Discontinuity in the unemployment rate series

By virtue of NSCB Resolution No. 15, series of 2004, a new definition of unemployment was adopted by the Philippine statistical system. Prior to this change, a person was considered unemployed if he basically satisfied 2 criteria:

1. Without work; and
2. Seeking work.

But starting the April 2005 LFS round, a third criterion was included so that the unemployed were those:

1. Without work;
2. Seeking work; and
3. Currently available for work (within the basic reference period or 2 weeks after the LFS interview date).

The inclusion of this availability criterion is in view of the desire to align the Philippine definition to international standards. As a result, this new definition removes the unemployment tag on those who are encountering temporary setbacks to availability due to, say, temporary illness, family responsibilities, or waiting for the school term to finish. They do not remain unavailable for too long, and soon enough they are back in the labor force. 

One problem arising from this is a discontinuity in the time series of unemployment starting April 2005. Specifically, there has been a reduction of the unemployment rate by as much as 2 percentage points, attributed to the definition change (Fig. 1). Some say that this time series disruption has prevented meaningful comparison of past and present unemployment.

Fig. 1. Source: PIDS database. Annual figures are the average of quarterly LFS rounds. Starting 2007, 2000-based population projections were used.

Albert [2014], however, shows a close correlation between the old and new time series (Fig. 2). 

Fig. 2. Source: Albert [2014].