YESTERDAY KYODO released the results of its monthly public opinion poll conducted over the weekend showing that support for the Fukuda Cabinet plunged 11.1 percentage points to 35.3%. It was also the first survey that showed the percentage of people who disapproved of the Cabinet exceeding the percentage supporting it.
Those who follow Japanese politics will not be surprised at the reason for the sharp downturn. Earlier last week, the government announced it would not be able to resolve the mess over mishandled pension accounts by next March, as they had promised, and that some accounts might never be straightened out.
A total of 57.6% of the poll respondents said the Cabinet had broken its pledge for a full resolution. The prime minister surely did himself no favors when he retorted that it was an exaggeration to say the government’s inability to deal with the problem constituted a broken promise.
One would have thought that Mr. Fukuda had more discretion, but then it’s not possible to go broke underestimating Japanese politicians. Everyone in Nagata-cho should have realized long ago that the public will severely punish any politician who fails to treat this issue with the utmost care.
Japanese Poll Volatility
Other results from the same poll, however, make one wonder just how seriously the Kyodo poll–or any of the mass media polls in this country–can be taken.
Consider the response to the question about the new bill to resume the Indian Ocean refueling mission by Japan’s Self-Defense Forces to support the NATO-led military operation in Afghanistan. A total of 46.7% of the respondents are opposed, while 38.8% are in favor.
Yet, the same poll shows that 41.2% would approve if the lower house passed the bill by overriding an upper house rejection, while 43.6% would be opposed. In other words, considering the margin of error, opinion is split roughly 50-50 when the question is put into a different context.
If the polls are to be believed, the public’s position on the refueling mission has switched from opposition to support and now back to opposition again in just three months.
Since there has been no change in situation in the Indian Ocean or Afghanistan, the public’s opinion of the issue is likely to have been influenced by unrelated factors, or even by the polling methods themselves. Numbers that unstable in the absence of significant events call into question a poll’s reliability to accurately gauge public sentiment.
The Public View of the DPJ
Consider also the numbers showing a large swing in support from the ruling Liberal Democratic Party to the opposition Democratic Party of Japan. According to Kyodo, only 28.5% now want an LDP-led government, while 44.7% say they would like to see a DPJ-led government.
But let’s review the events of this autumn. Six weeks ago, the opposition DPJ became embroiled in a political farce when party president Ichiro Ozawa quit during a televised press conference after party elders questioned his leadership. Mr. Ozawa broadly hinted that he would form an alliance with the LDP, only to return to head the opposition and resume blustering at the LDP three days later.
The party’s problems are compounded by their inability to formulate a coherent policy on major issues that all of its members would be comfortable with supporting. This is common knowledge in Japan.
Yet, just one month after the previous poll, the public now thinks the DPJ would be the better choice to lead a government? Here too, the results of public opinion polling showing support for and disapproval of the two mainstream parties has spun back and forth like a weathervane since the upper house elections in July.
One reason Mr. Ozawa got into trouble in the first place was because he saw more sophisticated private polling that indicated his party could not capture a lower house majority in the next election, thereby earning the right to form a government. These numbers have probably not changed.
The only sensible way to view these polls is to consider them as a way to take the temperature of the body politic–that is, if you think someone’s temperature needs to be taken every three hours, in sickness and in health.
Variation in Results
Here’s more food for thought: The Nikkei Shimbun released its own poll on Monday. It too shows the disapproval rating for the Cabinet higher than the approval rating, but by a smaller 46% to 43% margin. There is an eleven-point differential in the approval rating alone between the two polls.
Differences of this magnitude in the results of political polls conducted by major Japanese media outlets are commonplace. The polls from the two country’s two largest newspapers, the Asahi and the Yomiuri, almost invariably have a ten-point differential. It is likely that the most accurate public opinion surveys in Japan are those conducted by the parties themselves–which the public never sees—and the election results.
It can be argued that the public’s fickleness should be expected because neither party consistently offers a clear choice. While there are elements in both the LDP and the DPJ working to create an ideologically discrete two-party system, they have yet to fully emerge.
For example, the LDP has a growing semi-libertarian, semi-Reaganite/Thatcherite wing, but they are not yet strong enough to have gained undisputed control in the party.
In the same way, the opposition DPJ is clearly moving toward a policy of big government. One plank of their platform, called a manifest in Japan, advocates eliminating the tax break parents receive for children and replacing it with an outright monthly payment from the government to the parents. Just this week, they settled on a policy of giving government financial support to students who choose to attend private high schools.
The Real Purpose of Media Polling
Nevertheless, the problems resulting from the marriage of mass media and political polling are universal. For a case in point, Peter Hitchens dissects a British poll in detail to show that “polls…are not devices for measuring public opinion – they are devices for influencing it”. He also makes the same point that I’ve made frequently: “This is the story a lot of papers want, some to keep the flagging, desperately dull soap opera of Westminster alive (because vast numbers of political reporters make their living out of it).”
It happens in Japan, too: the media hounded former Prime Minister Yoshiro Mori out of office, and tried to do the same to Shinzo Abe.
The principle is the same as everywhere else: Public opinion polls keep the drama alive, and therefore keep the consumers reading. They also offer the media outlet a chance to influence the public.
Why should different media outlets conduct polls on the same question and get such widely divergent answers? The reason is almost always found in the political orientation of the pollsters.
Here again, the poll results from the Asahi and the Yomiuri illustrate the problem. The Asahi’s results tend to buttress its left wing views, while those of the Yomiuri have a definite slant to the right. That means neither are credible.
Thus, it’s not surprising that the polling figures from the Nikkei, Japan’s equivalent of the Wall Street Journal, show a more moderate disapproval rating for the Fukuda Cabinet than do those from Kyodo, which as a news agency is more interested in turning politics into a horse race.
When all the polls show similar numbers in the answer to the same questions–such the survey results just before the upper house elections–they are probably trustworthy.
But in Japan, that convergence of polling results seldom occurs.