How multiple-car ownership COE surcharge could work

This is not cool

This is not cool, you hear me?

Over 70% of people polled by the Land Transport Authority are in favour of imposing a surcharge on multiple-car ownership. The LTA itself was open to the idea, and specifically invited suggestions for this in the public consultation. It was only the difficulty of implementation that deterred it from doing so. As The Straits Times reported, experts agreed that there is no easy way.

Personally, I belong to the 30% who are not keen on the idea. Not because I own multiple cars or intend to, but because it means imposing more rules to fix a COE system that is very flawed (I think it should be phased out and replaced by very high ERP). Fundamentally, I also feel that how many cars you choose to waste money on is your own business, though it’s also true that a limited COE quota system makes it every other car buyer’s business.

The problem with imposing such a surcharge is how you define multiple ownership — by individual or by household? If by individual, you can circumvent by simply registering the car under another family member’s name; if by household, you unfairly penalise multi-generational households that have more than one car. When people think of multiple-car ownership, they think of a rich person owning several cars, without considering how many others there are in his household. Perhaps they are also against couples owning one car each, without recognising that two persons owning two cars uses the same resource per head as a single person living alone owning one. Also, if a couple has many kids, is it not reasonable that they might need more than one car?

Anyway, if this is the route we want to go down in the name of greater social equity, I decided to have a crack at playing policy wonk and see what I can come up with:

The Proposed System

Car ownership will be measured per buyer’s household by a metric, which I shall call Cars Per Person (CPP) in household:

CPP = 2 x Number of Registered Cars in Household / Number of Persons in Household

(Persons include members of all ages but exclude domestic helpers. Household will be based on officially registered address.)

The reason for including a factor of 2 in the equation is to normalise the baseline case of two persons to one car. In other words, two persons to one car will be taken as the average level of car ownership for a household, with a CPP of 1. This CPP becomes a measure of how “deserving” of a car a household is.

Based on the CPP, a car buyer will then pay a percentage of the COE price according to Table 1 below.

Table 1: CPP Threshold Levels

Table 1: CPP Threshold Levels

These are just sample threshold levels and percentage weights that can easily be fine-tuned. You will also notice that not only are there surcharges for a high CPP, there are discounts for a low CPP as well. Bidding will still be done based on full (100%) COE price before taking into account the one-off surcharge or discount. Value of COE after transaction and COE rebate on deregistration, on the other hand, will be based on actual amount paid.

Table 2 below gives a clearer illustration on how CPP affects different types of households:

Table 2: CPP by household make-up

Table 2: CPP by household make-up

For example, we see that in the base case scenario of a married couple with no child (Ref 2.1), they pay 100% of COE for their first car. Should they buy another car (Ref 2.2), the COE increases to 110% for the second car. In a multi-generational family, say a childless couple living with 2 parents, they pay 80% of COE for the first car (Ref 4.1) and 100% for the second (Ref 4.2). Similarly, for a couple with 2 children, they do not get penalised should they decided to own a second car (Ref 4.2) with COE at 100%.

What are the advantages of such a system?

Socially equitable: It becomes obvious that the amount you pay depends on how big your household is. Essentially, the bigger the household, the more reason to have multiple cars. In most cases when an additional car is added to the household, the level of COE percentage steps up as well.

Pro-family: This system not only does not penalise multi-generational families, as the main concern has been, it provides incentives through COE discounts. This is in line with the government’s objectives in not only encouraging married couples to stay with their aged parents, but also to have more children.

Positive self-selection: This system also favours households with a lower CPP, seen as “more deserving”, in the COE auction. For example, a four-person household bidding for their first car (Ref 4.1) needs only pay 80%. If the maximum they are willing to pay for the COE is $50,000, they will bid $62,500. In contrast, a 2-person household bidding for their third car needs to pay 120%. If the maximum they are willing to pay is also $50,000, they will bid $41,667. Therefore, the family with a lower CPP has a higher chance of winning the bid for the same amount both households are willing to pay.

Flexible: The CPP threshold levels and percentage weights can be adjusted continually to take into account prevailing COE prices, road conditions and even the state of the economy. For example, some may not agree that a single person living alone should be penalised, in which case the base level (Level 4) of 100% COE could be stretched from [0.9, 1.5) in Table 1 to [0.9, 2.5). Granularity can also be changed to have more or less threshold levels. In fact, it could also work on a pure surcharge basis such that any CPP below 1 will still pay 100%, though that will nullify some of the other benefits mentioned in this section.

Comprehensive: Other factors could also be incorporated into such a framework. For example, families with disabled members could be given “CPP upgrades”. So if, say, such a household is in Level 3 (90%), it will be bumped up to Level 2 (80%) — effectively be given a further discount. Likewise, if popular opinion holds that foreigners should pay more, such households could be downgraded by a level and end up paying more for their COE.

Auto-moderation: For households with a lower CPP, the discount they are eligible for helps to reduce the fluctuation in COE prices. This is illustrated in Table 3 below. As this surcharge/discount system is based on a COE multiplier, it also means that in times of high COE prices, the surcharge on multiple ownership increases automatically to discourage those with high CPP from bidding too high or from buying altogether.

Table 3: CPP lowers fluctuation on COE prices

Table 3: Reduced fluctuation in COE prices for households with lower CPP

What are the potential loopholes then?

There is nothing to stop a buyer in a high CPP household from registering the car under a friend or family member who lives in a different address. For example, a childless couple buying their second car could register it under one of their parents living elsewhere without a car. There are ways to discourage this though. For one, the government could make it compulsory for the registered owner to be the default named driver in the car insurance. Any additional named driver thereafter who lives in a different address will incur an annual levy, with exemptions granted on a case-by-case review basis. This will not penalise the majority who share cars while living under the same roof. If one chooses not to be a named driver, he exposes himself to a high excess for claims. It remains to be seen if a large number of drivers will take this risk and this could be a weak link in the system.

Buyers could also get around the rules by changing addresses, or bringing others into the household temporarily. The first loophole can easily be closed: if a buyer has changed addresses in the 6 months prior to buying, the highest CPP among the different addresses will be taken into account. Similarly, any additions to the buyer’s household in the past 6 months (excluding newborns) will not be counted towards its CPP. If it is too difficult for the government to track household movements, buyers will have to make signed declarations when bidding and be subject to random checks with penalties if falsehood is discovered.

There will be other loopholes, as with any system, and other considerations I have not addressed (e.g. what about company cars? Ownership transfers?), but this is a workable starting point to build on. What’s most important is that it serves the purpose for the vast majority of cases and there is sufficient deterrent against circumvention. As one may say, if some choose to go all out to cheat a system, you just have to give it to them. Ultimately, it always boils down to how many would go to such extent to gain that 10-30% of COE.

So, to summarise based on issues raised:

Easy to get around the rules: Not so easy here if you take into account entire household and close loopholes via insurance levies and mandatory declaration on household changes.

Surcharge does not make it easier for others to get a car: This does, through discount system, self-selection auction and “CPP upgrades” for needy families.

Surcharge will affect multi-generational families: Not with the CPP formula and threshold system. It incentivises them.

Surcharge will not lead to more reasonable COE prices: This will, through discount system and auto-moderation effect.

Surcharge will lead to more cars on the road because multiple-car owners don’t drive every car: There is nothing wrong with this. The COE quota system should be on the basis that every car is properly utilised, so any car that sits most of the time in the parking lot is a wasted resource for the whole society.

Tiered ARF and road tax already addresses social equity: This is on the assumption that people who own multiple cars also own more expensive cars, which is probably true most of the time but not always. The CPP system addresses social equity on multiple car ownership directly.