When the UAE’s fuel subsidy policy was modified, the petrol pump price went up. Then it went down, down again, and again. This of course means lower costs for households and businesses.

Problem is, this will be an issue later on. Not in the short-term, but whenever oil prices start increasing again. I am assuming here, and based on calculations I explained in a previous op-ed, that the fuel subsidy was not eliminated but rather reduced.

I am now more inclined towards that and here is why. As oil prices drop, the price we pay for a litre of petrol drops too. And this would probably be the case for another two to three years at least. For one, Iran would start increasing oil production soon enough. Secondly, Libya would be able to increase production once the country gets back into a state re-building mode. What happens after that?

In the beginning, increasing oil prices would not increase pump price of petrol in an alarming way. However, when it actually does — and it will, even if that takes a decade to reach previous peaks — the pump price of petrol per litre would exceed all previous ones since the UAE union was established in 1971.

This is why I have argued earlier that fuel subsidies could not have been fully removed as such a buffer would be required to hedge against higher petrol pump prices once oil prices move upwards. In other words, assume the subsidy rate has been reduced from its level in August (after the reform took place) to zero today because oil prices are too low.

Yet, and when prices increase, the subsidy rate could be increased again at a cost to the government up to the same rate of August 2015.

The question to ask is: What happens when oil prices increase far beyond their level in August 2015?

Step 1: Agree on an average petrol price... I will call it P here, which is fair enough in terms of market prices. Historic prices and trends are sufficient to produce such a petrol band of prices, whether based on five, 10, 15, 20 or even 25 years of historic data.

Such trends would cover troughs as well as peaks and would be accurate enough to come up with that average price.

Step 2: Introduce a petrol tax to inflate any price below P to P’s level. So lower the oil prices are, carbon emissions are kept at a desirable level and the government makes more money.

Step 3: Set a reasonable subsidy rate. This rate needs to cater to the government’s own forecasts and how much it is willing to spend to partially support prices. So for instance, and if petrol pump price is P+$1, the government’s subsidy rate kicks in at 1 per cent and increases whenever the petrol pump price gains by increments previously forecast and matches specific subsidy rates.

The current scenario cannot be sustainable. If oil prices went back to anywhere close to their level in 2012, filling a fuel tank would cost three to four times what it costs today.

That will then be way harder to deal with rather than having a calculated petrol pump price that factors in historic trends and prices, the government’s carbon emissions target, the tax revenues target and the amount of money that the government is willing to spend to support fuel prices when oil prices exceed a certain level.

Once such a system is established, it would be a matter of updating the petrol pump price, P, by incorporating each year’s data into the band of the number of years agreed upon — five, 10, 20, 25 years or so.

This will also ensure a stable petrol pump price all year round, enabling better forecasting and planning by households and businesses.

The last thought that I want to leave you with is: What should the maximum subsidy rate be?

The writer is a UAE-based economist. You can follow him on Twitter at @aj_alshaali.