Please view the main text area of the page by skipping the main menu.

Manual for inappropriate statistical method made by labor ministry in 2003

The Central Government Building No. 5 that houses the Health, Labor and Welfare Ministry is seen in this file photo taken in the Kasumigaseki district of Tokyo on Oct. 14, 2015. (Mainichi/Kimi Takeuchi)

TOKYO -- A manual permitting inappropriate methods for carrying out a fundamental statistical survey that led to millions of benefit recipients being underpaid was made at the Ministry of Health, Labor and Welfare in 2003, individuals linked with the ministry have disclosed.

The monthly labor survey manual was created a year before the inappropriate method of surveying only a portion of employers hiring 500 or more people in Tokyo -- rather than all of them as was required -- began being used. This calculation method resulted in roughly 19.7 million recipients of work-related benefits being underpaid by a total of about 56.7 billion yen.

The finding suggests that the section in charge was already preparing to shift to the sampling method in violation of rules for producing such statistics. Moreover, the ministry submitted a false explanation in a document to the Ministry of Internal Affairs and Communications in 2016 that said it was still surveying all companies with 500 or more employees.

The survey collects data on salaries, working hours and other employment statuses, and is one of the "fundamental statistics" stipulated by the Statistics Act. The data is used to calculate unemployment benefits and work-related injury payouts.

Officials who belonged to the labor ministry section in charge of the issue in 2003 are expected to be interviewed by the ministry's internal inspection team with outside lawyers to get to the bottom of the scandal. The team will have its first meeting on Jan. 17.

According to the people close to the ministry, the manual was found at the Employment, Wage and Labor Welfare Statistics Office in an internal probe conducted after the wage data problem came to light. The document was listed as being created in 2003, and explained that a sampling survey, not an all-out one, of employers with 500 workers or more concentrated in the capital, was sufficient to ensure the precision of the statistical analysis required.

This explanation was removed in 2015, according to other people associated with the issue.

Meanwhile, internal affairs ministry documents show that the labor ministry explained in papers submitted on Oct. 27, 2016, that all companies hiring 500 people or more were being surveyed. In addition, a senior labor ministry official at the level of division chief told a meeting of the internal affairs ministry's Statistics Commission around that time that a complete survey would continue to be carried out.

These findings mean that the labor ministry made false statements about the monthly labor survey on official occasions, even though it began using the sampling survey method in 2004.

(Japanese original by Shunsuke Kamiashi, City News Department)

Also in The Mainichi

The Mainichi on social media