China's government has unveiled a new measure of unemployment, acknowledging doubts about the accuracy of official reports that move markets around the world.
Last month, a top economic official announced that China's "polled unemployment rate" in March had dropped slightly from the previous month to 5.17 percent, state media reported, without giving the February figure.
The statement by Li Pumin, secretary-general of the administrative National Development and Reform Commission (NDRC), was the first reference to the new jobless measurement since last September, when Premier Li Keqiang cited "surveyed unemployment at around 5 percent" in a commentary for the London-based Financial Times.
The new figure marks a departure from China's practice of reporting "urban registered unemployment," an official gauge that has stayed suspiciously fixed at about 4.1 percent for years.
"That number has long been criticized as a biased and deflated one," the official English-language China Daily admitted in a report last month.
Since at least 2006, the rate has ranged only slightly between 4.0 and 4.3 percent, according to International Monetary Fund records. Under the last five-year plan ending in 2010, registered urban unemployment averaged exactly 4.1 percent, the National Bureau of Statistics (NBS) said.
The index does not reflect joblessness among some 630 million citizens listed as rural residents, or 269 million migrant workers, or those who are unregistered, making it an indicator of little value.
Given its concerns about stability and higher unemployment, the government's decision to go public with the new survey-based measure may be seen as a relatively bold move.
But it also renews questions about the transparency of NBS economic data and the reliability of its published reports.
One reason for shifting to the new index is that the old one had long since ceased to be believable.
"Even in 2009, a year when massive layoffs were common, the rate stood at only 4.3 percent," China Daily said in another report this month.
The reform can be compared to the government's belated recognition of the urban smog problem after years of reporting on "blue sky days."
The government only agreed to start providing readings of smog-forming fine particulates known as PM2.5 in 2012 after years of public pressure over visible pollution that could no longer be denied.
As with the smog crisis, official credibility may suffer if the government fails to give a better accounting on jobs.
Hidden from the public?
But reporting on the new index suggests the government has been keeping it from the public for years, although the NBS has been conducting monthly surveys for over a decade.
"The number is currently available only to policymakers and has not been made public," China Daily said in April.
"Actually, [the] statistics agency has been compiling the indicator since the 1990s as a reference for policymakers," the paper reported in May.
The disclosure may answer a larger question about doubtful NBS accounting that has long concerned China economists.
Do top officials have access to better data when they make economic policy decisions?
Derek Scissors, resident scholar at the American Enterprise Institute in Washington, said the answer appears to be yes.
"It fits with the consensus guesswork that the State Council [or cabinet] has better data than published," said Scissors. "That it's closely held is far from ideal."
The NBS has struggled for years against inaccurate and sometimes fraudulent economic reporting.
The problem is seen most often in estimates of gross domestic product (GDP), the broadest measure of the economy.
The agency regularly adjusts the numbers with an undisclosed formula, since totals from the country's provinces routinely exceed national GDP.
The NBS has tried to crack down on data fabrication by officials seeking to advance their careers. In one noted case uncovered by a National People's Congress probe in 2009, officials in Chongqing municipality inflated production figures of an enterprise by a factor of 10.
In 2012, NBS launched a direct reporting system for 700,000 businesses to avoid local falsification, but periodic problems have continued.
Last June, for example, the NBS found a town in southern Guangdong province was reporting production from 19 industrial enterprises that had shut down or moved, according to the ruling Chinese Communist Party-affiliated Global Times.
Trade data from the General Administration of Customs (GAC) has also been subject to distortion, notably in early 2013, when a reported export surge turned out to be the result of fake invoices, used to hide speculative monetary flows.
The problems have raised concerns that China's policymakers and regulators have been relying on concocted data in responding to the economy's slower-growth phase.
The new jobs number may ease some of that worry. But it may also raise concerns that the NBS has been keeping two sets of books, providing more accurate numbers to China's leaders while releasing false figures to the public and markets abroad.
Even if officials have access to better numbers, the conflicting reports on the new jobs index will hardly put concerns about Chinese data to rest.
For one thing, little is known about how or where the unemployment survey figures are compiled.
"Not much information has been released about the indicator, but it has been made clear that the index covers 31 large and medium-sized cities and is updated monthly," said a Shanghai-based financial analyst in the China Daily report this month.
But in April, the paper said the index was based on "a monthly survey in 65 major cities." Neither report specifies whether it includes migrants, the unregistered unemployed, newly-graduated students, or private firms.
Of little use
In any case, the new number may be useless as an indicator unless it is part of a consistent series of regular reports.
Last September, Premier Li said only that surveyed unemployment was "around 5 percent" in the first half of 2013, making it impossible to tell whether it has gone up or down.
Li's commentary in the Financial Times focused on the government's policy of keeping economic performance between "upper and lower limits" while resisting pressure for massive stimulus programs, arguing that the surveyed unemployment figure was "within the reasonable and manageable range."
Scissors said the government may have judged that it was time to concede that it had a better estimate of joblessness as a basis for its policies than the fixed 4.1-percent rate that it has publicized for years, still without being too precise about it.
"I think the timing is that it is now safe to say unemployment is a bit higher than the registered rate, since employment pressure is now less intense than during the previous decade," Scissors said.
The vagueness of the new jobless reporting also suggests that the government does not want to be pinned down or exposed to more public pressure on policy decisions by providing any more precise data on jobs.