Tyranny of numbers

I have my doubts about the rate of unemployment — 10.3% — recently published by the Statistical Center of Iran (SCI) for fall 2013 (Iranian year 1392), so in a piece that I just published in Lobelog.com I opted to report a rate of 14% that I estimated myself from the SCI report.  The difference between the published number and mine is, as in my previous post on unemployment, all in counting the reduction in the number of people in the labor force as discouraged workers and therefore unemployed in common parlance.

There is a wide margin of error in reporting of unemployment in Iran because it is easy to misclassify unemployed people as “out of the labor force”.  This is because searching for jobs in Iran does not have as clear a meaning as it does, say, in the US, where most of the unemployed collect unemployment insurance for which they have to be searching for work is some specific and well defined ways. The US Bureau of Labor Statistics (BLS) also does a much better job in finding out about those who are “marginally attached to the labor force” but are not actively searching and discouraged workers (who are a subset of the marginally attached).

Significantly, the data for fall 2013 shows a drop of nearly one percentage point in labor force participation relative to the same period a year ago, 36.7% compared to 37.5%.  If we assume that the labor force (10+ population) grew by 1.6% during 2012-2013 (the same as the 2006-2011 rate), the potential labor force will be 24.4 million workers, which is 1.1 million fewer than the number actually reported by SCI — 23.2 million.  I consider these 1.1 million as discouraged workers and add them to the 2.4 million unemployed to get a new number for the unemployed at 3.5 million.  This is about the same number that the census of 2011 counted as unemployed.  Since I do not see any reason why more people should be working in Iran’s depressed economy, especially after a 5.8% decline in the GDP, I would take this number as a lower bound estimate.  This yields a lower bound estimate for the rate of unemployment of 14. 3%, or 4 full percentage points above the published figure.

This number makes much more sense to me especially when I put it next to the 11.2% rate of unemployment that SCI reported for fall 2012: Roughly speaking, a decline of 5.8% in the GDP raises unemployment by 3.1 percentage points. The higher unemployment rate is also consistent with the contractionary policies that have been slowing inflation since last April.