ESTIMATES OF EXPENDITURE OF SINGAPORE FOR THE FINANCIAL YEAR - 1ST APRIL, 1998 TO 31ST MARCH, 1999
Sir, in the Committee of Supply in July last year, I announced that my Ministry would undertake a comprehensive review of the vehicle tax structure in conjunction with the implementation of ERP. We had two objectives in mind. One, was to achieve a better ownership between ownership and usage in managing our traffic, in better controlling traffic congestion; and, two, was to have a fairer, a more rational and a more consistent basis for taxing vehicles on the road. The new vehicle tax structure announced last week on 4th March should not be seen in isolation. It has to be taken as a total package, in other words, taken together with the ERP implementation plan as well as the package of road tax rebates. Just to give a very quick summary and the bottom line of the new tax package, a majority or 70% of vehicle owners, especially companies and businesses, will benefit from savings in road tax. The remaining 30% will see an increase in their road tax mainly because they have been paying a disproportionately low road tax relative to other vehicles when you consider the amount of road space they use or the amount of wear and tear they cause on our roads. However, to help them adjust, the Government has phased in the road tax increase over three years. We should take it as a total package, ie, the new tax structure, the phasing in and also the road tax rebates which all vehicle owners will enjoy. Many of these vehicle owners, especially motorcyclists, will end up paying lower or no road tax in the next few years. Reactions to the new tax structure have been generally positive and I would like to take this opportunity to express my appreciation to all those who have given their comments and feedback in the consultation exercise that my Ministry conducted last year. I would also like to thank Members for their support of the new vehicle tax structure and the ERP implementation plan. There have been a few dissenting voices on the ERP which I would like to deal with before I address the specific concerns that were raised by Members. Mr J.B. Jeyaretnam, in particular, questioned the principle of charging for road usage. He asked whether it is fair to charge for the use of the roads. For many years, most countries, if not all countries, operated on the principle that all motorists just need to pay a lump sum for the use of the roads and thereafter, the use of roads would be virtually free. So irrespective of how many times they drove or how much congestion or how much pollution they cause, there would be no incremental cost involved in using the roads. This was well and good, until more and more people wanted to also enjoy this free usage, and as more and more people wanted to use these same roads at the same time, the roads became more and more congested. The classic solution for many countries then was to build more roads, ie, not enough roads, roads are being congested, just build some more. If there is not enough land, you just build over, double-decker and triple-decker roads. But many cities are finding out today that their citizens are no longer prepared to have the road builders come along and rip up their country side or to see their cityscape destroyed by all these ugly concrete flyovers all over the place, and having all these cars polluting the country side or the city that they are living in. Why suffer all this in order to allow motorists to come and drive on the roads and pollute the environment? Sir, the reality today is that people are now coming around to the idea that road space is a scarce resource, as scarce a resource like water, oil, or any other commodity which is in short supply. So when Mr J.B. Jeyaretnam said that he does not oppose the idea of unclogging our roads, that it is a good idea, I was very happy to hear that because I think it is the first time that he has accepted that roads are a scarce resource. So now we are getting somewhere in the debate. But he does not favour the idea of charging. He says that if we charge for this road usage, this scarce resource, we are being unfair. He prefers to ration. He prefers to allocate this scarce resource by fiat. He decides, or the Government decides, or the civil servant decides who gets to use the road and who does not - black labels, red labels, no label, I guess, that would be the third category. So who gets the black label? Who gets the red label? It reminds me of some brand of whisky. But who are we to decide who is more deserving of using the road? The housewife who is driving a child to school, the businessman who needs a car to attend an important meeting, the salesman who has to take the car around to drive for appointments? Mr Jeyaretnam says salesmen, black label; doctors, black label; housewife with kids, red label; those who go to work, no label, because you have no right to be driving to work. I must say I sympathise with that point of view. But who are we to decide or to tell people that you cannot drive to work and that if you use your car just to drive to work, you get no label? Is it not better for us to give the choice back to the motorist to let him decide what he wants to do through a system of charging a fair price for the use of the roads? Is it not better for us to reduce the upfront cost of owning a vehicle, as we have done with the road tax rationalisation, to give back to the motorist a certain sum of money, not an inconsiderable sum of money, and then to let the motorist decide for himself whether he wants to give this money back to the LTA? 3.45 pm Many cities are coming around to this harsh reality that, first of all, road space is a scarce resource and, secondly, that the best way of allocating this scarce resource is to charge a certain price for using it, like any other scarce resource, whether it is water, land, oil, etc. In Singapore, we faced up to this harsh reality many years ago when we introduced road pricing in 1975, when we introduced the Area Licensing Scheme. At that time, the CBD was becoming congested. There were only so many roads you could build in the CBD and yet there were more jobs in the city. There were more people wanting to drive into the CBD. There were more offices, more meetings being held, more people wanting to drive their vehicles in. What do we do? We charge $3 if you want to drive in. If you have no business to be in there, do not pay the $3. That is one car less. And the motorist who pays the $3 suddenly realises, "Hey, I've got a little bit more space to move. My car moves a little bit faster. I get to that meeting a little bit earlier. I save time." And surely to a businessman, time means money. So at the end of the day, the $3 that he spends is more than recovered by the amount of time he has saved, by the amount of hassle he has forgone. We have only so much land. There are so many competing uses for our land. We have to make optimal use of our land for housing, for catchment areas, for factories, for playgrounds, for schools, etc. And among these competing uses, we have to build roads. Today, we already have 12% of our land devoted to roads. How much more land should we give up in order to build roads? We cannot be building more roads. So we manage the use of existing road space. We encourage people not to use their cars if they do not really need to. But how do we define need? Who are we to say that your need is greater than the next person's need? We do not wish to define that. We let the motorist decide. How do we make better use of the land set aside for roads? We give priority to buses. Why? Because buses make the most effective use of road space. And Mr Rai is absolutely correct and I am glad that, although he is in the minority, he has spoken up for the bus commuters, because this is also the reason why we are introducing ERP. We introduce ERP to encourage people to use our roads wisely. We say that pricing is the most efficient mechanism to allocate this scarce resource. At the end of the day, we collect revenue from people using this scarce resource. Mr Low Thia Khiang says that this is a revenue collecting mechanism. Yes, it is a revenue collecting mechanism, and I am not ashamed to say that it is a revenue collecting mechanism. But what do we do with this money? When we collect the revenues from people who voluntarily decide to use the roads, we take the money, which is the property of all Singaporeans, and we put it to good use, whether it is for public transport infrastructure, education or housing. It goes into consolidated revenue which is then used for the public good. And what is wrong with that? Mr Jeyaretnam then suggests that ERP is unfair because it only affects the poor, while the rich will not be affected because they can afford to pay. Has he not forgotten something? Most Singaporeans do not drive to work. They take the bus and the MRT. They will not be affected by the ERP. And, indeed, if we can make the roads clearer for the buses, then surely that must be welcomed by bus commuters if they can actually get to work on time and get to work faster. So, rather than saying that the ERP is actually discriminating against the poor, he has forgotten that actually there is a very large group of people in Singapore who will be benefiting, directly through faster travel, and indirectly through the revenues that we collect from ERP. Mr Low Thia Khiang and Mr Chiam See Tong made the comment that ERP is a revenue collecting mechanism. I have addressed that point earlier. Mr Chiam further asked whether COEs will be scrapped as a result of the ERP. Mr Low has already given him the reply, right answer but wrong working. I think he gets half mark for that. The right answer is that we cannot scrap COEs. We have to keep COEs, at the same time that we have to introduce ERP, because the two must go hand in hand. If we just carry on with COEs, we end up in a situation where it is very inflexible. We cannot continue to keep adding COEs. We are now talking about a 3% increase every year. Our land does not expand by 3% every year ad infinitum. At which point in time do we have to start to scale back on COEs? 20 years' time, 30 years' time, if we were to depend on just this mechanism alone? But if we also have in tandem a mechanism which regulates the usage of roads, which encourages people to use the roads more judiciously and if we put the two together, then we have two very powerful levers to allow us to manage road congestion but, at the same time, to allow for a regular increase in the number of vehicles on the roads, which is part and parcel of managing aspirations, managing people's desire to own cars. No, we cannot have ERP alone, we cannot scrap COEs, but neither can we have COEs alone without the ERP. Is it a revenue collecting mechanism? I say, yes. But we do it in a way that gives the motorists the choice whether they want to pay that fee for using the roads or not. The upfront taxes have already been reduced, slashed in some cases. The ERP revenue is an estimate. If the system is successful, there is really no reason why we should be collecting more revenue than what we are collecting today. On that basis, we have decided to make those tax changes on a revenue neutral basis today by giving rebates, by reducing taxes. The tax reductions alone, including the estimate of increased ERP revenue over the next five years, is going to cost the Government something like $1.2 billion, over five years. What happens after five years, Mr Low asked. The rebates will run out. By that time, people would have had time to adjust to the new ERP infrastructure. And in five years' time, if people continue to drive as they do today, if businesses continue to operate as they do today, in spite of the fact that now they will be charged for the use of the roads, that now they are going to be charged per pass, I would be very surprised. I expect that there will be changes and, in fact, this is what we hope will be achieved when we introduce the ERP, that we hope to change people's driving habits. And if motorists' driving habits are changed in such a way that they do not have to pay so much for ERP, so be it. That is the purpose of the ERP system. If they decide that they are going to drive irrespective of ERP charges, then we will collect the revenue, because the motorist willingly has decided that he is going to pay. How will the revenue situation be like five years from now? I do not know. All I know is that we have a mechanism in place which puts the responsibility firmly in the hands of the motoring public. And that, I think, is a fairer way of doing things. It is more rational and, I believe, it is more effective. Sir, I turn now to some of the specific concerns of Members. Mr Chay expressed some concern that motorcyclists are paying high ERP charges than taxis, and he felt that it was not fair. Dr Michael Lim also suggested that we should consider low ERP charges for those who use the roads more frequently. Sir, the principle of shifting upfront cost to the usage cost is really one of the fundamental principles of the ERP policy or strategy to manage demand for road space. Why? Because a vehicle owner who has paid a lot of money for his car or motorbike and pays very little or nothing for its use would have very little incentive to cut down on usage. So this is why in the new tax structure, we have reduced upfront taxes. We have cut registration fee. We have cut ARF and we have shifted more of the cost to usage, mainly through the ERP charges. The ERP is a pay per pass system. The more you use, the more you pay. It is fairer and it is more effective in encouraging road users to cut down on their usage. So on this basis, it is not possible to accede to Dr Michael Lim's suggestion that we levy a lower fee on those who, for one reason or another, need to use the roads more often. This is not fair to those who do not use the roads more often or who have managed to change their pattern in such a way that they do not use the roads more often. If we did that, it would defeat the purpose of the ERP. However, as Members would be well aware, we have recognised that this is a sudden change in the system. To the heavy users, it will come as a very sudden shock. To accommodate that, we have decided that we will phase in the changes over three or four years for taxis and commercial vehicles. And this is the reason why Mr Chay observed that motorcycles will be paying lower ERP charges than taxis, because we are phasing in taxi charges over three years. Of course, this only applies in the first few years. However, when the phasing-in period is over, ERP rates for taxis will be exactly the same as for private cars, by which time his concern is no longer valid. 4.00 pm Mr Chng Hee Kok made some suggestions on road pricing. Yes, we decided to introduce the concept of shoulder pricing because of this phenomenon of "peak within the peak" where everybody tries to jam into that particular stretch of road at that particular time. So we have decided that we will try and spread out that peak load by levying lower charges in the shoulder periods. However, "shoulder" pricing is a new concept, and I think we should start off with a more simple structure. Therefore, we have decided to have a peak price, a shoulder price and an off-peak price. It is a little bit discreet, I agree. But let us try the simple structure first and, if it works, we will extend it to other congestion points, and we will also look at varying the pricing to suit the traffic pattern more accurately, as suggested by Mr Chng. Mr Chng has some comments on gantries. Very quickly, I just wish to clarify that there are two gantries. Separate gantries for CBD and ERP are only a temporary problem. Come 1st September, there will only be one gantry. Location of gantries - Dr Michael Lim suggested that we review the locations of gantries so that motorists are not charged unnecessarily. Yes, I agree. With the present system, we already have some problems like that. So, with ERP, we will take the opportunity to review. But at the end of the day, I would advise all motorists to continue to plan their journeys well in advance. Mr Iswaran had a few queries about flexibility. We will be flexible, in terms of implementation and enforcement. But we hope that motorists will take the time and effort to get to understand the system, to use it and to be familiar with it, so that there are very little user-related problems, come 1st April. He asked about EMAS, what is EMAS. Basically, it is an early warning system, to put it very clearly. I am not familiar with the MINDEF jargon, but I can assure him that it is an early warning system, just like MINDEF has early warning systems, and equally useful for motorists once they get used to it. There has been a rather protracted period of testing which has confused some people and attracted some rather snide comments. But I would like to suggest that we give it a try when it starts and, if it works as well as planners feel, I think it will be a great boon to motorists. Incidentally, this is not something new. Such systems are well accepted in Europe and Japan. If you travel on highways in Japan, for example, you will see these numbers on the expressways which give you times. Those are the times which tell you how much congestion there is on that stretch of highway. So 50 minutes mean that you expect to reach the next junction in 50 minutes; please, if you can, avoid that junction. That is basically what those signs mean. But, as I said, it is something new, give it a try, and you will be pleasantly surprised at how useful it can be. Dr Toh See Kiat expressed concern about safety and reliability of the IU. I understand his concern because I think this is a matter which concerns every motorist, and for motorists to see a foreign object staring at him in the face in the car can be rather unnerving. But I have the assurance from the consortium which has tendered and is supplying the IU. It is an international consortium (Philips, Mitsubishi, Mitsui Electronics, CEI Systems in Singapore) and they have been awarded the contract. We have drawn up very rigid specifications for them to carry out very rigorous tests. In fact, the testing in the laboratory and the testing in the field took something like 18 months, longer than most new products which go to market. We had to be more than careful. We could have done the acceptance tests in a much shorter time. But we were dealing with 680,000 vehicles throughout Singapore and we wanted this system to last, we wanted the reliability of the system to be as high as possible, and we could not take chances. This is the reason why the testing took longer than usual. It is another reason why it cannot be miniaturised. The answer is that testing started off 18 months ago and during that time, many things had happened. But the suppliers assure us that it is safe. They have consulted, as far as they can, with motor manufacturers on where it should be located. The installers have also been trained to install it in such a way that the unit would not be dislodged. All in all, I think we have to take the results of the tests as reliable and we will implement it come 1st April, with the assurances that it will be safe and it will be reliable. But I will take Dr Toh's suggestion to put some of these details in the LTA Webpage, the Internet Website, eg, the FAQs, as he called it, on the IU and the ERP systems. If he is interested, the website address is "www.lta.gov.sg". If you cannot access that one, you can also go through the AA website which is "www.aas.com.sg", and this is hyperlinked to the LTA website. There were other concerns raised by Mr Sinnakaruppan. He is suggesting that we use other forms of payment for ERP, like NETS, cash, and credit card. When we first drew up the specifications, we had to decide whether we wanted to use a passive system, in other words, a centralised billing system whereby we will send you a bill every month, and Mr Sinnakaruppan would go and check his bill and see where he has been for the last month, and he would decide whether the bill is correct or not. That was one option. The other option was an active system, a smart card based system where you get deduction every time you pass through. No record is kept, no bills are sent to your house. We have decided to go for the smart card system for several reasons. First of all, in this way we do not need a centralised accounting and billing system, which is a very costly and very massive exercise. Secondly, by charging per pass, the moment you pass through a point and hear a beep, and you realise that your cash card balance has gone down by X dollars, you become more conscious of the cost of motoring. You become more conscious of the fact that you are actually passing through an ERP control point. Thirdly, by using smart cards, we can also help to move towards a cashless society. Smart cards are an important way of commercial transactions. They can become more important if we can encourage more people to use it. I believe that the ERP will provide one of these "killer" applications, as they call it, for smart cards in Singapore. And finally, last but not least, there is a question of privacy that I alluded to you earlier. Privacy is an issue for some motorists. In fact, I believe Mr Sinnakaruppan may have understated the concerns of motorists about privacy. It is a concern. If you recall, when Hong Kong was looking at the electronic road pricing system some years back, one of the reasons why they dropped it was because of this concern of privacy. And at that time the only system available was the centralised billing system. So I would not rule out this concern about privacy. But that is not the major reason why we chose the Cashcard system. As I said, there were many other reasons. Nevertheless, his suggestion is a valid one because if we can do it, it does help to make the system more convenient. It is a direct debit system. It is almost a Giro system. It is like the transit link card system where you do not have to top up your transit link card. Every time it falls below a certain balance, as it goes through the MRT turnstiles, it automatically tops up for you. So it is possible to be done. It is convenient. And we will look into that. But, again, considering the complexity of the system as it is today, I would like to take it step-by-step and see whether the system works first. And then later on, once motorists are more familiar with the system and are more assured of the reliability, safety and effectiveness of the system, then we can consider further refinements, such as those suggested by Mr Sinnakaruppan. Mr Goh Choon Kang suggested that the statutory lifespan of cars be extended to 12 years and the PARF benefits correspondingly. Sir, just let me point out that private cars do not have a statutory lifespan. In other words, you can keep your private car for as long as you want, so long as you pay the requisite taxes and so long as you pay the COE when it expires. So you can keep your car indefinitely, unlike taxis where you have to scrap the taxi after seven years, unlike buses where you have to scrap it after 15 years now, unlike rental cars, tuition cars, and so on. Private cars have no statutory lifespan. So the question of extending the statutory lifespan does not apply here. Yes, you can introduce an extended PARF scheme. But here I would be cautious because what we have tried to do with our vehicle population is to try and keep it young. We have an aging population problem with cars too, because the age of the car population is actually getting older and older every year. And today, about 25% of our cars on the road, one in four cars on the road, is more than 10 years old. So if we were to relax on this policy, I think we will see a much greater proliferation of old cars. And no matter how you look at it, no matter how well you keep your car, there is no doubt that, generally speaking, older cars are more likely to break down. Our statistics at the LTA inspection centres show that. The AA's statistics on breakdowns on the roads show that. There will always be exceptions. Many people can keep their old cars very, very well. But the exceptions do not prove the rule. The rule is that generally old cars tend to break down more often. So I would hesitate to tinker with this policy because I think it has served us well. Similarly, extension of COEs. First of all, we want to keep the population young. But I just want to highlight that if you extend your COE by two years, chances are that you are just going to bump up the COE prices, because the life of the COE has gone up by 20%, people are going to factor that in when they next bid. 4.15 pm Mr Gerard Ee suggested that we should lower the ARF further. The reason why the Ministry decided not to do this was because of the very significant knock-on effects of such a move. As he realises, when we made the change in 1991, we reduced ARF from 175% to 150% at that time. Very quickly the COE prices rose in tandem. So it is also going to happen this time. Of course, people may ask what is wrong with that. When COE prices go up, the prevailing quota premium (PQP) also goes up and that has an impact on all those who want to renew their COE after it has expired. So we have decided to err on the side of caution as far as this is concerned. We will reduce, if any, on a gradual basis. For this time, we have decided that a moderate 10% reduction is sufficient. Mr Gerard Ee also made a point about COEs for motorcycles. COE prices for motorcycles have fallen in recent months. In the past six months, about 90% of the quota of motorcycle COEs are actually used to register small bikes, those engine capacities of less than 500 cc. 90% of motor bike COEs are actually used for small bikes. So there is really no problem about small motorcycle owners getting their bikes registered or COEs registered. In fact, they are the ones who are really determining the price of the COEs, if anything. So I do not see any need at this juncture to do the separation. But, in principle, I must say that there is a case to be made for just one category for all vehicles. That is from the transport point of view. We just look at it from that point of view, we do not need to have Categories 1 to 4 for cars. But land transport tax structure, COEs, is not a straightforward matter. When we implemented the COE system, we divided it into four categories because we realised that, at that time, if buyers of small cars competed with buyers of large cars, there would be a general sense of unfairness. For that reason, we have had the COE system of today where there are four categories and where the price of each category is determined by how many COEs are there and how much demand there is for each one. Each category of COE price reflects the price that the motorist in that category is prepared to pay. It is not the Government that decides what price it is. It is the buyer. But there is a point that both Dr Wang and Mr Chng Hee Kok have made which I think is valid. We have had the COE system since 1990. It is timely for us to do a review from time to time. I agree that we should do a comprehensive review of the Vehicle Quota System this time round. In so far as timing is concerned, I think the appropriate time would be when the new vehicle tax changes are implemented and when the system settles down a little bit. Then I think we should do this comprehensive review and when we do that, I would welcome any and all suggestions on how we should rationalise the COE system. Sir, at this juncture, I would like to make an announcement on the COE quota for the quota year May 1998 to April 1999. The new tax structure that was introduced together with ERP implementation as well as the road tax package will not only allow us to control traffic congestion more effectively but, as I have repeated, it will also give us this opportunity to increase the number of COEs if traffic congestion is controlled. I have announced previously that we will issue about 15,000 to 20,000 more COEs over the next three to four years depending on the progress and effectiveness of ERP. If you may recall, we released about 5,500 additional COEs for this quota year ending this month, when we extended the Road Pricing Scheme to the two Expressways, and that was last year. In conjunction with the widening of the Road Pricing Scheme, we released the additional COEs. For the coming quota year, which is from May 1998 to April 1999, we will monitor the traffic patterns after the ERP has been implemented in the CBD - when we have extended it to the CBD and other RPS points on 1st September - before we decide on the appropriate number of extra COEs to be injected. Based on the current formula for computing COE quota, the total number of COEs for the next quota year is 47,885 which is higher than the current quota of 44,813 which included the extra 5,500 COEs. The LTA will release details of the quota of the individual COE categories shortly. Sir, I now turn to the taxi industry. I would like to thank Mr Seng Han Thong and Mr Goh Chee Wee for their support of the new vehicle tax structure. Mr Seng's concern was how the new structure can be translated to benefit taxi companies, taxi drivers as well as passengers. First of all, under the new tax structure, all existing taxis will enjoy the reduction in annual diesel tax. The annual diesel tax will be reduced from $6,600 to $5,100. LTA will meet the taxi companies and the Taxi Drivers' Association to work out the details on how taxi drivers can benefit equitably from these savings. Recently, I met the taxi drivers in a dialogue session. They suggested that the savings be used to top up their Medisave accounts. I will ask the taxi drivers, taxi operators, companies as well as the Association to put their proposals to the LTA and we will discuss together on how this can be done. I think it is a good idea. Mr Seng also asked about the future policy of diesel tax. Will it be removed? First of all, to answer that question, I think we must go back to the history of the diesel tax. What is the diesel tax? Very simply, it is a tax to equalise the difference between tax on diesel and tax on petrol. The tax on diesel is very much lower than the tax on petrol at the pump. There are many reasons for this. But that is a fact. So we try to make the cost platform for taxis and private cars more equal. This is where Mr Jeyaretnam just now asked, why do we tax taxi drivers so high, why do we not just remove the diesel tax, why do we not tax it on the same basis as public transport. Yes, we are taxing taxis as public transport but not mass public transport like buses. Taxis are a personalised form of transport. It is not a substitute for cars and therefore it cannot be taxed in the way that Mr Jeyaretnam suggested, like public transport, so that we allow everybody to take taxis. Why are we making it so expensive for people to take taxis? Taxis are not subsitutes for cars. But they are a form of public transport at the very high end and therefore they should, on the one hand, pay taxes similar to a car but, on the other hand, be given some concessions in recognition of their public transport role. This we have done in the new tax structure. So we have to impose the diesel tax to equalise this. We did this many years ago, calculated based on the average mileage travelled by taxis. With ERP, I would expect that the average mileage incurred by taxis will go down. Empty cruising will be reduced. Taxi drivers will be discouraged from driving unnecessarily looking for passengers as they do today. They will probably have to depend more on radio phones, radio calls or they will have to find some means of pooling their resources or whatever. The fact is that they will have to change their mode of operation in the same way that other industries will have to. If they do that effectively and if the mileage incurred goes down, then the basis of charging diesel tax, which is based on mileage, also suggests that diesel tax will trend downwards. But this is really up to the taxi operators, how they drive, how they react to the ERP system. The question of taxi cartels came up from both Dr Toh See Kiat and Mr Iswaran who were concerned that taxi companies may form themselves into a cartel. Sir, under the new tax structure, all taxis receive reduction in diesel tax. They will all receive road tax rebates. We design the tax structure so that taxes for taxis, as I said earlier, will be comparable to taxes for private cars. The new tax structure also has been configured in such a way that the total cost of registering and running a taxi after the new taxes are implemented will not be more than what it is today. So there is no reason for taxi fares to increase suddenly when they are deregulated on 1st September. I expect that taxi operators, as Mr Goh Chee Wee has stated in this House just now, will set their fares sensibly. They will have to see what the demand is like. They will have to match supply with demand. If they price their services too high, taxi commuters have other choices. They have got bus and the MRT. They can opt not to take a taxi. If they set their prices too low, then of course, there will be overwhelming demand and they will have to decide what is the correct level. It is a difficult task for the PTC to have to sit in judgement and to decide and arbitrate on taxi fare increases every time the taxi operators come to the PTC with their fare increase. They are not the ones who run the taxi companies. They are not the ones who drive the taxis. They have to look after the interest of the commuters. But in doing so, in looking after the interest of the commuters, they also have to balance the need to provide the taxi companies with a reasonable return and the taxi operators and drivers with a reasonable livelihood. This is a very difficult task. And every time they sit to arbitrate and decide on this, there is a lot of argument and at the end of the day, nobody is happy with the result. So it is appropriate now that we have set the tax platforms and the tax structure correctly, that we return this responsibility to the taxi operators and the taxi companies. Of course, I would expect taxi operators to be responsible, to recognise their public transport role and the service to their customers in return for the freedom to set their own fares. The PTC will continue to monitor the taxi industry after deregulation. It will need to set guidelines on taxi fare revisions which taxi operators must observe and it will have to ensure that commuters are given adequate information to make informed choices. 4.30 pm Dr Toh also asked the industry to set minimum standards. Yes, the LTA has introduced this performance evaluation scheme based on several criteria, and it will set the minimum performance standards for all taxi operators. We will watch and see what the situation is like, how it develops after deregulation. We will observe how the taxi companies behave, whether in fact, they are going to form a cartel to exploit the customer. If necessary, we will further deregulate the industry, as Dr Toh has suggested, to allow new entrants to come in to increase competition, and to let the market decide on the supply of taxis. Mr Shriniwas Rai posed the question whether we can make taxis and car rental companies more attractive as an alternative to private cars. This is, in a way, what we have done with the new tax structure. We have re-looked at SZ cars and taxis and made all their operating costs roughly more or less the same. For example, for new rental cars, we have increased the statutory life span from seven years to 10 years. Taxes for taxis, we have treated them more like cars. Fares will be deregulated. So the stage is really set for all these different modes of transport to compete for customers in the way that they know best, to offer better service, more comfortable rides, or cheaper fares on weekdays, Sundays, as Mr Goh mentioned just now. It is really up to the taxi companies now and they are free to do so after deregulation. With the tax changes, we are now, in a way, offering more choices. We are saying to those people who wish to continue to drive a car, here is a new tax package for you, slight reduction in upfront taxes, more weightage on usage, you decide what your lifestyle is like, which one suits you best. If you want the door-to-door convenience of a car, you want the reassurance of a car whenever you need it, you will buy a car. If you want to trade off some of that convenience for savings, you can either choose public transport or you can choose a combination of public transport and private transport. You can choose to rent a car, you can take a cab or you can join a car cooperative. So the choice is really up to you. Mr Rai also mentioned that we should convince Singaporeans, especially graduates, that it is not possible to allow every graduate to own a car. I would like to assure Mr Rai that this is what we have been trying to do all these years. It is not an easy dilemma that we have to resolve. It is a tough challenge because, on the one hand, with progress and development, rising expectations invariably mean a demand for cars; on the other hand, a growing demand for travel overall. So we have to resolve this dilemma. The Government has always stressed that cars are not like HDB flats. They are a luxury, albeit a legitimate luxury, as a recent Straits Times editorial put it, but a luxury nevertheless. So, while the Government can provide every household with a HDB flat, it cannot guarantee that everybody will own a car. This is the reason why we have put in so much effort into building up public transport, not just a public transport system that is so-so or can do, but we hope that we will improve the public transport such that it becomes a viable alternative to a car for most people. We will continue to try and resolve this dilemma. We will continue to convince people that public transport is a viable alternative. But, at the same time, we will continue to try and meet this aspiration for people to own a car. That is the reason why we have ERP, because with ERP we can allow more people to own a car. Mr Jeyaretnam said just now that he cannot understand why we want to issue more COEs. I could not understand why he could not understand that. If he has been on the ground and going around from house to house and talking to people, I think he will understand the strong desire of people to want to own a car. The other example of how we can meet this aspiration is to try and introduce what I call more personalised transport services. Under this category, I refer to taxis, rental cars and, recently, to the car cooperative scheme. I think there is just one last point that I have not addressed yet, and this is on Dr Toh See Kiat's question on the cost effectiveness of underground roads and flyovers. Basically, we have to build underground roads and flyovers because of limited land. There are many junctions in Singapore today which are very heavily used. It is no longer sufficient to just have a two-tier flyover. We need to have three-tier flyovers, because of the heavy flow of traffic in all directions. He asked about criteria. We have to take into account traffic volumes and traffic flows. We have to take into account the site constraints. We have to take into account the topography. We have to take into account the value of the land in the area, whether we go over or under, and so on and so forth. These are some of the considerations that have to be taken into account. The Holland Road three-tier interchange, for example, was selected after taking these factors into consideration. Design studies were made, evaluated against various criteria and the final choice was made. Basically, we needed more three-tier interchanges because we have to relieve the existing traffic congestion at these junctions, and because it is part and parcel of our plan to optimise the use of our roads. Junctions are where bottlenecks occur. So when we have major traffic arteries, we have to make sure that those bottlenecks are removed. Otherwise, we are under-utilising our roads. Sir, let me just conclude by taking this opportunity to remind Members that our land transport policy is not just ERP alone. Over the last so many minutes, I have been answering queries about ERP, but it is not just ERP. It is about more choices for the public in terms of good public transport. It is about expanding the road network. It is about building three-tier interchanges. It is about managing our roads to relieve congestion. It is about issuing more COEs so that more people can aspire to own cars. Since we debated the White Paper on Land Transport in this House in 1996, we have achieved a lot, but a lot more needs to be done. I am confident that as we press on with the implementation of the various policies and schemes that have been mapped out in this White Paper, we will achieve our vision of a world-class land transport system for Singapore.