Good question. How much more is reasonable? That question has been dealt with on several occasions in the past by people much more familiar with the situation than the nameless PFS [Person Familiar with Situation] referenced above.

On each of these occasions these people, professional arbitrators who did not hide in anonymity as PFS does, ruled against a government decision not to grant toll increases for the Eastern Harbour Crossing and made their rulings stick.

What is more, the government's arguments are exactly the same as those rejected by arbitrators in 2005, with the exception only that the government argued against toll increases last time on the grounds of economic slowdown. This time the economy is booming. If the situation is changed it would thus seem to have changed in favour of increases.

I therefore confidently predict that the government will again lose at arbitration and, more than this, that the bureaucrats know they will lose but inflict the cost of arbitration on the public purse again to preserve appearances.

The fundamental question is what the New Hong Kong Tunnel Company should be allowed as an internal rate of return (IRR) on a 30-year contract to build and operate the tunnel before transferring it back to the government in 2016.

Internal rate of return is a financial measure that expresses the net present value of all cash flows as an annualised discount rate. No, I can't do better than this. I would need a blackboard on which I could set out numbers and examples until I saw that 'Oh, I get it' look on your face. Just take it as read that IRR is the key measure of whether a capital project is worthwhile.

You can also take it as read that '...they have made billions of dollars, how much more is reasonable?' is not a measure that financiers use. Only political weasels who are bereft of experience in project finance, invoke an argument like this.

The point here is that in 1997 an arbitrator who was asked to set a figure for 'reasonable but not excessive remuneration' came up with an IRR of between 15 per cent and 17 per cent as appropriate for the company. This figure was accepted and confirmed in further arbitration in 2005.

Even the government concedes that returns are below this level but it now says that these arbitration decisions 'are not binding on subsequent cases. Each case has to be considered on its merit'.

It made the same argument in 2005 and was soundly rebuffed. As the arbitrator pointed out at that time, the tolls required to achieve the target IRR may change but the IRR is for the life of the project and even the government representative in 1997 asked the arbitrator to come up with such an IRR.

Now, I am not sure myself that 15 per cent to 17 per cent is the right figure. I have always thought it a touch on the high side. But that is what can happen when you do these build, operate and transfer projects. You need a high return to compensate you for losing the asset at the end of the period and, if it goes to arbitration, both parties have to accept the arbitrator's decision or we can give up on doing any more such BOT projects in Hong Kong.

It is cheap populist tactics then for a government spokesman (anonymous once again) to try pulling the public heartstrings by saying that the company should have 'due regard to public affordability and acceptability in devising their toll strategy'.

This is something that the bureaucrats should rather have borne in mind when conceiving the project and asking for private sector bids on it. Private entities cannot just take a loss on such a project and make it up elsewhere by pushing taxes higher, as governments can.

'Reasonable' is a precise return-on-investment figure to be determined by arbitration if the parties cannot agree. What we have here is bureaucrats too scared of a few noisy critics to stand up for what they earlier agreed.