Saturday, September 21, 2013

For COEs, best indicator of value is value itself, not proxies

From Tan Si An -

20 September

I refer to the Land Transport Authority’s (LTA) letter “Engine capacity and power better proxy” (Sept 17).

The LTA stated that the combination of engine capacity and power would be a better proxy for open market value (OMV) than engine capacity alone.

This is puzzling. How can a proxy for OMV be more correct than the actual OMV itself?

The LTA stated that using OMV directly is problematic because of fluctuations, and that even averaging (the value) would not solve this problem.
Under the new Certificate of Entitlement (COE) system, a Proton Exora would be in Category B, the same for a Lamborghini Aventador.

How does the LTA justify grouping two models with an OMV difference of about 3,000 per cent in the same value category, while arguing that a 40 per cent change in OMV over seven months for the same model renders the OMV a less accurate indicator of its true value?

Given the rationale that a market-based approach is the most appropriate way to allocate COEs — a limited and non-basic resource — it is puzzling why the LTA does not accept the information supplied by the market on OMVs.

Why choose a market-based approach then?

COE premiums themselves are susceptible to wild fluctuations from month to month. Does such volatility render these premiums invalid?

The COE system introduces a lot of distortion into the final price of a car. COE premiums can range from more than 10 times the OMV of low-end models to less than 10 per cent for a high-end vehicle.

This distorts market prices and can be viewed as implicit subsidy for the rich at the expense of mass-market buyers, which is undesirable from a social-equity point of view.

Removing all categorisations and simply using a percentage of OMV as the object of bidding would eliminate such distortions and inequity.

If one were to argue that a low-end model contributes to the same amount of congestion as a high-end one, then, again, COE categorisations are arbitrary.

Car ownership itself is not a source of externality; it is only a proxy for road usage, which is the source of the externality of road congestion.

Tackling car ownership does not address the problem directly.

[I have some bias here. There are flaws in the argument, but because I generally agree with the main points, I am refraining from critiquing the flaws. But the flaws are minor and does not detract from the main point which is that OMV is the best measure of OMV, and not a contrived, convoluted, complicated multi-component proxy.]

No comments: