I don’t know how Leong Sze Hian does it. Does he sit in front of a computer, staring at the Ministry of Manpower website, waiting for updates? The same day that this ministry released its Singapore Workforce 2010 report, he knew about it and even managed to write a piece on it for the Online Citizen. See Over 30% earn less than $1,200?
I am grateful for him drawing attention to the ministry’s report, for indeed it makes interesting reading, to say the least.
In his article, Leong’s point is that there is a large number — over 30 percent, he suggests — of people falling into one of these categories:
(a) those earning less than S$1,200 per month, based on 2001 dollars;
(b) those unemployed and looking for work;
(c) those unemployed and too discouraged to even look for work.
I don’t intend to cover the same ground, but to take the topic a little further. And it is this: The ministry’s data suggests that income inequality is rising again, though that conclusion cannot be reached by relying on the Workforce 2010 report alone, but from cross-referencing the ministry’s Earnings and Wages data with the Workforce report.
* * * * *
First however, I would like to present a few key facts from the Workforce 2010 report, which is based on a survey conducted in the middle of this year of citizens and Permanent Residents. In other words, the data discussed here excludes Work Permit and Employment Pass holders.
21 percent earn below S$1,200 per month.
The ministry’s report contains more details distinguishing full-time from part-time, but as Leong also pointed out, with “part-time” defined as working as much as 35 hours a week, it is pretty meaningless applying this distinction. Here, I too am largely ignoring it.
The actual number of people earning under $1,200 in the middle of this year was 400,100. Two thirds of them (262,700) work full-time (i.e. more than 35 hours a week) to earn what they earn, i.e. under $1,200.
* * * * *
Paragraph 3.6 of the Workforce 2010 report was what got me particularly interested.
First, it reported that the median income for all employed persons (i.e. full-time and part-time citizens and PRs) rose 3.3 percent in gross terms 2010 over 2009. Adjusted for inflation, it represented a rise of 1.0 percent.
However, 2009’s median income for all employed persons was 1.2 percent lower than 2008; 1.8 percent lower after adjusting for inflation. This means the median income has not yet recovered to the level of 2008 in terms of purchasing power.
I think Today newspaper painted the situation rather rosily when it wrote: “CIMB-GK regional economist Song Seng Wun noted that real wage growth was ‘just barely keeping pace with inflation’.” (Today, 1 Dec 2010, Median monthly income goes up)
* * * * *
I then searched for mean incomes and found it in the Manpower ministry’s Earnings and Wages data, hyperlinked above. Using only the Second Quarter mean income (the closest equivalent to the snapshot survey figures used for the mid-2010 Workforce report, I see that 2010 Q2 mean income increased by 5.8 percent over 2009 Q2 mean income. In dollar terms, before adjusting for inflation, it was S$3,819 in 2010 and S$3,609 in 2009. As you can see, it is a bigger increase than that for median income.
That said, one has to be careful because the data are not drawn from identical samples. The median income was based on a mid 2010 survey while the mean income is drawn from Central Provident Fund data for all employed persons (part-time and full-time) but leaving out self-employed persons.
Nonetheless, we can still compare trends over the last decade or more. Here is the data, comparing Second Quarter Mean Incomes (derived from CPF) with mid-year Median incomes (derived from surveys):
Presented graphically, you can see the mean income rising at a steeper slope than median income. What that means in terms of income inequality is discussed below.
The graph also incorporates the Consumer Price Index for the same period. It was 84.9 in 1996, rising to 100.0 in 2009. It indicates that the median income earner has 10 – 20 percent more purchasing power today than he had in the late 1990s. The mean income earner has improved his purchasing power by about one third.
* * * * *
At this point, some readers may need me to explain the significance of “median” and “mean”. Not everybody took classes in statistics.
“Median” I have explained above. It is the figure that is the halfway mark: half the population would be below the figure and half above it.
“Mean” is what the layman calls the “average”. In the case of incomes, it is naturally weighted by the number of persons within the same income band.
Because the distribution of incomes in virtually all societies is skewed, with lots of low-income earners and few higher-income people, there’s always a gap between median and mean. Generally speaking, the more skewed the income distribution, the larger the gap.
Take, for example, countries A and B, each with exactly 1,000 employed persons. They both use the same currency — the Fictional Franc. They have people earning as low as 500 francs a month and at the other end, some (but far fewer) who earn much more.
Both countries have medians of 4,000 francs. Exactly 500 persons in each country earn less than this threshold, exactly 500 earn more than this threshold.
But as you will have noticed from the graph, Country B has a longer “tail” of high-income earners. These high-income earners pull up the mean or average. Thus, while the mean for Country A is 5,051 francs, that for Country B is 6,058 francs. You can do the calculation to see for yourself.
Basically, it is this: If the gap between median and mean is increasing, it suggests a widening income spread, though it is not as hard and fast a rule as that — there are other possible, if unlikely, explanations.
Now, look back at Singapore’s figures for 2009 and 2010. Last year, the mean income (S$3,609) was 49.1 percent higher than median income (S$2,420). This year — 2010 — the mean income (S$3,819) is 52.8 percent higher than median income ($2,500). The gap is increasing. Again.
* * * * *
Another clue can be seen from this nugget I found in yesterday’s edition of Today newspaper:
Evidence of [reaping the benefits of reshaping the economy] can be found from the Inland Revenue Authority of Singapore’s annual reports. On a year-on-year basis, the number of residents with assessed income of $100,001 and above has been increasing. There were 132,399, 153,779, 187,856 and 215,467 residents with taxable income of $100,001 and more in the Years of Assessment 2006, 2007, 2008 and 2009, respectively. Note also that the number did not fall in the recessionary period from 2008 to last year.
— Today, 3 Dec 2010, Reaping the benefits of wealth creation
We discussed above how median income was rising only marginally, yet the the number of residents with taxable income greater than $100,000 ballooned 62.7 percent from 132,399 to 215,467 over four years, two of which saw a recession. Does that not suggest a widening income gap too?
A far better measure of income inequality would be to derive a Gini Co-efficient from individual earned incomes. As far as I know, this information is not publicly available. (The Gini Co-efficient occasionally announced by our Statistics Department relates to household income.) I think it is necessary for our Statistics Department and/or Ministry of Manpower to publish more details regarding earned income by income bands, together with a relevant Gini Co-efficient. This especially as circumstantial data highlighted above suggests a reason to be concerned.