In economic growth theory, the idea of convergence is the hypothesis that poorer economies' per capita incomes will tend to grow at faster rates than richer economies. As a result, all economies should eventually converge in terms of per capita income. In the literature, the idea has been applied not only to convergence across countries but also to convergence across regions (Barro and Sala-i-Martin (1991) "Convergence across States and Regions,” Brookings Papers on Economic Activity). Using the approach of Barro and Sala-i-Martin, there are several studies that found significant conditional β convergence across regions in Japan. However, these studies were based on the sample period when Japan experienced positive population growth. Thus it is far from clear whether the results are still robust when the population growth rate turns negative. In Japan, population aging is now becoming the most serious structural problem. The ratio of people aged 65 years or over to the total population reached 26.7% in 2015. The figure is expected to rise to about 40% by around 2060. The situation is more serious in rural areas than in urban areas. Therefore, it is very important to see whether convergence still exists across regions in Japan where the working age population have started to show a sharp decline. In the analysis, we apply various existing methods to test “β convergence”. We find that the β coefficient, which had significantly positive values (around 0.01 to 0.02) until the early 2000s, became insignificant in the 2000s. This implies that regional convergence in Japan, which had robustly held true until the early 2000s, no longer held true in the 2000s. In other words, a number of prefectures in Japan started to show weak divergence, not convergence, when the growth rate of the working age population turned negative. We discuss that this happened because in societies with aging populations, diminishing returns became more serious in poorer regions than in richer regions. We also discuss why labor market in Japan is tighter in urban areas than in rural areas even if working-age population is fewer in rural areas than in urban areas.