-convergence, absolute β-convergence, and conditional β-convergence. The author analyzes data from various regions, including 110 countries, OECD countries, U.S. states, Japanese prefectures, and European regions. The findings indicate that: 1. **σ-Convergence**: The cross-country distribution of world GDP did not shrink between 1960 and 1990, indicating σ-divergence. 2. **Absolute β-Convergence**: Poor economies did not grow faster than rich ones, as evidenced by the lack of σ-convergence. 3. **Conditional β-Convergence**: When variables that proxy for the steady state are held constant, the same sample of 110 economies shows a negative partial correlation between growth and initial GDP, with an estimated speed of conditional convergence close to 2% per year. 4. **OECD Economies**: Convergence in the sense of σ and β is observed, with a speed of convergence close to 2% per year. 5. **U.S. States**: The estimated speed of convergence is also close to 2% per year, with a significant negative correlation between growth and initial per capita income. 6. **Japanese Prefectures**: Absolute β-convergence is observed, with a speed of convergence close to 2% per year. 7. **European Regions**: Absolute and conditional β-convergence are observed, with a speed of convergence close to 2% per year. The paper concludes with four main lessons: 1. σ-convergence and absolute β-convergence did not occur globally. 2. Conditional β-convergence was observed, with a significant speed of convergence. 3. σ-convergence in OECD economies and within countries was observed but stopped around the mid-1970s. 4. The speed of convergence is slow, and the estimated capital share is higher than traditionally assumed.This paper discusses the classical approach to convergence analysis, focusing on σ-convergence, absolute β-convergence, and conditional β-convergence. The author analyzes data from various regions, including 110 countries, OECD countries, U.S. states, Japanese prefectures, and European regions. The findings indicate that: 1. **σ-Convergence**: The cross-country distribution of world GDP did not shrink between 1960 and 1990, indicating σ-divergence. 2. **Absolute β-Convergence**: Poor economies did not grow faster than rich ones, as evidenced by the lack of σ-convergence. 3. **Conditional β-Convergence**: When variables that proxy for the steady state are held constant, the same sample of 110 economies shows a negative partial correlation between growth and initial GDP, with an estimated speed of conditional convergence close to 2% per year. 4. **OECD Economies**: Convergence in the sense of σ and β is observed, with a speed of convergence close to 2% per year. 5. **U.S. States**: The estimated speed of convergence is also close to 2% per year, with a significant negative correlation between growth and initial per capita income. 6. **Japanese Prefectures**: Absolute β-convergence is observed, with a speed of convergence close to 2% per year. 7. **European Regions**: Absolute and conditional β-convergence are observed, with a speed of convergence close to 2% per year. The paper concludes with four main lessons: 1. σ-convergence and absolute β-convergence did not occur globally. 2. Conditional β-convergence was observed, with a significant speed of convergence. 3. σ-convergence in OECD economies and within countries was observed but stopped around the mid-1970s. 4. The speed of convergence is slow, and the estimated capital share is higher than traditionally assumed.