4. SOCIAL APPRAISAL OF THE KENANA PROJECT: INTRODUCTORY POINTS

The project was undertaken in the context of continuing attempts by the Sudan government to promote economic growth through the agency of development planning. A ten-year plan was initiated in 1961-62 followed by a five-year plan for 1970-75, introduced in 1969. This five-year plan was amended by the Interim Programme of Action (1972/73-1976/77) and it was this programme which contained provision for the Kenana scheme.

A linking theme of the plans to date, in common with most other LDCs, has been the need to promote economic growth and the (possibly) conflicting objective of more equitable distribution of wealth.

These concerns imply that the project's contribution to savings and to the consumption of poorer groups, be given direct expression in the analysis. In this appraisal we adopt the `numeraire' advocated in the UNIDO (1972) manual of aggregate consumption benefits. The technique involves the `weighting' of consumption by different groups.

One of the important decisions to be made in the analysis concerns the treatment of the capital costs of the project. Since a substantial part of these costs is to be met by foreigners' equity investment (cf. Table 3), these capital funds will be treated as if they had no opportunity cost for Sudan. This would be the usual assumption for foreign direct investment, since the alternative for the foreign firms of the project in question would be to invest in another country and not another project in the same country. A complication in this case, however, is that the political conditions of the time suggest that the Arab governments may well have been willing to invest similar amounts in other projects in the Sudan. If the volume of funds could be regarded as freely available to the Sudan, it would be sensible, from Sudan's viewpoint, to channel them to the relatively lower yielding public sector projects with Sudan taking a larger share of the most profitable ones. This would be analogous to the situation where a country is contemplating the alternatives of a purely foreign-owned direct investment and the possibility of doing the same thing from its own financial resources. It is argued, for instance, by Little and Mirrlees (1974, p. 121; 1990) that in these circumstances the government would choose the latter if the expected rate of return on the foreigners' equity participation appeared to be higher than the country's own accounting rate of interest (by which it discounts returns on their public sector projects). In the Sudan's case, this suggests that the project should be penalised if it is expected to produce such a premium for foreign shareholders since they would be receiving a higher return on their funds than the implied cost of capital for the Sudan economy.³ This appears unlikely given the acute shortage of foreign exchange faced by the country, to which further reference will be made when foreign exchange premia are introduced.

Before conducting the full analysis the following points may be noted:

1) In addition to the value of the `conventional' output from the project a further indirect benefit of some significance is generated by housing and welfare facilities for about 11,600 Sudanese employees.<sup>4</sup> In assessing these benefits we impute a 16% rate of return on investment of S£ 33.92 million in these facilities. This in turn suggests an annual `social' return of around S£ 5.2 million.

2) Table 11 provides further details of the information given in Table 10 in the form of local and foreign currency costs which will be the foundation of the analysis. Of further importance is the breakdown provided regarding skilled and unskilled labour. The former refers to engineers, surveyors and all employees with higher education. Semi-skilled workers, such as drivers, are regarded as unskilled since the error of undervaluing their opportunity cost is probably small and since some will receive `on-the-job' training⁵ for which the project should receive credit.

Unfortunately, full details of the breakdown of labour skill and domestic and foreign exchange cost of materials used in production were not available and assumptions have had to be employed.

The principle being followed here is to give approximate breakdowns to all social costs and benefits and to use the overall percentages throughout the project life-span in the hope that pluses and minuses cancel or nearly so (a point of view also expressed in the UNIDO, 1972, and by Little and Tipping, 1972, in their case study of the Kulai Oil Palm Estate).

Luckily the foreign exchange component of capital costs, which represents about 70% of the total investment cost, proved to be relatively easy to split between imported materials and foreign personnel. Also costs of production have been reported in a helpful manner by putting expenditure on materials and on wages and salaries separately. In the absence of any relevant information, we guessed the

Table 11: Benefits, Costs and Cash Transfers of Kenana Sugar Project(Part 1)

(in million S£)

Year	1	2	3	4	5	6	7	8	9
Price (FOB Port Sudan) 2. Repatriated Profit (Foreign Exchange)					0.1	20	26	28	27
3. Repatriated Equity (Foreign Exchange)
4. Scrap Value (Foreign Exchange)
5. Reclaimed Working Capital: 5.a Foreign Exchange 5.b Domestic Materials
6. Indirect Benefit: Welfare & Housing (Domestic Materials)	5.2	5.2	5.2	5.2	5.2	5.2	5.2	5.2	5.2
7. Capital costs  7.a Foreign Exchange 7.b Foreign Personnel 7.c Domestic Materials 7.d Skilled Labour 7.e Unskilled Labour	6.4 2.9 1.6 1.1 0.45 0.32	70.4 32.4 16.8 12.0 5.0 3.6	139.2 64.2 33.4 23.4 10.0 7.0	120 54.8 28.6 20.0 8.5 6.0	56.8 26.2 13.6 9.6 4.0 2.8	30 13.8 7.2 5.0 2.2 1.4
8. Costs of Production 8.a Foreign Exchange 8.b Domestic Materials 8.c Skilled Labour 8.d Unskilled Labour				11.8 1.8 3.2 1.2 5.6	14.8 2.28 4.0 1.3 7	21.8 3.2 5.8 3.4 10.2	44.44 6.4 12 6.9 20.8	7.2 7.3 12.8 7.2 22.8	50.8 6.6 13.6 6.6 23.8
9. Transport to Port Sudan 9.a Foreign Exchange 9.b Unskilled Labour 9.c Skilled Labour						1.92 1.4 0.4 0.2	4.6 3.2 1.0 0.46	6.3 4.4 1.2 0.54	7.6 5.4 1.6 0.8
10. Renewal & Replacement Cost 10.a Foreign Exchange 10.b Skilled Labour					4.64 3.50 1.16	5.12 3.8 1.28	5.6 4.2 1.4	6.08 4.6 1.52	6.08 4.6 1.52
11. Cash Transfers 11.a Income Tax 11.b Potential Subsidy (Foreign Exchange)					11.8	7.0	13.8	21.8	29.2
12. Working Capital					16.0	6.9

Table 11: Benefits, Costs and Cash Transfers of Kenana Sugar Project (Part 2)

Year	10	11	12	13	14	15	16	17-30	31
1. Output at World Sugar Price (FOB Port Sudan)	112	110	112	115	108	110	110	110	-
2. Repatriated Profit (Foreign Exchange)	27	26	27	28	25	11	11	13	-
3. Repatriated Equity (Foreign Exchange)									191.4
4. Scrap Value (Foreign Exchange)									83.4
5. Reclaimed Working Capital: 5.a Foreign Exchange 5.b Domestic Materials									22.8 16 6.88
6. Indirect Benefit: Welfare & Housing (Domestic Materials)	5.2	5.2	5.2	5.2	5.2	5.2	5.2	5.2	-
7. Capital costs 7.a Foreign Exchange 7.b Foreign Personnel 7.c Domestic Materials 7.d Skilled Labour 7.e Unskilled Labour
8. Costs of Production 8.a Foreign Exchange 8.b Domestic Materials 8.c Skilled Labour 8.d Unskilled Labour	50.8 6.6 13.6 6.6 23.8	50.8 6.6 13.6 6.6 23.8	50.8 6.6 13.6 6.6 23.8	50.8 6.6 13.6 6.6 23.8	50.8 6.6 13.6 6.6 23.8	50.8 6.6 13.6 6.6 23.8	50.8 6.6 13.6 6.6 23.8	50.8 6.6 13.6 6.6 23.8	- - - - -
9. Transport to Port Sudan 9.a Foreign Exchange 9.b Unskilled Labour 9.c Skilled Labour	8.16 5.72 1.64 0.82	8.16 5.72 1.64 0.82	8.16 5.72 1.64 0.82	8.16 5.72 1.64 0.82	8.16 5.72 1.64 0.82	8.16 5.72 1.64 0.82	8.16 5.72 1.64 0.82	8.16 5.72 1.64 0.82	- - - -
10. Renewal & Replacement Cost 10.a Foreign Exchange 10.b Skilled Labour	6.08 4.6 1.52	6.08 4.6 1.52	6.08 4.6 1.52	6.08 4.6 1.52	6.08 4.6 1.52	6.08 4.6 1.52	6.08 4.6 1.52	6.08 4.6 1.52	- - -
11. Cash Transfers 11.a Income Tax 11.b Potential Subsidy (Foreign Exchange)	29.8	28.8	27.8	27	27	26.6 27	26.6 27	32.8 27	- -
12. Working Capital					16	6.9

(in million S£)

breakdown of wages and salaries, but this is a very minor item. When it came to a domestic currency component, matters become somewhat more complex and consequently our procedure is a rough-and-ready method. We believe that little damage was done to our evaluation by this procedure since the capital expenditure in terms of local currency was rather small (29% of the total). Moreover, the disaggregation process will hopefully minimise the errors.

The last items which deserve some explanation are the breakdowns of rail transport and replacement costs. Following Little and Tipping (1972, p. 84), where their calculations were based on a Malaysian input-output table, we have put transport costs from Kenana to Port Sudan into the following proportions:

60% foreign exchange (e.g. fuel and spare parts)

30% unskilled labour and 10% skilled labour.

With regard to renewal and replacement cost, an attempt was made to separate the cost of equipment from that of buildings. In the end, we have settled for the assumption that maintenance and repairs of buildings is rather insignificant since about 78% of Kenana's work-force (i.e. semi-skilled and unskilled employees) live in grass cottages and huts. Renewal and replacement of equipment is obviously a heavy consumer of fuel and spare parts and a user of skilled manpower. The split between them was guessed as follows:

75% foreign exchange (imported materials)

25% skilled labour.

3) It should be noted finally with regard to Table 11 that foreign equity participants are assumed to repatriate their share of anticipated future profits. This is shown as item 2 in the table. Moreover, capital costs under item 7 will be reduced by 60% of S£ 330 million being the foreign equity contribution of the project in line with the discussion above.

4.1 Evaluation of Aggregate - Consumption Benefits

Our evaluation will be in successive stages of approximation. The first step is to assess the social benefits and social costs under the assumption that market prices adequately reflect the consumption benefits and cost involved. On this basis the benefits of Kenana consist of items (1), (4), (5) and (6) in Table 11 and the consumption costs (i.e. sacrifice of consumption possibilities) include items (2), (3), (8), (9), (10) and (12).

All these cost items represent payments for resources that could have been used elsewhere in the absence of the project.

The first approximation of net aggregate-consumption benefits in any given year will be measured as follows:

A = (1) + (4) + (5) + (6) - (2) - (3) - (7) - (8) - (9)

- (10) - (12) + 0.6 (S£ 330 m) (1)

To proceed to the second approximation, we have to relax our assumption of prices truly reflecting scarcity. We introduce market distortions which necessitate the adjustment of the market prices of specific resources wherever these prices do not represent the real contribution of the resources to our objective of aggregate-consumption benefits. Since imperfect market prices can occur in a number of ways, for simplicity and clarity we shall assume that all resources other than foreign exchange, skilled labour and unskilled labour are correctly priced by the actual market prices.

With regard to foreign exchange, Sudan has had a two-tier exchange rate system since June 1978 to encourage inflow of foreign exchange remittances. Arrears of unpaid commercial debt amounted to $ 3.6 billion and since the summer of 1978, the development programme has been slowed down due to balance of payment problems. Sudan has exercised strict quantitative import controls and export subsidies to maintain the dollar value of its pound. As the Sudanese pound is clearly overvalued we apply a positive foreign exchange premium which will be represented by . The opportunity cost of foreign exchange relative to the official exchange rate of the pound can be denoted by (1 + ).

Skilled labour is fully employed in the Sudan. However, the fact that 1988 alone witnessed a number of strikes over payment disputes by Engineers, Doctors, Accountants, University Teachers and Agriculturalists Trade Unions may point to a wide recognition of underpayment. Therefore, we assume that the marginal skilled worker contributes more to aggregate-consumption benefits than the salary he/she commands. The social premium on the market wage of skilled labour can be denoted by x.

Finally we come to the case of unskilled labour which is assumed to be surplus in the Sudan. Obviously the case of surplus unskilled labour is opposite to the two previous cases. The social premium on unskilled surplus labour, which is negative, can be symbolised by and the opportunity cost -- relative to the market wage rate -- will be denoted by (1 + ).

Before we proceed to the formula of the second approximation we have to make one more correction. Foreign exchange in our case falls into materials and salaries to foreign personnel as part of capital costs. It is sensible to assume that foreigners will consume part of their salaries in the Sudan and repatriate the rest. For the part consumed in the Sudan there is no need for correction, since the dollar value of this part is converted at the official exchange rates and thus Sudan does not lose the extra value of the foreign exchange. We shall denote the portion of the salary consumed in the Sudan by . This parameter is between zero and one and the repatriated portion is (1 - ).

Assuming that all the correct factors (opportunity cost premiums) will remain constant throughout, we can have our second approximation as follows:

B = A + F + L + xW (2)

where

F = (1) - (2) - (3) + (4) + (5.a) - (7.a) (1-) (7.b) - (8.a)- (9.a) - (10.a) - (12.a) + 0.6 (S£ 330 m) (2a)

L = -(7e) - (8.d) - (9.b) (2b)

W = -(7.d) - (8.c) - (9.c) - (10.b) (2c)

The term F corrects A for the opportunity cost of foreign exchange by multiplying benefits and costs component of foreign exchange by the positive foreign exchange premium . The term corrects A for the opportunity cost of unskilled surplus labour, L, by the negative labour premium, ; and the last terms, xW, corrects A for skilled labour with the positive premium x.

We now come to the third and final approximation which is to account for the fact that in developing countries development is normally given a priority to the extent that the value of funds devoted to investment exceeds the social value of the same amount of funds devoted to consumption. This phenomenon is clear in the Sudan. The fact that Sudan has accepted in principle the terms of assistance dictated by the IMF which includes severe cuts in public expenditure, lifting of subsidies on petrol, sugar and some other food items and devaluation of the Sudanese pound may all point to the inability of the government to bring about optimal savings deemed necessary for development (see also Wynn, 1980; World Bank, 1990). Moreover, it can be assumed that aggregate savings in the Sudan will remain suboptimal in the near future.

This requires the adjusted net benefits in equation (2) above to be `allocated' to the groups that will gain or loose from them. For this purpose, three sectors are assumed, the government (GO), the private sector (V) and unskilled labour (L) for which the net effects have to be specified.

The `government' (including El-Nilein Bank and the Sudan Development Corporation) controls the foreign exchange market so that the net benefits of the project may be allocated to the three sectors as follows:

B = B^GO + B^L + B^V (3)

where

B^GO = A + F - (6) (3a)

B^L = [(7.e) + (8.d) + (9.b] + (6) (3b)

B^V = -xW (3c)

The important points here are that unskilled labour gets two benefits. First its `excess' payment over marginal product (the proportion ) and the further benefit of the investment in housing (6). The private sector, however, loses the 'surplus' it has previously enjoyed from the underpayment (by proportion x) of skilled labour drawn to the project.

B^GO, B^L and B^v must now be amended to adjust for the proportions which each group will consume or save (since savings are assumed to be at a premium).

For instance, if the marginal savings propensity of unskilled workers is S_L then the social value of net consumption benefits flowing to unskilled labour is as follows:

C^L = [(1-S_L) + S_LP ^INV]B_L (4)

where P^INV = the shadow price of investment.

The same equation applies if S_GO and S_V are the marginal savings propensities of (GO) and (V) respectively.

Based on the preceding information we may write the final approximation as follows:

C = C^GO + C^L + C^V (5)

4.2 Benefits to the Sudan

In order to throw light on the question of how profitable Kenana could be for the country, we have to establish numerical values for the various national parameters which we have been specifying throughout the course of the discussion in the preceding section. Obviously the parameters we have been discussing are essential for the assessment of the desirability of proposed public sector projects in the light of the national interests. Consequently, in estimating them, information that pertains to the state of the economy and the policies of the government are essential.

Naturally, the required information is of both factual and normative nature. Presumably this is the main reason why the two manuals presented by UNIDO in 1972 and that by Little and Mirrlees in 1974 for evaluation of industrial projects in developing countries suggested some specific roles for a central office for project appraisal and planning as part of the Ministry of Economy. Unfortunately this office has no counterpart in the Sudan. Accordingly, we have to resort to some reasonable assumptions about these parameters based on the available information on the country and the economy. On this basis, we put forward the set of assumptions in numerical values. The range of values given is believed to be appropriate to the Sudanese economy. A few comments are in order.

In view of the fact that Kenana is making use of land which was bare savannah a few years ago and where "... people used to live the same nomadic lives which their forefathers had followed for centuries" (KSC, 1980, p. 33), it may not be unreasonable to regard the opportunity cost of unskilled labour as equal to zero. A further support to this point of view can be gained from the fact that Kenana province is one of the relatively densely populated regions in the country; thus it is not expected that local production will be adversely affected by the project. On the contrary, it is almost certain that the nomads of the area will find a lucrative market in the township of Kenana for their milk, meat, etc. In any case it is to be noticed that unskilled labour costs represent a surprisingly small proportion of the total costs (7% of capital costs and 15% of the total costs of production).

In contrast, the opportunity cost of skilled labourers is estimated at between one-and-a-half and twice the market wage rate. We have already mentioned that a number of strikes took place.

In discounting the social costs and benefits we adopted the fairly safe assumption of the Social Discount Rate, i, being 0.08 and 0.13. With regard to the marginal propensity to save, S, and the marginal rate of return to investment, q, we followed the rule of upper and lower limits. Namely q > i and i > Sq which is consistent with the UNIDO (1972) procedures. Within these limits and in the absence of any information, we regard our set of values for S and q as a reasonable guesses. Having numerical values for i, S and q, the shadow price of investment is calculated by the aid of the following formula:

P^INV = (1-S)q (6)

i-Sq

As for propensities to save, it is assumed that while the government may devote its savings to investment, the unskilled labourers will consume all their wages. Also, it is tentatively assumed that the private sector will consume between 40% and 50% of its profit.

As for the foreign exchange premium, the formula for estimating the Shadow Exchange Rate (SER) suggested by the UNIDO⁶(p. 215 onwards) is considered by many including the authors themselves as possibly, but certainly not necessarily, correct, especially if the traded goods in question are intermediate products subject to varying tariffs or quotas or both. In the absence of any reliable operational procedures on the issue we have guessed as 0.2. This is a very conservative guess suggesting a 20% implicit premium on foreign exchange. Our narrative so far indicates that this is probably a considerable underestimate. For foreign personnel we have assumed that they will consume between 40% to 60% of their salaries in the Sudan. We have guessed these last two percentages with the high rate of inflation in mind.

All the parameters listed in Table 12 are assumed to be constant. Consequently the time flows shown in Table 11 are converted into their present value in the base year (Table 12) since that is equivalent to making separate calculations for each year of the project. Having the values in Tables 12 and 13, we set in the equations given earlier. The results are given in Table 14.

Table 12: Values of National Parameters: Assumptions Set

(1) Foreign Exchange Premium	= 0,2
(2) Domestic Skilled Labour Premium	x = 0.5
(3) Unskilled Labour Premium	= 1,0
(4) Marginal Rate of Return to Investment	q = 0,20
(5) Marginal Rate of Savings	S = 0,25
(6) Social Discount Rates	i = 0,08; 0,1; 0,13
(7) Shadow Price of Investment	PINV = 5,0; 3,0; 1.8
(8) Marginal Propensities to Save (MPS): (a) Government (b) Unskilled Labour (c) Private Sector	SGO = 1,0 SL = 0,0 SV = 0,6
(9) Proportion of Foreign Personnel Salary Spent in the Sudan	= 0.6

Table 13: Present Value in Year 1 of Items in Table 12

(in million S£)

Item	Social Rate of Discount
	8%	10%	13%
1. Output (Foreign Exchange)	801.40	620.80	439.1
2. Repatriated Profit (Foreign Exchange)	123.20	97.45	70.3
3. Repatriated Equity (Foreign Exchange)	17.60	9.97	70.3
4. Scrap Value (Foreign Exchange)	7.70	9.97	4.3
5. Reclaimed Working Capital 5.a Foreign Exchange 5.b Domestic Materials	2.10 1.50 0.6	1.20 0.83 0.37	0.52 0.4 0.12
6. Welfare & Housing (Domestic Materials )	53.4	44.2	34.3
7. Capital Costs 7.a Foreign Exchange 7.b Foreign Personnel 7.c Domestic Materials 7.d Skilled Labour 7.e Unskilled Labour	320.5 145.2 80.1 55.3 23.15 16.07	302.7 137.12 75.67 52.06 21.79 15.13	276.1 125.07 69.58 47.49 19.88 14.08
8. Costs of Production 8.a Foreign Exchange 8.b Domestic Materials 8.c Skilled Labour 8.d Unskilled Labour	363.71 56.37 82.20 57.10 167.31	280.75 43.52 63.45 44.08 129.14	199.33 30.89 45.05 31.29 91.69
9. Transport Cost to Port Sudan 9.a Foreign Exchange 9.b Unskilled Labour 9.c Skilled Labour	49.7 29.82 14.91 4.97	39.34 23.60 11.80 3.93	27.00 16.2 8.1 2.7
10. Renewal & Replacement Cost 10.a Foreign Exchange 10.b Skilled Labour	42.3 31.72 10.57	32.74 24.55 8.18	26 19.5 6.5
11. Cash Transfers 11.a Income Tax 11.b Potential Subsidy	94.93 184.5	64.7 142.06	32.7 99.06
12. Working Capital 12.a Foreign Exchange 12.b Domestic Materials	15.25 10.89 4.35	13.82 9.93 3.89	11.99 8.68 3.31

Table 14: Present Value of Net Benefits in the Base Year Under Assumptions in Table 12

(in million S£)

Item	Equation	Social Rate of Discount
		8% PINV=5	10% PINV=3	13% PINV=1.8
A	(10.1)	+130.34	+91.73	+58.80
	(10.2a)	+112.35	+89.50	+67.30
(n.b. < 0)	(10.2b)	+198.29	+156.07	+113.87
xW (n.b. x.0)	x(10.2c)	-47.89	-38.99	-30.18
B	(10.2)	+393.09	+298.31	+209.79
BGO	(10.3a)	+189.29	+137.03	+91.80
BL	(10.3b)	+251.69	+200.27	+148.17
BV	(10.3c)	-47.89	-38.99	-30.18
C	(10.5)	+1035.32	+525.58	+286.74

The final row of the table represents the effect of adjusting the net receipts in row 1 for foreign exchange premia (row 2), unskilled, and skilled labour scarcity indices (rows 3 and 4). These add to the total benefits in row 5. The allocation of these benefits to the three sectors is shown in rows 6, 7 and 8.

The final adjustment for each of these benefits is that for the investment premium attached to saving (e.g. equation (4)). Thus for the government, the value B^GO is multiplied by 5 (i.e. P^INV = 5), since all government benefits are assumed available for saving. The consumption equivalents so derived add to the final C in row 9. At each discount rate used the project appears from these calculations to be socially worthwhile.