Recalling the analysis in Module 2, an obvious comparison would be between the total expenditure on subsidies and total government expenditure. (It would be misleading to include tax-expenditures in the total aid figure, for tax-expenditures are not a component of government expenditure. (As already explained, tax-expenditures offer a subsidy through reductions in tax obligations.))
Table A1.2 extracted from the EC Report cited in Module 6 (EC 1992) shows how such a comparison may be used to show the relative importance of subsidies in the government budget of different EC countries:
| Italy | 5.9 |
| Ireland | 4.9 |
| France | 3.9 |
| Germany | 5.2 |
| Belgium | 5.7 |
| UK | 2.8 |
| Netherlands | 2.1 |
| Denmark | 1.9 |
As one important justification for subsidies is their employment-creating capacity, some comparison might be sought between subsidy expenditures and the number of persons employed in subsidized sectors. A simple way of doing this is to work out the total subsidy element per person in employment, as shown in Table A1.3.
| Italy | 999 |
| Belgium | 1046 |
| Ireland | 635 |
| France | 757 |
| Germany | 967 |
| Netherlands | 521 |
| UK | 306 |
| Denmark | 397 |
As in the previous example, the dispersion between countries is quite wide.
For good measure, we add a third comparison. It may be of policy significance to investigate the comparative importance of subsidy in relation to the total value of output in different sectors, e.g. industry, agriculture. The EC study produces a comparison between total subsidies with the gross value added in industry and in agriculture. The results obtained are shown in Table A1.4.
| Manufacturing | Agriculture | |
| Italy | 6.1 | 12.9 |
| Ireland | 5.7 | 5.5 |
| France | 3.7 | 9.2 |
| Netherlands | 3.1 | 6.8 |
| UK | 2.3 | 8.7 |
| Germany | 2.6 | 20.1 |
| Denmark | 2.0 | 7.9 |
There are two points to be made about your comprehension of the classification and comparison of aid:
You are not expected to memorize data of the kind presented. In any examination paper, the data will be given. What you will be asked to do is to make the data tell an interesting story.
If you choose to answer a question on the interpretation of data, you will find it useful to employ the techniques learnt in the Quantitative Methods Course. Although the sample of countries in the tables presented above is small, you can practise your skills in interpretation by measuring the degree of dispersion between countries in each comparison, and by calculating some simple correlations between the different series presented.
If the government uses taxpayers' money in order to increase the capital stock, this must deny the use of the resources for some other purpose – other government expenditure, private investment or private expenditure. There must therefore be a prima facie case for applying some rate of discount to future returns from this addition to the capital stock. A simple rule would be to apply a discount rate which represents a rate of return at least as high as that applied to private investment projects. The British Treasury, for example, has used a rate of 5 per cent for some years, with exceptions made in particular cases. This rate of discount is sometimes known as the social opportunity cost of capital (SOC).
A ‘rival’ approach to the problem of discounting emphasizes that a government is not like a business – an argument to be explored further below – and that any discounting procedure must consider the preferences of individuals who elect governments. Furthermore, individuals who elect governments are not like shareholders who are presumably interested in obtaining an income from a business at least equal to that obtained by investing elsewhere. They may wish to consider the interests of future generations who have no say in the present-day decisions of the government. Therefore, the rate of discount should reflect the time profile of the benefits of all those whose preferences are considered relevant. This social time preference rate (STP) is often used as an argument for a lower rate of discount than that suggested by the SOC approach. Thus projects which would not pass the SOC test might be accepted. This argument has a particular appeal to those who place a high value on preserving the environment, e.g. in slowing up the depletion of natural resources.
The argument needs to be taken further, for, as Section 6.2 demonstrates, the benefits from subsidized projects frequently include non-marketed items. Thus expenditure on roads may take into account a valuation of the reduction in the number of deaths and injuries estimated to result from additions and improvements in the road network. Conversely, if a hydroelectric scheme were to destroy an area of great natural beauty, which cannot be re-created, a valuation of the forgone benefit from this environmental loss, should be treated as a cost.
Finally, businesses distribute the profits from investment projects according to the distribution of shares in the company. Elected governments must also take account of the distribution of the benefits and costs of investment projects. Thus a new international airport may pass the ‘discounting test’, if the distribution of the benefits and costs are ignored, but may not be acceptable to the voters if it is perceived that the benefits to travellers are outweighed by the costs to those who have to suffer extra noise and pollution.
Note that the reason for taking the argument further than the discussion of the method of discounting arises from the fact that it is often argued that ‘adjusting’ the discounting rate is an alternative way of taking into account non-marketed benefits and costs, and their distribution between present and future generations. Economists, at least, seem to be agreed that this is an inadmissible procedure. Apart from the empirical evidence presented by the existence of positive interest rates reflecting a preference for present over future benefits, analysts of government projects would have the impossible task of selecting different STPs for different projects. Instead of using this dubious procedure, it seems more sensible to reflect non-market considerations in the valuation of costs and benefits before applying an overall discount rate. (For further discussion, with particular reference to environmental factors, see Chapter 6 of the now famous Blueprint for a Green Economy (Pearce et al 1989).)
Following the hint offered in the text, the first step is to calculate the expected value of the outcomes for each project. To do this we simply multiply the value of each outcome by its respective probability and add them up.
| Project A |
| +10.0×(0.1)+1.0×(0.4)−1.0×(0.4)− 10.0×(0.1) =+£0.0m. |
| Project B |
| +10.0×(0.1)+5.0×(0.4)−4.0×(0.3)−10.0×(0.2)= −£0.2m. |
From the point of view of the 50 per cent probability of total recovery rule, both projects appear satisfactory. In both cases the subsidy will be more than recovered with a probability of 0.5. However, this does not mean that the expected value of the excess recovery is positive. If the government continued to finance projects of type B it would recover the subsidy in 50 per cent of cases, but it would lose money over the long run. This is simply because, in the 50 per cent of cases in which the subsidy is not recovered, the expected deficit (£−6.4m.) exceeds the expected surplus (£+6.0m.) generated by the 50 per cent of cases in which total recovery is achieved.
This question exemplifies an important feature of government policy towards industry – considering the costs and benefits of alternative courses of action. In this case, the government is considering the relative merits of using a subsidy or a tariff in influencing the output and employment in a domestic market in which overseas suppliers have an incentive to sell.
The first stage of the answer is best completed by redrawing Figure 6.1 to produce Figure A1.3. Taking Figure 6.1, the first step is to reproduce the situation where there is no subsidy. Output is Qd and price Pw, for the domestic firm, with output OQ−OQd supplied by overseas producers. Now a tariff is to be introduced which will raise price sufficient to induce the domestic supplier to offer OQ’d. This is shown by shifting the (competitive) price line BP upwards until B’F = BE. At the higher price Pt(=OB’), the total quantity demanded is reduced from OQ to OQ’.
Figure A1.3 The effects of a tariff on imports
The domestic firm is placed in the same position with the tariff as with the subsidy. Although this is not shown, the fixing of the tariff clearly depends on the information available about marginal costs of production, which are only known to the firm, so that the asymmetric information problem remains. This additional complication is not shown in the redrawn diagram, but it can be left out of account in considering the relative position of the different income groups affected by the tariff.
The overseas suppliers both lose part of their market, as with the subsidy case, and in fact lose more because total quantity demanded has fallen as a result of the rise in price – Qd’Q’ is less than Qd’Q. Their output is subject to a tariff equal to BB’ per unit of supply, totalling FEE’P’.
Domestic consumers have now to pay a higher price. If the product supplied is a raw material, this will probably be translated into a rise in prices of domestic final products, with a loss of welfare to final consumers. On the other hand, domestic consumers, as taxpayers, no longer have to meet the cost of a subsidy and the receipts from import duties as a result of the tariff present an opportunity for the government to lower other taxes.
Some highly specialized economics concerns itself with working out the welfare effects of alternative policies, with a view to determining which policy will be optimal or at least how alternative policies should be ranked. Our concern here is simply with an examination of the kinds of considerations which will have to be taken into account by governments concerned with balancing the various interests which affect their own position. It will be noted that the diagram only shows the short-term position. The imposition of a tariff may provoke retaliation against a country's exported goods and services by other governments, and is a much more transparent form of discrimination in favour of domestic production than a subsidy. It may not even be a feasible option, if a country subscribes, as in the case of EC countries, to a code of practice which rules out barriers to overseas trade. It is noteworthy that much more progress has been made in the EC towards removing tariff barriers than towards the banning of subsidies on goods which are tradeable between countries.
