Frustrating numbers: household income

As you can see from the graph above, the richest households saw the greatest improvement in their incomes in 2011. The data refer to “resident households”, i.e. households headed by Singapore citizens or permanent residents, and was provided by the Department of Statistics.

However, the Straits Times story of 15 February 2012 preferred to focus on the fact that everybody “earned more last year”.

Before going on, I need to give you the numerical data behind the above graph:

There are three important facets of the above numbers I need to highlight: Firstly, the monthly incomes from work include CPF contributions made by employers. Since the total mandatory CPF contributions (employer + employee) make up about 30% of “income” (lower for employees over 50), what it means is that the average household in the first decile, earning a gross $1,581 per month in 2011, actually had only about $1,100 in available money. As I will show below, that household had 3.75 persons to feed in 2011.

The second thing to note is that the decile ranking is not based on household income but on household income per member.

Thirdly, households with no employment income are not included at all in this analysis. Page 5 of the booklet issued by the Department of Statistics (Key Household Characteristics and Household Income Trends, 2011) indicates that 9.3 percent of households have no employment income. This figure includes the 5.8 percent classified as “retiree households”, defined as those comprising solely of non-working persons aged 60 years and older.

Unlike in the case of the previous two articles on Yawning Bread, where I was of the view that those numbers were hard to interpret, being selective and piecemeal, in this case the numbers are quite clear. The frustration they produce however, is simply that nothing is changing. The income gap continues to widen.

But that is an observation buried fairly deep inside the Straits Times’ version. The newspaper’s report begins by focussing on the median:

The median income of a Singapore family has gone up, and it rose faster than inflation last year, according to a government report.

This income, which is the mid-point in a range, jumped 11 per cent to $7,040 a month last year as more family members went out to work in a growing economy overflowing with jobs.

As inflation was around 5 per cent, the family enjoyed a real income growth of 5.6 per cent, said the report of Key Household Characteristics and Household Income Trends in 2011, released yesterday by the Department of Statistics.


Economists like Mr Irvin Seah of DBS Bank cheered the achievement, saying: ‘The fact that real income has risen across the board suggests all Singaporeans have benefited from the healthy economic growth in recent years.’

— Straits Times, 15 Feb 2012, Singapore households earned more last year, by Cai Haoxiang [emphasis added by Yawning Bread]

It’s only when you get to the lower half of the story that you sense the reporter trying to tell us it isn’t all sweetness and light.

But Prof [Randolph] Tan is concerned about the widening income gap between the top and bottom 10 per cent.

The Gini coefficient, a measure of income inequality, went up to 0.473 last year, indicating the gap has widened. It was 0.472 in 2010.

— ibid.

Then the story took an interesting turn, which perked me up. It reported that even though the household income increased more for rich households, in these well-to-do households, the income per member increased by less. That’s quite a trick, I said to myself, but I quickly guessed that it was due to changes in household size, a point confirmed by the Department of Statistics.

Besides overall household income, the report also gave the average income per household member.

It showed that wages of those at the bottom increased at a faster rate than those at the top. The reason, said a Department of Statistics spokesman, is the change in the family size, albeit slightly.

The family size of the bottom 10 per cent shrank last year, while that of the top 10 per cent grew.

— ibid.

Although it was explained in the story, I had to check the data for myself. Indeed, found that the richest decile of households increased size by about 5% while the poorest decile decreased in size by 5%. Click thumbnail for data.

Why is that happening? The figures won’t be able to tell us. One possibility is that the better-off are able to have more children, and those in the lower-income brackets are not. This would be consistent with the observation made by others that many Singaporeans just cannot afford to have children based on what income they have.  At a broad social level, even as we respect individual choice, this cannot but be a bad sign.

The question that seizes us is this: At what point does personal and family stress spill over into social collapse?

This essay is the last of three on ‘frustrating numbers’.  The others were on the number of scholarships for foreign students and foreign immigration.

12 Responses to “Frustrating numbers: household income”

  1. 1 Chanel 24 February 2012 at 17:30


    You have to do a 4th one in this series…….that of “resident” gamblers at the 2 casinos. The numbers given by MCYS acting minister Chan Chan Sing are far more frustrating!

    • 2 Descended 4 March 2012 at 15:50

      IT IS SAID – you can make numbers tell you anything you want. My takeis that the richer are getting richer and the poor are getting poorer,

  2. 3 Anonymous 24 February 2012 at 18:33

    Eugenics at play? Something similar to the graduate mum’s scheme?

  3. 4 Godwin 24 February 2012 at 20:34

    “Eugenics at play?”

    No, just common sense and being responsible to your children.

  4. 5 XH 24 February 2012 at 22:07

    Hey Alex, I think a fall in birth rate among the lower income (and an increase in birth rate among the higher income) cannot affect household size fast enough to have a significant effect on year-to-year household size data. What is more likely going on is more larger households moved into the higher income brackets while more smaller households moved into the lower income brackets – which makes some sense if larger households have on average more members able to work.

    • 6 yawningbread 24 February 2012 at 22:53

      But the deciles split is based on household income from work per member (not total household income from work). In any case, what would make a household “larger”? More children seems the most likely reason. In the end, as I said, these numbers will not get us to the answer; we can only speculate.

      • 7 XH 24 February 2012 at 23:10

        The way larger households could move into higher income brackets even when the metric is household income per member is that those with more household members may have more “spare capacity”, i.e. non-working members who can take up (perhaps part-time) employment when the economy improves. So the average income per member would increase in those larger households by more than in smaller households who are at full capacity.

        But you’re right that we won’t really know for sure without more data (partcularly longitudinal data)! I just thought this seemed like the more likely explanation, since an increase in birth rates takes an awfully long time to translate into more births – an increase in a TFR of 0.1 just means the average woman has 0.1 more babies over her reproductive years, which is a long period of time, much longer than a year. Seems unlikely to translate into a 5% increase in household size over a year. And it also can’t explain a reduction in household size among lower-income households.

  5. 8 Anonymous 25 February 2012 at 06:08

    I think what is missing from this is a time-series analysis over an economic cycle or generation, and it needs to trace the household from its classification in the first year of the dataset. The reason is that generally one would expect the higher-income to be taking more risks with respect to their income, therefore in an upturn they are expected to gain more. On its own, greater income inequality isn’t a problem, but when combined with socio-economic immobility and a barely subsistence level of income at the lowest deciles, it makes the continued disregard by policy-makers and the positive spins put on by the media utterly reprehensible.

  6. 9 Lye Khuen Way 26 February 2012 at 11:38

    May I venture a guess as to why, some household may have more members, beside more babies ?
    How about more grand parents/ aunts-uncles / sibilings through the immigration route ! These statistics were for BOTH true blue Singaporeans & PR’s right ?

  7. 10 Yong Siang Woodlands 29 February 2012 at 16:17

    Hi Alex,

    referring to “Thirdly, households with no employment income are not included at all in this analysis. Page 5 of the booklet issued by the Department of Statistics (Key Household Characteristics and Household Income Trends, 2011) indicates that 9.3 percent of households have no employment income. This figure includes the 5.8 percent classified as “retiree households”, defined as those comprising solely of non-working persons aged 60 years and older.”

    theres 3.5% of households below 60 without income. it would be interesting for Dept of Stats to reveal their profiles.structurally unemployed, imprisoned or what?

    thank for

    it shows only those 7th decile and above are increasing family size. stats here can help the population planning dept. 🙂

  8. 11 Anonymous 2 March 2012 at 22:38

    The way i see it, households were forced to combine due to ppl selling their flats and being priced out of the market, hence increasing the number of members per household, and increasing household income. another reason is because ppl are getting married later or not getting married, so parents work, children work, household income goes up. That’s why some ppl prefer to look at household per capita income.

  9. 12 Rajiv Chaudhry 4 March 2012 at 01:22

    Fascinating analysis. I have long been seeking a clear exposition of this subject and have not been able to obtain it elsewhere. Your figures show a very clear picture, although as Anonymous says above, it would have been interesting to see the data over a 10 or 20 year period to see how incomes have moved over time.

    Although you say in this case “the numbers are quite clear”, would it not have been much more straightforward to present the figures to for all working individuals, rather than households, which creates all sorts of grey areas? Its probably politically more expedient but, in my view, is still obfuscation of a sort.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: