| A | B | C | D | E | ||
| F | G | H | I | J | ||
| K | L | M | ||||
| 40 | 5 | 5 | 20 | 5 | ||
| 5 | 5 | 100 | 5 | 5 | ||
| 20 | 5 | 30 | ||||
| +200 | +10 | +10 | +20 | 0 | ||
| +10 | -10 | -100 | -40 | +20 | ||
| +60 | +20 | +60 | ||||
| +80 | 0 | 0 | +40 | +10 | ||
| 0 | -5 | +50 | +5 | +10 | ||
| +60 | 0 | -60 | ||||
| +5 | +2 | +2 | +1 | 0 | ||
| +2 | -2 | -1 | -8 | +4 | ||
| +3 | +4 | +2 | ||||
| +2 | 0 | 0 | +2 | +2 | ||
| 0 | -1 | +½ | +1 | +2 | ||
| +3 | 0 | -2 | ||||
Michael Baker - Dissertation - Equity in Transport Planning
About | Contents
Previous: How Equity Might Be Considered | This: An Alternative Approach | Next: Section III
In an attempt to make 'rational' decisions, that is decisions which follow the 'synoptic ideal'[1] , planning methodology has been developed along the lines of the "alternatives" approach. The method is, having ascertained the goals and objectives, to postulate alternative solutions which may achieve the goals and objectives. The alternatives are evaluated and the one which satisfies the goals and objectives best is chosen. This method, due to the limited number of alternatives tested, is very unlikely to find the 'best' plan, especially when the criteria for 'best' is expressed as a function which has to be maximised.
Further difficulties arise if there is more than one objective which is to be maximised. Either,all but one of the objectives must be reduced to constraints, or a composite objective must be produced. There is a severe problem in formulating a single composite objective, it is; what weight should be given to each of the individual objectives? Often changing the weighting of the various objectives will alternative which is chosen.
This makes the procedure very unsuitable in a political process since the only part the politician plays in the procedure is the setting of the objectives and their weighting. If there is more than one politician it may not be possible to obtain a set of weightings which is acceptable to a majority let alone one which is acceptable to all.
Over and above the problem of finding an acceptable set of weightings is that of the politician not being able to comprehend the effects of different weighting systems. This is because the setting of weights is so far removed from the final result in the decision making process. However it is at this point at which the politician, and more particularly the public in the case of public participation, must act if any effect is to be made on the result, within the confines of this type of decision making process. If the public do not affect the decision this way, or even if they do, they are unlikely to accept the results of a decision which is made on the basis of a weighting system which is different from that which they would have chosen individually.
These problems of lack of involvement and of what weightings to use will also occur in cost-benefit analysis if the values of the costs and benefits are to be weighted on equity grounds. There is a fundamental problem of trying to reduce a description of the incidence of costs and benefits to a single value, since a single value can cover a multitude of possible distributions.
Testing a few alternatives is a poor way of finding the solution which maximises a given function. It is even worse when there are several objectives or considerations of the distribution of the function over the population are concerned.
The "alternatives" approach to decision making could be replaced by a "learning" or "iterative" approach. It would be more likely to find a closer approximation to the 'best' plan, especially when 'best' is expressed in terms of maximising a given function. The procedure involves, the formulation of a solution, the testing of the solution against the objectives, and then in the light of the results of the evaluation, the solution is improved. These stages of 'testing' and 'improvement' can be repeated as often as is cons idered des irable.
There is scope in this procedure for a greater involvement of the politicians and public, especially when multiple objectives or distribution are concerned. That is at the stages between testing, and plan modification. At this point the politician or public can say in which areas the modifications should be made.
This is a proposed method for identifying who bears the costs and who receives the benefits in the testing of a transport plan. The results would not be used to choose between plans but rather to indicate areas in which the plan should be modified. This would lead to an iterative process of producing the plan. It should be possible to devise a process whereby the effects of changes in the transport network can be analysed without having to re-run the entire transport model every time modifications are made [2] .
The basic framework for this type of analysis has been proposed elsewhere in unpublished papers by Michael Tyler and John Popper. The intention is to identify all groups which bear costs and receive benefits. Of prime importance will be the incidence of costs and benefits on the consumer. That is because it is in this sector that it will be most difficult to make transfer payments from net-gainers to net-losers, or to make the distribution of benefits as even as possible. Of course it is also important to consider the effect on other sectors such as on commerce and industry who use commercial vehicles, where benefits and costs do not directly affect the final consumer. It is necessary to consider them separately since it would be very difficult, if not impossible, to trace the benefits to commerce and industry through to the final consumer, as well as to its owners and to the government (from changes in tax revenue). It is of considerable importance that the effect which any changes will have on the transport operators should be taken into account. It has been pointed out, in Foster & Beesley (1965) that the level of benefits is affected by the level of fares, and by the level of tax revenues to the government.
A fundamental differentiation can be made between those costs and benefits due to differences in operation of the transport system[3] , which affect the users, the operators and the government, and all other costs and benefits. These 'other costs and benefits' are both those occuring when physical changes are made to the transport system[4] , and those which occur due to the operation of the transport system[5] other than those mentioned above. Secondary effects such as the benefit of an enlarged catchment area to a shop pose problems. There is both the danger of double counting and more importantly, they are extremely difficult to analyse and quantify.
The first set of changes can be predicted directly from the transport model[6] . Person trip user costs can be evaluated for all trips originating in any zone. Since most trips are home based, it is possible to identify which households receive benefits so long as non-home based trips are considered separately.
Some of the public transport operators' costs are fixed, but are dependent on the scale of operation and some vary with the length and density of routes served. The public transport operators' revenues can be calculated by summing the product of the individual elements of the inter-zonal fare matrix, which can be produced from the inter-zonal distance matrix. (These matraces are created as part of the working of the transport model)[7] .
The tax payed by private road users can be estimated by multiplying the number of interzonal private trips by the corresponding interzonal distances by the tax rate per mile and summing over all trips. This does not include the road fund tax payed on vehicles, but if the level of provision of roads is assumed not to affect car ownership, and since only differences in tax revenue are being considered, this does not matter.
The incidence of costs and benefits can be represented in Tabular form such as in Figure 2 based on a table in Tyler's Paper. The important part is the incidence of costs and benefits as they fall on persons living in specific areas.
Figure 2 Framework for showing the incidence of costs and benefits
| Sectors to which benefits accrue | Users, area specific | Users, non area specific | Goods area specific | Operator(s) | Government | Social Benefit |
| Effect on Public Transport | ||||||
| user benefits | + 160 | + 20 | 0 | 0 | 0 | + 180 |
| revenue | - 40 | - 5 | 0 | + 45 | 0 | 0 |
| operating cost | 0 | 0 | 0 | - 30 | 0 | - 30 |
| Effect on Road System | ||||||
| (Private) user benefits | + 190 | + 10 | + 80 | 0 | 0 | + 280 |
| tax | - 10 | + 5 | - 10 | 0 | + 15 | 0 |
| accidents | + 20 | + 5 | 0 | 0 | + 10 | + 35 |
| Capital cost and maintenance cost | 0 | 0 | 0 | 0 | - 200 | -200 |
| Total benefit | + 320 | + 35 | + 70 | + 15 | - 175 | + 265 |
All entries are made in terms of present value resource costs.
The "Social Benefit" of items such as the public transport operators "revenue", and "tax" will be zero since these are transfer payments.
The "Total benefit" is the effect the change under study has on each of the identified sectors.
These can best be presented in map form, showing the level of costs and benefits per person for each zone. An example of how user benefits might be presented is given in figure 3.
Figure 3 User cost-benefit distribution surfaces
|
|
| ||||||||||||||||||||||||||||||||||||||||||
| Area of Study | Schematic Representation | ||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||
| Population ('000) | |||||||||||||||||||||||||||||||||||||||||||
|
| ||||||||||||||||||||||||||||||||||||||||||
| total Car user benefits (£M) | Total Bus user benefits (£M) | ||||||||||||||||||||||||||||||||||||||||||
|
| ||||||||||||||||||||||||||||||||||||||||||
| Per capita Car user benefits (£'000) | Per capita Bus user benefits (£'000) | ||||||||||||||||||||||||||||||||||||||||||
Both the costs and benefits due to construction and the 'other' costs and benefits due to the operation of the transport system fall mainly on people who work or live in specific areas. They will also be more difficult to calculate than the previously mentioned costs and benefits which can be calculated in the main by manipulating matraces which are produced as part of the transport model.
The costs which are associated with construction of transport infrastructure are, the direct resource costs of the construction which are borne by the government, and costs imposed due to the disturbance caused by the construction. Here and elsewhere government is taken to mean both the local authority and the national government. Disturbance costs are borne by the people who live and work and own property on or near the site of the construction, and by those who are delayed or otherwise inconvenienced on the existing transport system.
To the extent that compensation is payed for the disturbance caused, the incidence of the cost is transferred to the government. The cost of the disturbance to the owner of land which is used, can be taken as the market value of the land. Since the government pays market value as compensation, the cost can be considered to be transferred from the owner to the government. If full compensation is not payed, (for example it may be considered that market value is not adequate compensation for displaced property owners) only part of the cost is transferred to the government, the rest (that which is over and above the compensation) is still borne by those affected.
The costs caused by the operation of the transport system but not borne by the users are those such as noise and air pollution, which can in general be identified as falling on people in given areas unless the government pays compensation for these costs, in which case the incidence of cost is transferred to the government.
These costs both due to construction, and the operation of the transport system, which are not compensated for and so transferred to the government, can also be represented in map form, showing the level of costs per person for every area.
Having produced an anlysis of where costs and benefits fall it should be possible for the politicians and public to determine whether the distribution is acceptable or if an attempt should be made to change the costs and benefits falling on people in any given area. This analysis can also be used to test the effects, of raising or lowering the fares, or making other fare structure changes, on the users of public transport, other road users, and to the profitability, or need of subsidy, of the public transport operators. Conversely it can be used to assess the effects of subsidies to the public transport operators.
As an example of bow the procedure would work, changes under consideration might give rise to an incidence of user benefits as depicted in figure 3. This would indicate that car owners from G, H and I, and bus users from G and M would suffer net losses. Car users from A would receive a considerably higher net benefit than those in other zones. It may be considered that all of these are unacceptable and in an attempt to make an improvement changes could be made to the transport plan. An orbital motorway from L to B could be deleted from the plan, the traffic management scheme in G H and I could be improved, and an increase in size and frequency of bus services to and from M could be added to the plan. These changes would then be tested and the new incidence of costs and benefits displayed.
There are problems, unfortunately, inherent in the proposal. They are similar to those which affect present methods, many of which are outlined in the next two sections. The final section deals with how they affect the proposal.
[1] The "Synoptic ideal" is that the decision maker
[2] See Halder (1970) where a method of determining "..... the effect on the trip assignment of changing parts of the network" is outlined.
[3] These costs and benefits are negative and positive reductions in costs and increases in revenues between the "do nothing" system and the system under test.
[4] These include the capital cost which fails on the government and/or operators, and disturbance costs which fall on those whose housing is affected by construction, and those users of the pre-existing system who suffer delays.
[5] Such as costs imposed due to noise and air pollution.
[6] The basic method is that proposed in Popper's paper.
[7] All the user costs and benefits could be calculated without difficulty from data which is used within present transport models.
About | Contents
Previous: How Equity Might Be Considered | This: An Alternative Approach | Next: Section III
Copyright © Michael Baker 1974,2005. All Rights Reserved.