Within a space of slightly over two hours Friday night, two friends of mine mentioned Tharman Shanmugaratnam’s ‘pay one dollar, get four dollars’ sales pitch. “What do you think of that?” asked the latter of the two.
I knew Singapore had become one of the gambling capitals of Asia, but it was still depressing to see our politics adopt the same mindset: If you pay taxes, you win!
Clearly, that was not the point the finance minister was trying to make. I believe he was trying to show that the help lower- and middle-income Singaporeans were getting from the government exceeded what they paid in taxes, over a lifetime. But from the twinkle in my friends’ eyes as they repeated Tharman’s soundbite to me, I had the awful feeling it was received quite the wrong way. My eyebrows rose.
I had seen that comment from Tharman in the morning’s edition of the newspaper, but the math didn’t strike me as noteworthy — of course, the low income would get more back from the state than they put in, I told myself — and I did not pay it any further attention until my two friends mentioned it.
For every dollar that low-income Singaporeans paid in taxes in their lifetimes, they now get back $4 in the form of government aid.
Middle-income earners have also gained, the Finance Minister said, as he wrapped up Parliament’s three-day debate on the Budget. They receive about $1.50 in benefits for every tax dollar they pay. If they own a car, that falls to around $0.80.
— Straits Times, 2 Mar 2012, Grow economy to forge inclusive society: DPM; Strategies to tackle inequality are in place, Tharman assures MPs, by Lydia Lim
(The graphic alongside this story in the Straits Times, indicated the following ratios to two places of decimal: Low-income 4.25; middle-income without car 1.55; middle-income with car 0.81)
Straight away, a critical-minded person would take issue with the vague terms he used. What do ‘low-income’ and ‘middle-income’ mean? What does the term ‘benefits’ include?
You should also note the qualifier ‘over a lifetime’. This means the multiples the minister mentioned do not apply when we only consider a person’s working years. The payback only reaches these multiples when the amounts spent by the government during a person’s childhood, adolescence and senior years are included. Thus my guess that ‘benefits’ includes education and healthcare spending, in addition to to the more tangible help from housing grants, rebates, workfare payments, baby bonus, etc.
Anyone who wants to check those numbers would either have to wait till Tharman (right) clarifies, or make some heroic assumptions.
Even so, I hope to show readers why I didn’t think those numbers were anything out of the ordinary. To do so requires me to make a few assumptions — not too many; I don’t want to play hero more than necessary.
Let’s start with a bird’s eye view. Individuals who don’t own cars pay, in the main, two kinds of taxes: personal income tax and the Goods and Services Tax (GST), which is currently rated at 7 percent. As you can see from the pie chart I found from the Finance Ministry’s website (above), each of these two types of taxes contribute about 15 percent to the total budget revenue. (Personal income tax is the pale lime slice; GST is the strawberry cream slice). Together, they make up about 30 percent. In other words, the government has more than than three times as much money as they collect from individuals for the various kinds of governmental spending.
Some of that spending cannot be considered ‘benefits’ except by a great stretch, e.g. defence or certain infrastructure projects. But still, there’s more money available for ‘benefits’ than the 30 percent collected from personal income taxes and GST. Hence, to have an overall Benefits-to-Personal Taxes ratio of 1.5 or more is therefore not remark-worthy at all.
Looking at the same question from a micro perspective is harder, because it requires several assumptions.
Let’s assume that by ‘low-income’, we mean someone whose income from work is at the 10th percentile. By ‘middle-income’, it is someone at the 50th percentile, i.e. the median income from work. The Ministry of Manpower has on its website the 2011 median gross monthly income from work of full-time employed residents, excluding employer’s CPF contribution, of $2,925.
Getting the figure for the 10th percentile proved much harder. The closest I could find were these figures from Table 32 (page T47) of the Manpower Ministry’s Labour Force Report 2011. It gives the number of employed residents in each wage band:
(By the way, don’t you just love the title the Report gave to the table: “. . . monthly income from work and sex. . .”?)
The table indicates that the 10th percentile income earner would have a salary around $1,200 a month.
Next is an outline computation of the personal income tax and GST that these two persons would have to pay:
I assumed the typical person would get a 13th month bonus and I also put in the main tax reliefs that a male citizen might enjoy. The personal income tax is then computed based on the current rates.
For GST, I assume the low-income guy would spend just about all the disposable income he has, perhaps managing to save only his 13th month bonus. As for the expenditure likely incurred by the middle-income guy, I made an estimate by referring to the Report on the Household Expenditure Survey 2007/2008. Page iv has a table that shows average monthly expenditure by households in the third quintile (41st to 60th percentile) to be $3,571. Page 2 has a table that shows average household expenditure per household member, for households in the third quintile to be $974. Together they suggest a household size of nearly four persons.
I’ll assume that in such a middle-income household, there are roughly two income earners, and that each income earner supports two persons. Thus, the expenditure per income earner is about twice $974, or $1,948.
The second block of the green tables above shows how much in GST the low-income and middle-income workers would have handed over to the government based on their assumed expenditure. The bottom line is that the low-income worker would be paying a total of $756 a year in GST, while the middle-income worker would be paying a total of $1,652 a year in income tax and GST.
The table at right shows my estimate of how much the ‘low-income’ and ‘middle-income’ guys would be paying in taxes over their working lives, and how much ‘benefits’ they get in return per year on average over a lifespan of 75 years.
Considering that Workfare for the low-income, GST rebates, baby bonus already run into hundreds and thousands of dollars per head, and housing grants into the tens of thousands, the bottom figures (averaged per year over a lifetime) aren’t that much. They certainly look less than any initial impression one might get from Tharman’s multiples of 4.0 and 1.5, which to some sounded like “pay one dollar, win four dollars.”
As a soundbite, it was a good one. But as with all soundbites, it is worthwhile looking beyond it.