Back To Top
Opinion

[Editorial] No window dressing

Change of sample, survey method raises questions about reliability

The top 20 percent of earners in South Korea made 5.41 times as much as the bottom 20 percent in the first quarter, Statistics Korea said last week.

The statistical agency used a new sample and survey method for the first time to calculate the income disparity.

The multiple figure for the first quarter of last year was 5.18 times when calculated by the new methodology and 5.8 times by the old one.

If the old standards were applied, the figure for the first quarter of this year is reportedly estimated to be more than six. This would be the worst ever, breaking the previous record of 5.95 times higher for the first-quarter 2018.

The agency also excluded one-person households from the sample. One-person households account for about 30 percent of the total, and consist mostly of old and young people with relatively low income.

Statistics Korea announced May 7 that average household consumption expenditure fell for the second successive year last year. The agency said household consumption declined because it increased the size of household sample from 1,000 to 7,200. It added that comparison with data gathered from the old sample is difficult.

The agency said on May 13 that the number of the employed decreased 476,000 in April from the previous month -- the biggest drop in 21 years.

The number of temporary and daily laborers decreased, too, by 782,000, the largest decrease since related data was first compiled in 1982.

Unexpectedly, however, the unemployment rate fell 0.2 percentage point in April from a year earlier to 4.2 percent.

Statistics Korea said many of the self-employed who shut down their shops largely due to the impact of the new coronavirus were not counted as unemployed because they were not actively seeking employment. Statistically this may make sense, but one cannot but question if such data reflects the dire employment reality properly.

Controversies over the reliability of data announced by Statistics Korea have dogged the administration, particularly after its commissioner was replaced abruptly in August 2018.

The agency under the new commissioner made major changes to the sample and survey method for household statistics. The government once considered discontinuing the household survey in 2017, but decided to continue it. There were calls in the ruling camp for the need to confirm the effect of the income-led growth policy.

The change in sample and survey method has made time series analysis difficult.

Furthermore, the government is itself to blame for questions about the reliability of its statistical data by interpreting them to its advantage. When employment shrinks, it blames demographic changes or the weather. When employment increases, though this is largely thanks to public sector part-time jobs for senior citizens, the government trumpets it as a product of its income-led growth policy.

The government created many public sector part-time jobs for elderly people and counted them as employed. Then it had to suspend most senior citizen employment programs due to the new coronavirus crisis.

And yet Statistics Korea classified suspended part-time jobs for seniors as employees on furlough who are counted as the employed, and announced that employment increased 492,000 on-year in February. Employment statistics, if obtained this way, would be hard to trust as they are.

The government and the ruling party have seldom admitted faults in their policies. Hong Jang-pyo, chairman of the presidential Special Committee on Income-led Growth, boasted on May 13 that income redistribution situation improved thanks to income-led growth policy. But nine days later, Statistics Korea announced household income statistics that pointed in the opposite direction.

Statistical data on household and employment disprove the effectiveness of the policy.

Changes of sample and survey methods are just “window dressing.” Time series analysis must be made possible. The government needs to conduct surveys with both old and new methods -- at least for the time being.

Statistical data announced by the government must be as objective and impartial as possible. The government must grasp the economic reality as it is and let the people know it. Window dressing statistical data is an act of deception. It distorts the effects of policies, misleading policymakers and the country.
MOST POPULAR