Recently I got a request from a reporter to comment on the labor force participation rates (LFPR) of Singapore. An interesting table is one that compares 2000 statistics with 2005 statistics (click this link to see the table, published in the Singapore Yearbook of Manpower Statistics 2006). It shows that LFPR of those older than 55 is higher in 2005 than in 2000:

  • LFPR of men 55-59 years old : 74.4 (2000) -> 76.4 (2005)
  • LFPR of men 60-64 years old : 49.6 (2000) -> 51.3 (2005)
  • LFPR of men 65 and older: 18.5 (2000) -> 18.9 (2005)
  • LFPR of women 55-59 years old : 29.6 (2000) -> 38.0 (2005)
  • LFPR of women 60-64 years old : 15.3 (2000) -> 19.6 (2005)
  • LFPR of women 65 and older : 4.1 (2000) -> 5.0 (2005)

This upward change was interesting especially among men because LFPR of  men younger than 55 appeard to have dropped by about 1 percetange point or more. The Straits Times ran the story about this on July 15th, 2006 (Saturday), titled "More aged 55 and up are still working" (Link, but you may need to subscribe to the site), which included an edited--to a great extent, I must add--comment of mine.

Now, let me take a look at this apparently "large increase" from a broader perspective (I didn't have time or will to do this when I responded to the reporter's request, which I regret now). When I look at the in-between statitistcs, I realize that the "large increase" is actually a ruse. LFPRs of older men in years 2002-2004 (Go to the Table) were actually much higher than that of 2005. For example, LFPR of 55-59 year-old men was 79.3 in 2002, 78.8 in 2003, and 78.0 in 2004. LFPR of 60-64 year-old men was 53.5 in 2002, 49.8 in 2003, and 53.0 in 2004. Now it looks like that LFPR of older men are actually decreasing. It doesn't seem that "More aged 55 and up are still working", does it? (For my defense, I should add that I responded to the reporter that I expected the LFPR of older men to decrease, not to increase, in the future, although I erred and tried to find a reason for the "increase" which was not real.)

[Caveat: LFPRs of 2002-2004 are computed using the Labour Force Survey (LFS) results, while LFPRs of 2005 are computed using the General Household Survey results. They are not, therefore, strictly comparable. However, unless the LFS sampling scheme is gravely wrong, the LFS should give us a fairly accurate picture of what is going on in Singapore's labor force.]

This all shows the danger of trying to read too much into short-term fluctuations in statistics (responding to a short notice from a reporter). Things are bound to change from day to day, month to month, and even year to year, without any significant "economic" reason. If you can explain every change you see in monthly and yearly statistics, you should be called a "seer" not an "economist."