The desirability of FDI, however, depends on its "net social benefit" accruing to the host country, the estimation of which is much more complicated. As in the case of social cost-benefit analysis of domestic investment projects, all inputs and outputs need to be estimated at their "economic" prices, which represent their "real" value to society (the so-called opportunity costs). This is important because domestic prices are often largely distorted. For example, any subsidies on input prices that make actual prices lower than their "economic" prices have to be deducted in estimating the benefit of FDIs. Likewise, when FDI produces for a domestic market that is highly protected by import tariffs, the value-added that it creates in terms of domestic currency is much higher than its true worth in terms of foreign exchange (that is, at international prices).

This is not to argue that FDI should not be given the benefit of any subsidized inputs or other incentives like tax breaks, but that these should be taken into account in assessing the benefit of FDI. There is a need to provide reasonable incentives to attract FDI. But, unless the benefit for the country is assessed in relation to "economic" prices, one cannot be sure that the proposed investment is genuinely profitable enough to result in a win-win situation.

Fortunately, not all FDIs require such detailed evaluation to see that the country is getting a fair deal. This would include FDIs that work within the market discipline, such as those in labour-intensive manufacturing industries catering to the export market or a highly competitive domestic market. The very fact that foreign investors are attracted to make such investments under market determined competitive prices is itself an indication of a win-win situation.

On the other hand, the economic benefit from FDIs that operate in a non-competitive domestic market is not self evident. These include the FDIs involving "administered" prices, such as under purchase or sale contracts with the government. Some of these FDIs, such as in electricity generation, may be potentially highly desirable because of their being in strategically important sectors; but their benefit depends on the terms of agreement regarding prices and other incentives. International competitive bidding can be a mechanism to ensure that an "exploitative" arrangement is not imposed on the country; but this is easier said than done.

Even a competitive bidding process does not do away with the need to determine whether the investment is at all genuinely profitable so as to be able to yield a win-win result. The problem is compounded by the fact that, for a country like Bangladesh having a poor international country-risk rating, the perceived risks of investment will be factored in while prospective foreign investors offer their bid. Also, in a perceived risky environment, foreign investors may have a too strong preference for getting their profits sooner than later, which may result in an overexploitation of exhaustible resources like minerals.

Another problem may arise from the fact that FDIs in the "non-tradable" sectors, such as in electricity generation, telecommunications or other infrastructure provision, do not directly contribute to earning or saving foreign exchange. True that these investments provide funds for making strategically important investments that may greatly contribute to overall economic growth. But the fact that there is no immediate involvement of the country's resources may lead to the neglect of the problem created by too large future obligations for profit repatriation. In going for such FDIs, the implications for the long-term viability of the balance of payments should be taken into account.

In promoting the case for FDIs, some indirect benefits are often cited. While these benefits may be quite important, their nature needs to be properly understood in evaluating FDIs, particularly since these benefits are largely of an intangible nature. The following are three most commonly cited benefits in the context of Bangladesh:

First, beyond the direct economic impact, the investments may yield indirect benefits by creating jobs and promoting economic activities in other related sectors through the so-called forward and backward linkages that is, by purchasing inputs (or outsourcing) and/or producing goods that are inputs for other activities. Whether creating demand for other sectors is an additional benefit from FDI depends on the nature of production constraints in those sectors. In a capital-scarce and labour-surplus economy like Bangladesh, the expansion of most production activities is not constrained by demand deficiency, but by a lack of investment to create additional production capacity. However, there are many low-productivity labour-intensive activities, mostly services, which can easily respond to increased demand without much investment. The employment and incomes generated in this later way can be considered as a genuine spill-over benefit from the inflow of FDIs. As regards forward linkages, the benefit will depend on whether the FDI is for producing essential inputs for other activities that would not be otherwise produced because of lack of technical know-how or would be available only at a higher price, say, from imports.

Second, FDIs may save foreign exchange by producing tradable items -- exports or import-substitutes. This will be an additional benefit depending on whether there is an effective balance of payments problem that constrains economic growth. In such a case, a taka worth of output that can be internationally traded will carry a premium over a taka worth of non-traded output hence the likely benefit arising from FDI's contribution to the balance of payments. Recent development experience in Bangladesh suggests that the balance of payments can indeed be a problem at times of accelerating economic growth.

Third, the claim that an initial FDI may promote investor confidence and reduce the perceived country risks by future prospective foreign investors -- hence the case for providing it some extra incentives. While this may be correct, an "unfair" deal may create future resentments and other problems that will do damage rather than promote the investment environment. The best way to improve the country-risk factor is to address the underlying causes rather than offer too generous concessions to foreign investors.

Tata's investment proposal is a complex one with several components. Of these components, the proposed urea plant is an entirely separate project with no link with the rest of the investment proposal; as such, its merit is better judged separately. This aspect is mostly blurred in the way the overall impact of the project on the Bangladesh economy is discussed in the report of the Economic Intelligence Unit (EIU) as well as in much of the documentation produced by Tata. The other part of the investment project is an integrated one involving steel production, power generation and coal mining. Even for this integrated project, the possibility of generating power by gas and thus leaving out the coal mining component may be kept open as an alternative, given the many unresolved issues regarding coal mining.

About the EIU report, it is mainly concerned with the likely economy-wide or macroeconomic impact of Tata's investments; it has very little to say about the likely net social benefit of the project in terms of a cost-benefit analysis as discussed above. It does not deal with the key issues nor make any attempt to quantify the key variables that will directly determine the part of the project benefit accruing to Bangladesh - such as the pricing of gas and the future streams of wage incomes and tax yields generated from the project. Instead, the EIU report mainly dwells upon what may be called indirect project benefits, such as those arising from the impact on the balance of payments and from any backward and forward linkages. As noted above, these indirect benefits may be additional considerations in deciding on the desirability of the project, but these are not the basic ingredients for estimating the net benefit of the project. Even in estimating these indirect benefits, the EIU report does not often take into account some of the common caveats of such estimates, some of which have been mentioned earlier. We come back to these issues towards the end of this report, but for the time being, let us concentrate on the basics.

Steel production is highly energy-intensive, while urea production is a non-energy use of gas. According to the data available from Tata, the urea plant will produce $127 worth of value-added annually by using 0.04 tcf of gas (total of 0.8 tcf in 20 years). The value added from steel production will be $495 million annually against the annual use of 0.028 tcf of gas (0.7 tcf in 25 years.). If captive power generation for the steel mill is to use gas instead of coal, the total annual use of gas for steel production will be 0.033 tcf (0.825 tcf in 25 years). Including power generation and coal production, Tata's total project, when in full operation, is estimated to generate $972 million value-added annually.

The critical importance of the pricing of gas in determining Bangladesh's benefit from Tata's investment can be readily seen from some aggregate figures. If the price of gas were to be reduced by $1 (which will be a subsidy if the real "economic" price of gas is taken as the benchmark), this will reduce Petrobangla's sale proceeds from this project by $68 million annually (by $83 billion if captive power were to be generated from gas). Against this, the direct benefit will include the part of value-added accruing to Bangladesh in terms of wages and salaries of local employees and the tax revenue of the government. The salaries are estimated to be $20-$30 annually (the true social benefit may be half of this if we assume that the "shadow" wage rates representing the opportunity cost of labour in Bangladesh is half the market wage rate). This is a very low proportion of the estimated total value-added -- about 2 to 3 percent. This is an implication of the fact that although the production processes in the various components of the project are highly resource-intensive, their labour-intensity is extremely low.

Tata estimates that the annual taxes payable after the tax holiday period will be approximately $120 million per annum. This would include mainly corporate tax on profit at the prevailing rate of 40 percent, and also royalties for coal mining and other indirect taxes (presumably net of various tax exemptions sought by Tata as part of the incentive package). Applying a time discount rate of, say, 8 percent annually, the above tax revenue can be estimated as equivalent annual revenue for a 20-year project life. This works out to be $38 million annually if we assume that the tax holiday is for the initial 10 years, but the amount rises to $63.5 annually for a six-year tax holiday period.

The above aggregate figures suggest some broad features of the benefit to be derived from Tata's investments:

(a) If we consider the net benefit from the direct investment impact, this will be quite sensitive to the pricing of gas. While some amount of subsidy in gas sales (compared to the "economic" price of gas, see later discussion) may be accommodated, a large subsidy may easily make the net direct benefit negative.

(b) The period of tax holiday can make a large difference to the net benefit, since tax revenues seem to be the main direct benefit from the investments.

(c) Tata's projected profits from the investments seem to be large enough, particularly for the steel-power complex, so as to provide ample scope for bargaining in order to arrive at a fair win-win deal. There also thus does not seem to be any strong case for allowing tax-breaks or other incentives beyond what are allowed under the existing structure of incentives for such investments. Special incentives beyond the existing rules also create precedence for giving such incentives to other prospective investors.

The last point is a bit problematic because the information that is available from Tata's documentation and the EIU report is rather insufficient and, in some respects, also appear inconsistent. To make decisions regarding the separate components of the project (the urea project in particular), component-wise detailed information is necessary. There seems to be also very large discrepancies between the estimates of Tata's annual profits as implied by the annual tax revenue to be paid after the tax holiday period and as can be directly estimated from the annual value-added net of taxes and salaries (the later seems to be much larger). To help evaluate the investment proposal component by component and, thereby, negotiate the terms of contract, Tata should be encouraged to share more detailed information for each component.

Wahiduddin Mahmud is a renowned economist. Part II of this report will appear tomorrow.