One of the most-commented on features of Japanese politics is the short lifespans of Japanese premierships since the long rule of Koizumi Jun'ichiro, abetted by the spectacular meltdowns of Cabinet popularity ratings in the months following the election of a new prime minister.
(Source - Kyodo figures with typo uncorrected)
Perhaps we have been looking at these curves on these graphs the wrong way. Rather than depicting the popularity of a prime minister precipitously declining over time, perhaps the graph shows the unreasonably high expectations of the public at the inauguration, followed by an actually quite gradual fall of each Cabinet's popularity to a low-level norm, conditioned by economic stagnation, an unhealthy number of workers in temporary or contract work, demographically-determined economic and social challenges and the poor stewardship of the nation's wealth in the pre-Koizumi era.
What is striking when one reads the responses to the questions of why a particular prime minister is not supported is the stability of the answers given. Those not supporting the Cabinet explain their views as the result of a lack of faith that the government delivering on policy, that the government is pursuing the wrong policies or that the prime minister lacks leadership. These explanations remain in the same relative positions even as the identity or personal style of the prime minister changes or the Cabinet's popularity travels drifts down the curve. They are always in double digits in the polls, whilst all other explanations remain in the single digits.
We should consider as likely that these three explanations are the three legs of an idée fixe regarding the prime minister and his cabinet. What the curves are indicating is a return to a normal level of cynicism based upon the three principles, completely disassociated from the actual policies or leadership skills of the prime minister, with rare exceptions.
The current curve of the popularity of Prime Minister Noda's premiership is proceeding according to this internal dynamic. Noda is doggedly pursuing a radical path of forcing a doubling of the consumption tax through the Diet, a choice that should precipitate a fall in popularity even more severe than the one we are seeing. Compare the Noda curve with that of the popularity of Kan Naoto, which suffered a sharp fall after Kan merely mentioned the possibility of raising the consumption tax. In addition, whereas Kan suffered from a fully twisted Diet and a vicious campaign to undermine him led by Ozawa Ichiro for most of his tenure, his path was not unsimilar to his four immediate predecessors who had either a supermajority in the House of Representatives or, in Abe Shinzo's and Hatoyama's case, majorities in both houses of the Diet (Abe indeed had both). If actual performance mattered rather than perceived performance, the curves of all the prime ministers should have been different. Furthermore, the prime minister in the weakest political position, with the least-liked policy changes and the least chance of passing them, survived the longest -- though it cannot be denied that Kan's tenure was extended by the disaster of 3/11.
All of which suggests that the norm for voters is mistrust of the cabinet. The initial high popularity ratings of a new cabinet represent not an actual feeling of hope in the new person taking over as prime minister but a shrugging temporary suspension of an ingrained feeling of pessimism and despair.