The system, officially called the Vehicle Quota System, was put in place in May 1990 as the Government deemed the prevailing vehicle growth rate of 3.5 per cent unsustainable in the long term.
The system was to cap growth at 3 per cent per annum, to be in line with the pace of road network expansion and me-dian income growth of Singaporeans.
An analysis of growth trends 20 years before the system was introduced in 1990, and 20 years after it, shows the car population has, on average, been growing faster post-COE than pre-COE.
According to data from the Registry of Vehicles and the Land Transport Authority, the average growth rate was 3.5 per cent per annum from 1971 to 1990. It rose to 4 per cent from 1991 to 2010. The growth rate after COEs were introduced was 14 per cent faster than before.
Why? Here's one theory: Before the quota system, consumer behaviour was influenced more acutely by factors such as economic slowdowns, oil crises and changes in tariffs. But with a quota system in place, consumers become shielded from natural market forces. While the quota system put in place a growth ceiling, it also established a 'floor' that prevents the market from collapsing - even in the depths of recession.
An example of the latter was well demonstrated in 2004, when Singapore sank into a deep recession. Car sales grew 19.2per cent that year.
The year before, when the country was in the throes of the Sars epidemic, sales grew 29.1per cent.
But what would have happened if there had been no quota system in place? Would Singapore have become as gridlocked as cities like Jakarta, Manila and Bangkok?
Maybe. Then again, maybe not.
While growing income tends to fuel the desire to own cars, there comes a point when the pattern goes into reverse.
A paper written by United States transport researcher Joyce Dargay and economist Dermot Gately (Income's Effect On Car And Vehicle Ownership Worldwide: 1960-2015) asserts that car ownership grows slowly when income levels are low. When a country reaches 'middle-income' status, the ownership growth rate becomes twice as fast. And when income goes beyond a certain level, the car ownership growth rate actually reverses.
The global car market today confirms this. Car sales in developed markets such as America, Europe and Japan are slowing, while sales in China and India are in the fast lane.
Singapore's per capita gross domestic product ($63,000 last year) is among the highest in the world. So, even if its car ownership growth rate was relatively fast in the 1990s (assuming the COE system was not in place), the pattern could well have gone into reverse gear by the mid-noughties.
Today, when COE premiums are heading towards record levels, the quota system causes unintended ill-effects too. High inflation is one. Ironically, public transport fare adjustments are influenced greatly by the inflation rate.
The other major ill-effect is social discontent. Like all auctions, the COE system favours those who are most able to pay. When the supply of certificates is constricted (as it is now), COEs invariably end up mostly with wealthy consumers, who buy bigger and costlier cars.
Interestingly, this contrasts with a common criticism of the high ad valorem taxation regime preceding the quota system. When car registration taxes added up to more than 220 per cent of a vehicle's open-market value (OMV, or roughly its pre-tax cost) before 1990, well-heeled buyers complained that the system placed an inordinately high penalty on bigger, pricier cars.
Car taxes today amount to 120 per cent of OMV.
If there was an attempt to incorporate social equity into the quota system by segregating COEs according to car engine size, it is now negated by an increasing number of premium brands with small-engine models.
In the first quarter of this year, brands such as Mercedes-Benz, Volvo and Audi accounted for 45 per cent of sales in the up-to-1,600cc COE segment. Five years ago, they had a mere 0.3 per cent slice of the segment.
This only serves to widen the gap between the haves and the have-nots.
So, after 20 years, perhaps it is time we took a long hard look at the COE system, to examine its relevance and to see if it can be improved upon.
In doing so, it might be useful to look at how other cities without a quota system have coped - Hong Kong, for instance.
Hong Kong has 59 cars per 1,000 residents - half of Singapore's 117. Taipei has 250 cars per 1,000 residents, but the annual mileage of cars there is half that of those here. Ditto Tokyo.
New York is another example. The Big Apple is one of the wealthiest cities in the US, but its car ownership rate is among the lowest (230 per 1,000 residents).
These cities share a common denominator: a superior rail network and limited parking facilities. In the case of New York's Manhattan, 60 per cent of work trips are made by public transport. In Hong Kong, the figure is 90 per cent.
None of them has a quota system, or even high taxes on cars. They rely instead on letting the motorist bear the brunt of driving: jams and the frustration of not being able to find parking.
That, however, exerts a huge environmental cost. Singapore can do better. We have electronic road-pricing. All we need now is a policymaker brave enough to expand the gantry network and treble or quadruple ERP rates.
After all, it is ultimately congestion we are tackling. ERP, if priced right, will be more efficient at controlling congestion than a vehicle quota system.
And if a quota system is deemed still necessary, we should have one without the sharp supply fluctuations that send premiums to $90,000 in one year, and $5,000 in another.
