State divestiture as a policy instrument in developing countries

World Development, Vol. 16, No. 12, pp. 1465-1479,1988. Printed in Great Britain.

0305-750x/88 $3.00 + 0.00 0 1988 Pergamon Press plc

State Divestiture as a Policy Instrument in Developing Countries TONY KILLICK and SIMON COMMANDER Overseas Development Institute, London Summary. - Reducing the size and role of the public sector in developing countries and emphasizing prices and markets as allocative mechanisms has led to increased privatization measures, of which divestiture has been the most common variant. The paper examines this experience and indicates that ownership transfer is neither a necessary nor sufficient condition for achieving more efficient use of resources. In the design of divestiture, the weighting given to enhancing freedom of entry and intensity of competition remains critical. At the same time, the paper notes the existence of multiple constraints - including profitability and capital market limitations that continue to impede the implementation of divestiture measures. The conditions for divestiture to be both appropriate and successful are shown to be restrictive.

2. BACKGROUND

1. INTRODUCTION The recent drift away from Keynesian and more interventionist approaches to economic management in the developed economies has had as one of its components the explicit intention of reducing the size and role of the public sector. The rehabilitation of prices and markets as allocative mechanisms has favored policies designed to achieve deregulation of markets, to loosen restrictive practices and, most dramatically, to effect a transfer of ownership and management away from the public sector. The latter policy option - divestiture - while only a part of a broader process of privatization in some cases, has nevertheless emerged as a particularly stark illustration of the reversal in policy preferences that has occurred in the 1980s. Moreover, despite the fact that the rhetoric has outstripped the reality, divestiture now figures increasingly prominently on the agendas of a growing number of developing countries. This paper examines the experience with divestiture as a policy option in less developed countries (LDCs), and is divided into four main sections. The following section outlines the context in which privatization issues have come to the fore. The third section is concerned with theoretical issues relating to the instrumental efficiency of divestiture while the fourth section considers the constraints on divestiture. A concluding section follows which attempts to draw some general lessons.

The recent and growing emphasis on privatization in both developed and developing countries can be understood primarily as a response to the earlier growth in the public sector, the expansion of which has commonly been argued to have been excessive and premature. Yet measurement of the size of the public sector has proven difficult. Using public expenditure as a share of GDP as the measure, in Britain public expenditure comprised roughly 44% of GDP; by the early 1980s in West Germany the figure was around 42%; and in the United States federal government spending amounted to over 22%. Other indicators likewise point to significant shares of fixed investment and employment being accounted for by the public sector. This has been especially pronounced outside the developed economies and one estimate has suggested that around 44% of non-agricultural employment in LDCs was in the public sector (Heller and Tait, 1983). Figures on the share of national income generated by public enterprises remain more elusive. This derives in part from data constraints and a lack of transparency in national accounts. Nevertheless, at a global level, public enterprises have been estimated to account for around 10% of GDP on average in both developed and de-

*This article earlier paper 1465

has been developed from the authors’ in Cook and Kirkpatrick (1988).

1466

WORLD

DEVELOPMENT

veloping countries. It has further been estimated, albeit on the basis of a limited sample, that in 1980 state-owned enterprises were typically producing lO-20% of GDP and accounting for 2@60% of aggregate investment in LDCs (Berg, 1987). The growth in the public sector, however measured, has undoubtedly been most rapid outside of the Organization for Economic Cooperation and Development (OECD) countries. In Ghana, for example, the parastatal Cocoa Marketing Board alone had swollen to around 20% of formal sector employment by the early 1980s while in Egypt possibly as much as a third of the labor force is currently employed by the state.’ World Bank estimates point to between a quarter and a third of total modern sector employment being accounted for by the state in Africa and Asia (World Bank, 1986b, p. 21). In general, the growth in the public sector can be traced to the period stretching from the late 1950s to the mid-1970s. This process was notable across a wide range of economies - in Korea, for example - and was not limited simply to those countries which explicitly espoused importsubstituting policies or else sought, as in many Sub-Saharan African countries, to build a modern state through a rapid consolidation of the administrative system and direct state intervention in production. Justification for the proliferation of public enterprises and enhanced public expenditure was derived not only from arguments regarding market failure and the need for economic diversification and control over natural monopolies or strategic concerns, but also on overtly political grounds. This could be expressed in terms of the equity or social benefits that could be derived from public activity, but - and this was particularly the case in much of Sub-Saharan Africa, where the post-colonial state was marked by its fragility.the growth of a state sector and its use as a source of patronage and employment was viewed as the glue that held the structure together (Bates, 1981). Exclusively economic arguments were thus rarely invoked in justifying the expansion of the public sector. Since the late 1970s however, the combination of terms of trade and financing shocks, coupled with domestic imbalances, have compromised governments’ ability to sustain growing levels of public expenditure. Public sector deficits grew averaging 6-7% of GNP for both low- and middle-income economies by 1983/84 - fuelhng domestic inflation rates, as well as, in some contexts, crowding out private sector activity. Such short-run effects were often cited as corollary to dynamic consequences on the the adverse,

supply-side that resulted from excessive expansion of the public sector. This was held to be the case particularly when public sector monopolies, such as the West African Marketing Boards, were associated with high levels of implicit taxation of producers2 The primary effect of the downturn in economic activity since the late 1970s has thus been to place pressure on the public finances of many LDCs and, particularly in Sub-Saharan Africa, to force governments into lowering public expenditure levels. This has been most pronounced where governments have had recourse to International Monetary Fund (IMF) and World Bank stabilization and adjustment loans. Indicative of the growth in adjustment lending is the fact that while in 1979/80 structural and sectoral adjustment lending accounted for under 0.6% of total World Bank loans, between 197986 they accounted for over 6% of total lending and had risen to over 17.5% in fiscal year 1986 (World Bank, 1986a). Use of such lending instruments has been accompanied by a growth in policy conditionality. Consequently, restraint of central government current expenditures featured prominently in nearly 85% of cases where IMF adjustment programs were adopted in Sub-Saharan Africa between 1980 and 1984 (Zulu and Nsouli, 1985). Likewise, World Bank structural adjustment lending has normally placed heavy emphasis on improving resource use efficiency, and that of public enterprises in particular. Budgetary constraints have been a key factor in engendering a widespread review of the size, role and efficiency of public sector institutions. Yet it has not only been public finance considerations that have driven this review. Policy shifts in the developed economies have also played an important part. Although the leading OECD economies that have explicitly attempted to reshape their macro and microeconomic pohties - Britain, West Germany and the United States are the most prominent examples - have not actually achieved very much in terms of reducing the share ‘of public expenditure in GDP, there has nevertheless been a significant shift in emphasis away from the public sector. Possibly the most striking manifestation of this has been the privatization that has become a major plank in economic policy. Such measures have ranged from deregulation of markets to the more dramatic transfer of ownership from the public to the private sector. In Britain where divestiture by the state has been a particularly prominent feature of the privatization process the proportion of households owning shares has apparently risen from under 10% in 1979 to about 40% in 1987. Divestiture has been promoted on the grounds

STATE

DIVESTITURE

IN DEVELOPING

that the earlier preference for public enterprises entailed too high a cost in inflation, lower output and productivity than a narrower emphasis on efficiency considerations would warrant. In this light, management questions attract particular importance (Matthews and Minford, 1987). Changes in economic policy preferences in the developed countries have had a direct, albeit lagged, impact on developing countries. The strong advocacy of the United States government has been particularly important and this has percolated not only through US Agency for International Development (USAID) lending policies but also through multilateral and other bilateral donors. A telegram issued by US Secretary of State George Schultz in 1985 to USAID missions spells out the major issues which USAID officials should raise in the course of policy discussions with LDC governments.3 Take first the general statements about the desirable policy orientation of aid recipients: Policy dialogue should be used to encourage LDCs to follow free market principles and to move away from government intervention in the economy. This allows the market to determine how economic resources are most productively allocated and how benefits should be distributed. To the maximum extent practical governments should rely on the market mechanism -on private enterprise and market forces - as the principal determinants of economic decisions. Now

take

what

it has to say on parastatals:

Parastatals are generally an inefficient way of doing business. They usually require subsidies and/or preferential treatment. Government should scrutinise the need for and activities of such entities. In most cases, public sector firms should be privatised. Where this is not practical, such firms should in so far as possible be subject to market forces and the discipline of the marketplace. (Emphasis added) USAID subsequently handed down a directive requiring most of its African field missions to be involved in “an average of at least two privatisation activities” by the end of fiscal year 1987.4 While this is a particularly blunt statement of preference, let alone a prescription for policy in developing countries, privatization has featured less prominently and more pragmatically in the lending conditionality of multilateral agencies (Mosley, 1988). Privatization was only an explicit condition in three (Senegal, Jamaica and Pakistan) out of 10 Structural Adjustment Loans recently evaluated by the World Bank (1986~) but, it should be noted that emphasis on public expenditure reduction has been widely associated with a transfer of functions away from the public sector, if not directly with divestiture. In those

1467

COUNTRIES

cases where divestiture has occurred, it appears that closure of public sector enterprises has been a more common outcome than transfer to private sector owners and management. While much of the recent preference for reducing the size and role of the public sector can be traced back to the ideological and political imperatives of the developed country governments, disillusionment with the state and its instrumentalities has ranged beyond conservative ideologues. The sheer weight of evidence pointing to low efficiency and returns to investment among public sector enterprises in developing countries has sponsored a growing re-appraisal within those countries. Such a process has also begun to occur in a number of centrally-managed economies - China and Hungary are perhaps the most marked examples as also in Africa, where poor public enterprise performance has proven to be a chronic and widespread phenomenon. In a growing number of cases - Senegal and Togo are the most prominent - divestiture has emerged as a key feature in the adjustment programs being implemented by governments. Even so, it should be emphasized that the connections between political philosophy and attitudes towards ownership or the merits of the market mechanism remain intensely close. Privatization has emerged in close association with the marked conservative shift in political opinion in Europe and North America.” Whatever the politics of the matter, however, we turn now to explore what light economics can throw on the efficacy of divestiture as a policy instrument.

3. DIVESTITURE AND ECONOMIC EFFICIENCY (a) Developmental

efficiency

Since we are dealing with divestiture as it relates to developing countries, the most important issue is what effect a change from public to private ownership might have on the pace of economic development. Unfortunately, standard microeconomic theory has little to say on this subject. It is a widely recognized weakness of the dominant neoclassical school, with its focus on maximization and the establishment of equilibrium, both that it actually has little to say about the firm as an entity and about dynamic processes.6 As Baumol (1968, p. 68) has put it: “In all these [maximising models] automaton maximisers the businessmen are and automaton maximisers they remain. And this shows why our

1468

WORLD

DEVELOPMENT

body of theory . . offers us no promise of being able to deal effectively with the description and analysis of the entrepreneurial function.” Thus, although the pace of technical progress and innovation are of central importance for economic development, models of competitive equilibrium cannot handle this subject in a satisfactory way. The normal recourse to comparative statistics as a way of handling time within the neoclassical framework is a methodology of crippling limitations for the study of development, with its lack of concern with processes. Writing outside the neoclassical framework, Schumpeter (1934) has been the most influential writer in this area. He saw competition through innovation as deviation from routine behavior, not readily amenable to marginal analysis, and yet of far greater long-term importance than competition through price. He regarded (private) monopolies as more likely to innovate than competitive firms both because they could use their monopoly profits to finance research and development (R&D) - an activity not so attractive to potential outside sources of capital - and because monopolies would have greater assurance of being able to use their market power to reap the benefits of their innovations. If his view were borne out in practice, it would tend to undermine the pro-market thrust underlying the privatization movement. Since. his original insights, however, limited progress has been made in bringing this type of hypothesizing within mainstream theory, although there has been a good deal of empirical testing. The evidence unfortunately does not take us much further because it indicates that market structures intermediate between highly competitive and monopolistic have the most intense levels of innovation, and that R&D is more efficient in medium- than in large-sized firms.’ As Kamien and Schwartz (1982, p. 217) have put it, with regard to the influence of monopoly on technical progress, “Our present knowledge is inadequate to allow for sweeping generalisations or universal formulas.” In any case, Schumpeter was of course writing of privately-owned firms and it is difficult to derive from his-or any other-writings any clear inferences for the relative developmental merits of public and private enterprises. A potentially more promising line of approach is to examine the actual record but that too is a neglected area. The example of South Korea - and before it of Japan - is often cited as a favorable example of public enterprises playing a key role in the earlier phases of industrialization, to be privatized later (Jones, 1975). On the other hand, a recent survey of public enterprise performance in Sub-Saharan

Africa came to a very (Nellis, 1986, p. ix):

unfavorable

conclusion

PE earnings

are generally low; many run losses; often these losses are of a large magnitude. Far from contributing to government revenues, African PEs have regularly become a heavy burden on already strained budgets. Few PEs generate revenue sufficient to cover operating costs, depreciation and financial charges; a good percentage do not cover operating costs alone. In many instances where PEs are classed as profitable, closer examination reveals distorted prices, direct subsidies, hidden transfers, preferential interest rates and a host of other elements which if properly accounted for-would reduce the paper profits of the PE in question. The conclusion is that African PEs present a depressing picture of inefficiency, losses, budgetary burdens, poor products and services, and minimal accomplishment of the non-commercial objectives so frequently used to excuse their poor performance. Though every African country has one or more PEs which perform well by the most stringent of standards, on the whole PE sectors are not fulfilling the goals set for them by African planners and leaders. In a review of the performance of industrial public enterprises in Africa, one of the present writers similarly came to a strongly negative overall conclusion about their performance and cautioned against the creation of such enterprises as a means of stimulating industrialization (Killick, 1983). Overall, if the developmental dimension is viewed largely in terms of the reinvestment levels and innovatory spective dynamism of the two, we suspect the evidence would be in favor of private ownership, on the grounds that returns to capital (and hence the size of the reinvestible surplus) are usually lower among public enterprises, and that they have a generally poor reputation for product innovation and diversification. However, we have already mentioned exceptions to this conclusion and the chief point to make is that a definitive study of this subject has yet to be published, so we remain in much ignorance.

(b) Allocative efficiency The privatization movement, as we have seen, is part of a more comprehensive reassertion of the merits of market solutions. Private ownership, it is claimed, will lead to a more efficient, higher productivity use of resources. Some would also claim that efficiency considerations cannot be divorced from the system of rules and rights the constitutional contract - within which exchange takes place. When the voluntary nature of exchange between individuals is valued for its

STATE

DIVESTITURE

IN DEVELOPING

own sake, market-like solutions, they would (Buchanan, argue, will always be preferable 1986). The standard neoclassical position, derived from the theory of perfect competition, argues that the market mechanism tends in the direction of producing a Pareto-optimal result, where resources are so distributed between alternative uses that no changes could be made that would not make someone worse off. It is, of course, accepted that economic conditions in developing countries are far removed from the demanding assumptions upon which the perfect competition model is built but it is nonetheless asserted that market-determined resource allocations will result in higher-productivity uses than non-market forms. The normative pro-market prescriptions of standard neoclassical theory are, however, vulnerable at a number of points. One well known difficulty is the existence of increasing returns to scale, giving rise to natural monopolies and allocations of resources which deviate from the perfectly competitive optimum. Since increasing returns to scale are a common feature of modern industry this is an important consideration and it is precisely the existence of such natural monopolies that is used as an argument for public ownership, to capture the benefits of scale economies for the public at large rather than for the owners of monopoly capital. To some extent, the theory of “contestable markets” is a response to this problem and draws attention to the special importance of freedom of entry and exit.’ Under this theory, the possibility of entry by a competitor acts as a constraint upon the profit-maximizing behavior of the natural monopoly. When entry and exit are literally free (which implies the absence of sunk costs in the industry) hit-and-run competition is feasible and the constraint on the abuse of monopoly power is powerful. The implications of this development of neoclassical theory for the privatization debate are unclear, however. In many cases, sunk costs and other entry and exit costs will be large, with limited real threat from potential competitors, especially of the hit-and-run kind. On the other hand, it is not uncommon for public enterprises to be given a legal monopoly, with private firms forbidden from operating in the industries in question. When this is the case, divestiture can increase contestability and thus create a new restraint on monopoly power. In any case, some writers would question whether the existence of natural monopoly power arising from increasing returns constitutes a valid case for public production of the good or service in question, advocating instead the adop-

COUNTRIES

1469

tion of franchising techniques, as has been done in France and elsewhere. Under this solution, open bidding is invited from private firms for the delivery of an output of defined volume and quality, with the franchise given to the lowest bidder.” A far larger source of difficulty for standard micro theory is its inability to cope with important aspects of reality in actual market structures. Critics question whether optimization is a useful postulate in understanding the workings of firms and market processes. Many follow Simon (1959, 1965) in suggesting that firms have neither the information nor computational ability to optimize, that theirs is a “bounded” rationality, and that “satisficing” may be more useful than optimization as a key to understanding enterprise behavior. Given this array of difficulties, it is not clear how much remains of the normative conclusions of standard neoclassical theory. Most economists, including the present writers, believe that in general allocative efficiency is likely to be furthered by more rather than less competition. But the weaknesses in the theoretical basis of this prescription are such as to allow plenty of exceptions and to indicate the need for caution in making policy recommendations. In any case, a basic difficulty of arguments for divestiture on the grounds of an alleged superior allocative efficiency is that, while mainstream microeconomic theory does point to the preferability of competition, it is actually silent on the ownership issue. Thus, advocates of privatization have been accused of confusing private ownership with competition (Shackleton, 1984; Kay and Thompson, 1986). There is, of course, no necessary connection between the two. Public enterprises can be required to compete on equal terms with private concerns. Private enterprises often possess much monopoly and/or monopsony power, a condition particularly likely to be prevalent in developing countries with their small internal markets and generous protection from import competition (Killick, 1981, Chap. 10). The extent of freedom of entry in the markets in question will be a key issue (Yarrow, 1986, p. 344). One of the great difficulties for divestiture is to avoid simply converting public monopolies into private ones. This is not to say that such a conversion would be without significance. It is possible that there would be greater freedom of entry in the case of private ownership giving the markets in question more of the character of contestable markets with smaller adverse efficiency and welfare effects. It is also possible to argue that, in contrast to public monopolies, their private counter-

1470

WORLD

DEVELOPMENT

parts are subject to the discipline of financial markets, including the possibility of takeover bids. This would give them a greater incentive to keep down production costs - but not to pass on the benefits to consumers (Tomlinson, 1986). Transferring monopoly powers from a public to a private corporation will also tend to result in a shift in value-added from owners of labor to owners of capital, but with the net effect on consumers and on total factor productivity indeterminate. Mainstream neoclassical price theory thus apparently fails to provide conclusive reasons for favoring private over public enterprise. Two caveats should be entered, however. First, it should be possible to design a privatization program in ways that do increase the probability or extent of competition. Many public enterprises in developing countries are either statutory monopolies or are given advantages over private competitors by means of open or indirect subsidies, or in other ways. It is still rather rare for a government to allow a state concern to go out of business because it cannot compete. In this sense, standard price theory is not quite as silent on the ownership issue as is sometimes suggested. Second, moreover, there are certain outgrowths from the standard theory which take a more positive view of the connections between ownership and efficiency. We.particularly have in mind here the “property rights” school which is principally concerned with the relationships between ownership rights, incentives and economic efficiency.“’ This deviates from mainstream theory by rejecting profit maximization as the guiding behavioral postulate, focusing instead on the actions of managers who are assumed to pursue their own self-interests. In the classical firm operating in a highly competitive market the property rights of the owners are strong and direct. Managers will be severely constrained from diverting the firm’s resources to their own ends, they will have little scope for discretionary behavior, and the final outcome in this situation is likely to approximate that of standard neoclassical theory. In the large limited liability corporation, on the other hand, the property rights of the owners are more attenuated, leaving them with less control over the managers, who thus have more discretionary power to maximize their own welfare. The greater the attenuation of property rights, according to this school, the greater is the likely deviation of the allocation of resources from optimal efficiency. Attenuation of property rights and the dilution of economic incentives is seen as at its greatest in state enterprise (Hanke, 1986, p. 16):

What distinguishes public and private enterprises is the fact that public assets are not “owned” since they cannot effectively be transferred. This lack of transferability means that decisions taken by public bureaucrats and employees do not readily translate into changes in the market price of the firm’s assets, and the “owners” have little incentive to monitor public managers’ and employees’ behaviour Without changing the underlying property rights arrangements we cannot expect their behaviour to conform to that needed to maximise the value of a public enterprise’s assets.

While the language is unfamiliar and property rights theory also rests on restrictive behavioral assumptions, the general thrust of the argument can easily be related to a number of complaints commonly made about public enterprises: that shirking is easier and incentives to efficiency inadequate; that there is no takeover threat; that it is difficult to exert meaningful control over managements and to make them effectively accountable to the representatives of the public. Jackson and Palmer (1988) illustrate the intrinsic difficulty of instituting satisfactory performance measurement in public enterprises. The same point is made by Kay and Thompson (1986, p. 18): “the view that [privatization] contributes to economic efficiency is derived from the belief that private sector managers are subject to incentives and disciplines different from, and more demanding than, those which apply to their public sector counterparts.” The insistence of property rights theorists that the only way to achieve improved results is through a shift in such rights to private ownership is also paralleled by others who argue that inefficiency is intrinsic to the multiple goals and “political interference” which characterize public enterprise. We should note, however, that the property rights school would predict that the degree of improved efficiency to be derived from divestiture would be strongly influenced by the manner in which it was done, i.e., by the extent of attenuation of property rights in the newlycreated private enterprise. Their position, in practice, is probably rather close to the mainstream neoclassical advocacy of using divestiture to create more competition, rather than private monopoly power. Finalljr, under this heading we should ask about the evidence on the relative allocative efficiency of public and private enterprises. So far as the position in the industrial countries is concerned, the property rights literature is full of predictions of higher cost structures in public enterprises. Hanke (1987a, p. 977), for example, has no doubts: “Considerable evidence suggests that the public cost incurred in providing a given quantity and quality of output is about twice as

STATE

DIVESTITURE

IN DEVELOPING

great as private provision. This result occurs with such frequency that it has given rise to a rule of thumb: the ‘bureaucratic rule of two’.” Sales per employee are lower; operating expenses are higher; profitability and returns to capital are lower; productivity grows more slowly. From a survey of empirical studies that attempted to assess the relative performance of public and private enterprises in industrial countries Yarrow (1986, pp. 373-375) draws more circumspect conclusions. For US utility industries cost comparisons “tend to slightly favour public ownership.” In markets which are less monopolized, “the comparative performance studies suggest a more favourable verdict on private enterprise Even here, however, the evidence is far from overwhelming.” On the other hand, he reports a clear consensus on private cost advantages for activities that can be franchised on the basis of relatively simple contracts. In a 1983 study Millward (p. 258) concluded that there was no systematic evidence that public enterprises are less cost-effective than private firms. More recently (1988) he has studied the comparative evidence as it relates specifically to developing countries concluding (a) that there is no evidence of a statistically satisfactory kind to suggest that public enterprises have lower levels of technical efficiency, although there is a tendency in that direction; (b) that technical efficiency in public enterprises is not uniform, ranging from best to worst practice; (c) that public enterprises tend to be large and to have problems associated with size (as distinct from ownership); (d) that differences in product mix complicate public-private enterprise comparisons; and (e) that no satisfactory studies of cost-effectiveness have yet emerged for developing countries. It is impossible for observers lacking independent sources of information to choose between these alternative interpretations of the data and easy to see how in the absence of definitive results prejudice, pro or con, could flavor what ought to be a factual enquiry. There is a final matter to clear up in this discussion of comparative microeconomic efficiency, namely the prices in which efficiency is measured. We are referring here, of course, to the familiar distinction between market prices and social (or economic, or shadow) prices. The discussion is almost invariably conducted in terms of market valuations but this seriously biases the debate against public enterprises, whose pricing and production policies may be based, by government decision, upon social valuations. This question is begged, for example, by Peacock (1984, p. 6) when he asserts that economists are agreed that the desirability of privatization is to be

COUNTRIES

1471

judged by the effect it has on the present value of the excess of consumers’ and producers’ surplus over cost. Given the various influences in developing economies that can drive wedges between market and social valuations, to concede that efficiency should only be judged in market terms is to concede too much. On the other hand, to go the route of shadow pricing introduces all the additional difficulties of arriving at a satisfactory set of social valuations - and further reduces the possibility that economics will be able to offer any clear pointers in the privatization debate.

(c) Macroeconomic

efficiency

Opinion on the contribution of public enterprises to macroeconomic efficiency has gone full circle over the last two or three decades. One of the arguments for public enterprise was that it would facilitate national planning, giving control over the commanding heights of the economy and providing the state with an additional set of instruments through which to achieve its objectives. That is not an argument which nowadays has a very persuasive ring. The experience has not in many cases suggested that a large public enterprise sector has facilitated development planning, partly because of the absence of effective public accountability already mentioned. There have been difficulties in synchronizing the planning periods of public enterprises with those of national governments. And there have not infrequently been large discrepancies between public enterprise plans and achievements. In some cases - commercial banks provide an example - it has seemed difficult to bring public enterprises within national planning precisely because they were in the public sector and politically powerful. The more common tendency now is to include privatization in “structural adjustment” programs. Generally, the connection is through the budget. Because public enterprises are largescale claimants of budgetary subsidies of various kinds (and also represent an unutilized part of the tax base), divestiture is often seen as an important way of reducing the government’s budget deficit, with all its implications for inflation and the balance of payments, although Mansoor (1988) suggests that the budgetary effects will be keenly affected by the extent to which divestiture results in increased competition.” The connection between privatization and balance-ofpayments adjustment may also be directly through the export sector, for example when inefficient marketing boards discourage cash crop

1472

WORLD

production or nationalized mines mize foreign exchange earnings.

DEVELOPMENT

fail to maxi-

(d) Distributional efficiency The privatization movement represents a value shift as well as a change in perceptions about the most productive way of allocating resources.12 The desirability of providing adequate incentives for improving efficiency of resource use is seen as pulling against egalitarian concerns and the latter are given less weight than the former. Indeed, this distributional aspect has rightly been described as the suppressed question of the privatization debate. It arises in connection with the distribution of both income and wealth. Taking first the income dimension, we can ask who is likely to gain and who to lose from an act of divestiture. Actual outcomes will naturally vary from case to case but some general considerations do apply. First, the microeconomic arguments for divestiture - or, more strictly, for increased competition-emphasize the potential benefit to the consumer from lower cost production. If this is indeed achieved the consumers in question will have higher real incomes and the distributional effects will then hinge on the income classes from which these consumers are drawn. However, it is common for public enterprises to be used by governments as a means of subsidizing consumers, for example for wage goods or the delivery of economic services. The substitution of market-determined prices for the previously subsidized prices will create a group of unambiguous losers. These may or may not be made up of poverty groups, for state subsidies are a notoriously inefficient way of alleviating poverty. Subsidies tend to have a strong urban bias and even in the rural economy a disproportionate share of the benefits is liable to be captured by local elites. Where public enterprises are indeed used as a means of subsidization of consumers, a further consideration in assessing the distributional consequences of divestiture relates to the progressivity of the increases in other government expenditures, or reductions in taxation, made possible by the budgetary resources thus released. It is thus impossible to generalize about the expected net effect of the factors just considered on the overall degree of income inequality. Stren’s (1988) examination of the delivery of urban services in Africa well illustrates the complexities and the impossibility of any blanket identification of public provision with distributional progressivity. We can perhaps go a little further when consid-

ering the functional distribution of income. We suspect that public enterprises usually have larger work forces than could be justified in competitive conditions. There may also be a tendency for public enterprises to pay higher wages than would obtain under market conditions. The influence of trade unions tends to be stronger in public enterprises and they usually oppose divestiture. There is thus a rather strong presumption that the short-run effect of divestiture will be to reduce employment (although if the efficiency gains are large enough the longer-term outcome could be for more employment, as the enterprise expands). There may also be a tendency towards lower real wages, perhaps achieved through the erosion of a frozen level of nominal wages by inflation. Whether the benefits of lower labor costs are passed on to consumers or retained by the new private owners of the capital is, along the lines of standard microeconomic theory, likely to be highly sensitive to the degree of competition to which the enterprise in question is subject. If, as suggested above, the privatized concern will often retain considerable monopoly power (and there are no effective regulatory restraints on the exercise of this power) then the benefits will be retained, in whole or part, by the owners and there will, at least in the short term, be a shift in the functional distribution of income in favor of capital, tending also to increase inequalities in the size distribution. This brings us to the ownership of wealth. Of course, if divestiture results in a sustained shift in the functional distribution of income in favor of owners of capital this itself will aggravate wealth inequalities, as income from profits is converted into assets. But there is also the question, who will buy the enterprises to be privatized and at what price? Even though the buyers have undoubtedly been drawn from the more well-to-do segments of society, it may be that the widespread and highly fragmented sale of shares in privatized concerns has actually reduced the skewness of wealth ownership in countries like Britain. It is unlikely that a similar result could be achieved in most developing countries, however. On the contrary, the general expectation must be that, if ownership remains in national hands, it will pass to an already wealthy elite. When enterprises are sold at prices that do not fully reflect their market value divestiture will tend to accentuate inequalities in the distribution of wealth, and hence also in the distribution of income.13 Indeed, there have been complaints in a number of developing countries that one of the motives behind the divestiture movement has been a desire to promote a “crony capitalism,” from which

STATE DIVESTITURE

IN DEVELOPING

the relatives, friends and supporters of members of a ruling group are the chief beneficiaries. However, in some of the poorer developing countries there may be few or no potential buyers of public enterprises in the private sector other than multinational investors. If indeed ownership passes into foreign hands the effect will tend to be to widen international inequalities in the distribution of wealth, with the associated income streams increasing the gap between domestic and national income. However, in this context the main point will often be that the sale of national assets to foreign owners will be unacceptable to citizens and governments alike. An absence of national purchasers is thus more likely to be a constraint on divestiture than a source of widening international disparities.

COUNTRIES

1473

Divestiture thus attempts to substitute the relatively straightforward criterion of the marketplace for the multiple goals of the public enterprise; and to promote greater efficiency by clarifying its meaning and reducing its ambiguities. We think this simplification of objectives is a rather important aspect of the privatization movement. But divestiture itself is by no means free of contradictions. In Britain, for example, the desire to use this instrument as a way of reducing public sector borrowing has provided the government with a motive for maximizing the revenue from the assets it is selling. This can be done by offering the private buyer monopoly privileges, thus undermining the potential microeconomic efficiency gains.

4. CONSTRAINTS

ON DIVESTITURE

(e) Instrumental efficiency and multiple goals Finally, under the heading of efficiency we should revert to the multiplicity of policy objectives which public enterprises are commonly asked by governments to promote: profits for the government’s revenues and for reinvestment; increased and low cost output; assistance to underprivileged groups or regions; increased employment; price stability; a source of party or personal patronage; and so on. The point here is that efficiency can only be assessed relative to the goals being pursued, and much of the perceived “inefficiency” of public enterprises results from judging them by criteria which only partially correspond to the policy objectives they are intended to promote. This is a highly ambiguous defense of public enterprises, however. If there are too many goals and these contradict one another (as is commonly the case) then “efficiency” loses meaning. This further increases the difficulties of monitoring the performance of public enterprises and of safeguarding against “inefficiency.” The movement for divestiture is symptomatic not merely of a change in values but also of a simplification of the goals that previously public enterprises are expected to pursue. The intention is to substitute the single and simple goal of profit maximization for the multiple objectives of the state. In principle, such a simplification could be achieved without privatization, by governments telling the managers of state enterprises to pursue profit maximization (or some other single objective). But for many observers multiple goals (or perhaps multiple principals14) are among the defining characteristics of public enterprises and the likelihood that governments would stick to any one objective is slim.

Despite the fact that most attention has been paid to the issue of ownership transfer, a survey of the privatization process in developing countries argues that the most common form of divestiture has been informal closure, concluding that “the number and scale of actual divestitures in developing countries seems extraordinarily small in the light of the considerable rhetoric that the idea has generated” (Berg, 1985, p. 19). Moreover, other modes of privatization - such as leasing of state-owned enterprises through private management contracts, contracting out of services to private agents or simple withdrawal of public agencies from the provision of goods and services - have also been options that, as yet, have been employed to a very limited extent. While the transfer of ownership from public to private hands is perhaps the most clear-cut mode of privatization, this mechanism faces major is problems. l5 The first and principal constraint the smallness of the capital market available to divesting authorities in most LDCs. In the advanced economies, ownership transfer has largely been through the sale of equity. Although this has posed problems of valuation and in terms of the relation between the realized proceeds from equity sales and the net present value of future earnings, the instrument for transfer does not itself pose a problem so long as the enterprise is regarded as being commercially viable. But the limited capitalization and number of traded issues of developing country stock remains a major barrier to such methods of transfer. In most developing countries there has been very limited equity financing, while local markets have rarely been able to provide satisfactorily long-run investment resources. Consequently, debt financing has emerged as a more common

1474

WORLD

DEVELOPMENT

method of raising capital. The absence of organized capital markets is particularly pronounced in economies where the pressure from donors for ownership transfer is most acute. In Sub-Saharan Africa, for example, formal stock markets exist only in Lagos, Harare, Nairobi and Abidjan but since 1980 these markets have declined when measured in terms of capitalization and issues (Moose, 1986, p. 155). Moreover, it is instructive to note that the total listed capitahzation of LDC markets barely comprises 10% of that of the London Stock Exchange and only 3% of the London and New York exchanges combined. Where equity sales are actually a feasible mode of ownership transfer, the limited size of most developing country stock exchanges has implications for the consequences of this approach. What is likely to be engendered is a reverse crowding-out effect, with sales of equity mopping up available capital at the expense of other private sector companies. There is some evidence that this has been the case with major flotations in the UK, but clearly any such effect would tend to be magnified in the presence of a greatly restricted market. The factors inhibiting the development of capital markets and investment in developing countries are various, with some more susceptible to remedy than others. On the former category, one could include entry barriers, such as restrictions on the nationality of investors, but a more problematic issue relates to the general structure of expectations and its impact on the level of foreign investment and on domestic savings mobilization. Investment will clearly be structured in part by expectations regarding future actions by the state with respect to ownership and tax policy, as also by the general economic environment. Where such perceptions are bleak, the access that financial institutions have to savings will be stunted, while capital flight will commonly be a major feature of the economy. This will have direct implications for the fiscal profile. In Mexico and Argentina, for instance, capital flight in 1983 may have amounted to around 10% of the fiscal deficit and for Venezuela this share may conhave exceeded 75%. l6 Lack of economic fidence and political stability remain critical factors ensuring the continuation of massively sub-optimal levels of investment in a large number of LDCs. The barriers to a more accelerated withdrawal of the public sector from the provision of goods and services are, however, not merely those associated with the instruments of transfer. Absence or constriction of capital markets can be surmounted by recourse to direct cash sales or direct

transfers by the state to chosen private agents. In the majority of instances this has been to the previous owners from whom the enterprises had been confiscated or transferred, as in Bangladesh where extensive denationalization in the manufacturing - particularly textile - sector has recently occurred. But in many countries such sales or transfers are likely to attract considerable domestic political opposition. This opposition would tend to come not only from the existing work force but also from political lobbies. Special sensitivities are liable to be aroused by any suggestion that ownership could pass to foreigners or to racial minorities. Even if the basis of this opposition may be related to particularistic interests, and hence be irrelevant to an economic calculation of the likely benefits or costs of ownership transfer, the fact remains that in the absence of sufficient support for such a policy the political costs will remain high.” If this is the case then the expected pace of implementation will remain sluggish. As fundamental in determining the nature and pace of divestiture is the question of potential profitability, itself in part a function of the level of the debt obligations of any enterprise that is proposed for ownership transfer. Although it can be argued that some public enterprises have been inefficient users of resources, much public sector activity is concentrated in areas of low profitability. Provision of “merit” goods has rarely been a profitable area, especially when generalized. Where there is greater scope for profits as with national energy companies or toll authorities - these tend to be the least likely candidates for divestiture on the grounds that they are natural monopolies or providers of public goods. Many public enterprises in developing countries are in fact loss-makers. Divestiture in such circumstances is only likely to be feasible when significantly sweetened by market and tax concessions. As suggested earlier, this will further militate against increasing the level of competition. Foreign exchange shortages and weak or volatile demand for output - a common feature where absorption-reducing policies are combined with falling terms of trade-would be other contributory factors. Thus, the restoration of reasonable prospects of profitability is liable to depend on the implementation of a whole package of “adjustment” measures, and on the scale of external support which this attracts. In addition, where public enterprises are profitable the rationale for divestiture becomes less obvious if the objective is to reduce public deficits, as against a more general allegiance to the virtues of private ownership. Where the sale of profitable enterprises is entertained, familiar problems con-

STATE

DIVESTITURE

IN DEVELOPING

cerning valuation will surface. In addition, where very high rates of inflation have been sustained - as in many Latin American economies - it will be difficult to calculate the real worth of an enterprise. If ownership transfer remains a problematic option in many LDCs, it is increasingly clear that other less visible methods of divestiture are being employed. Most common among these has been the tendency for public agencies simply to withdraw from the provision of particular services. In many cases, as with the supply of inputs to farmers in large parts of Sub-Saharan Africa, this has occurred over a relatively protracted period of time. With inadequate recurrent budgeting and weak management systems one widespread consequence has been a dilution in the range, frequency and suitability of public services. Where more acute resource constraints have emerged as in the period since 1980 - this has exacerbated existing trends, leading to de facto withdrawal by public agencies. In some cases, such as education, this has led to greater private sector substitution but where public agencies have had monopoly import rights to scarce commodities such as pesticides - simple transfer of function has not proven widely feasible. In addition, the limitations to simple transfer of functions to the private sector can be derived from the actual structure and depth of markets in which such substitution might be deemed desirable. This can most clearly be seen with regard to the agricultural sector in developing countries. By and large, there is agreement that areas such as extension and agricultural research are unlikely direct candidates for the private sector. What is more feasible is the introduction of charges for the services provided, although with the state remaining the executing agency. In contrast, it is increasingly held that areas such as the supply of inputs and credit are more suitable fields for the private sector to take over. The factors inhibiting divestiture are well illustrated in the agricultural sector, although the experience with regard to the transfer of functions hitherto undertaken by the state has so far been limited. In a few countries - such as Senegal - the state has largely withdrawn from the direct provision of some basic agricultural services, such as credit and seed supply (Abt Associates, 1985). But even where the will exists among governments to raise the share of the private sector in agricultural provision, the obstacles remain significant. First, there is the problem of a relative lack of profitability in such activities when compared with trading, urban sector activity and rentseeking that commonly constitute alternative avenues for organized capital. Where agriculture

COUNTRIES

1475

is subject to high levels of risk and meager levels of capitalization - creating collateral constraints - the variability in the demand for inputs will tend to depress profitability. In such circumstances, the paradoxical consequence may be that in order to sustain current - almost always sub-optimal - levels of purchase of say, inputs, the state will have to offer a combination of subsidies and preferential market conditions to potential private suppliers. This may be particularly necessary where an allied feature of adjustment is the elimination of subsidies on agricultural inputs, a further demand constraining factor. Second, with regard to credit, the enforcement costs of lending to agriculturists are commonly high and private financial institutions tend consequently to be reluctant to enter this area. In consequence, in almost all developing countries, informal capital markets predominate in the rural economy. But the reasons for establishing public sector financial institutions in the first place other than for facilitating the transfer and utilization of agricultural sector savings - still hold. Informal sector lenders tend to charge a high risk premium, that sometimes extends to controls over marketed output. While there is no inherent reason why high interest rates should necessarily deter agricultural innovation, in the context of many low-income countries the absence of an organized banking and credit system would tend to depress the use of productivity-raising inputs and capital formation. A third issue relating to the delivery of services to agriculture concerns their coverage. While public interventions in the delivery of extension and other services may be highly imperfect, using de facto rationing, retreats from generalized coverage to levels that may be financially viable will have both distributional and political implications. Scaled-down private sector interventions would inevitably imply a trade-off between coverage and possible efficiency gains.

5. CONCLUSION Rather than attempt a systematic summary we shall try finally to pull some loose ends together and suggest some general conclusions. The distinction between the objectives to be promoted and the instrumental efficiency of alternative means of achieving them is an organizing principle that underlies much of the foregoing discussion. We have suggested that the divestiture movement is symptomatic of a value shift among the electorates and governments of many of the major Western countries which has

1476

WORLD

DEVELOPMENT

been exported to developing countries both as part of the general spread of ideas and through the policy conditions attached to external assistance. It is a value shift which places less weight on egalitarian concerns, although we have also drawn attention to the difficulties of arriving at firm conclusions about the distributional consequences of public enterprise. Nevertheless, even if for the moment we assume that divestiture would result in improved efficiency of resource use, no conclusion can be drawn from this improvement about the desirability of this change in ownership unless it is further assumed that the improvement in productivity results in increased utility or welfare distributed in a manner regarded as satisfactory (by whom we leave aside). The desirability or not of some particular act of privatization cannot be settled in abstraction from the values of society, or of its representatives, and from the relative importance attached to its various goals. This consideration also relates directly to the simplification of enterprise objectives which is an important aspect of the privatization movement. The gain is clear: it clarifies the meaning of efficiency, gives management a well-defined goal and thus makes the monitoring of performance far easier. The loss is that it will often be inappropriate to assess enterprise performance by the simple test of the market because society’s needs are too complex to be reduced to that criterion. Almost the same point is made by questioning the appropriateness of replacing evaluation at shadow prices by market prices, given the many conditions in developing countries that prevent market prices from accurately reflecting social values. We hasten to add, however, that probably only a modest part of observed public enterprise “inefficiency” could be justified along such lines. A further conclusion to be drawn from this set of considerations is that when divestiture occurs there is likely to be a need to establish regulatory mechanisms for protecting the public interest: safeguarding against the abuse of monopoly power, ensuring some minimum coverage of the services provided and so on. At the same time, we hesitate to place much faith in such devices, for regulatory authorities are notoriously prone to be captured by the special interests they are intended to oversee, or to be ineffective in the face of such interests. On the question of instrumental efficiency, we have shown that the claims of divestiture should by no means be taken for granted, notwithstanding the admittedly poor record of many public enterprises in developing countries. There are essentially two difficulties. One is that main-

stream neoclassical theory does not provide a sufficient analytical basis for strong normative conclusions. It is particularly weak in the treatment of the dynamic aspects of enterprise performance which are of special importance for economic development. Even when it comes to static allocative efficiency, the unrealism of much formal theorizing, and its preoccupation with optimization and equilibrium, result in a shakier theoretical basis for claims about the superior efficiency of market allocations than many of its devotees are willing to admit. In any case, the policy thrust of neoclassical theory is to favor competition rather than private ownership per se. It is commonly claimed that advocates of privatization fall into the error of confusing divestiture with competition. We have not entirely agreed with that view, pointing particularly to the arguments derived from the property-rights literature. Nevertheless, it is awkward for its advocates that the main normative thrust of microeconomic theory is in favor of more competition and that divestiture is neither a necessary nor a sufficient condition for the enhancement of competition. We might add here the particular difficulties associated with the theory of the second best: that there can be no assurance of improved allocative efficiency when privatization occurs (as it always must) in the presence of a large number of factors preventing a Pareto-optimal allocation of resources. These considerations leave the claim that divestiture will lead to a more efficient use of resources in ambiguity. There is a further weakness, namely that devotees of privatization have a tendency to compare a worst-case picture of public enterprise with a more idealized view of the private sector. This overlooks some important economic realities: that public corporations were often created to fill vacuums left by private enterprise, with little assurance even now that entrepreneurs would be ready to move in; that there are manifold factors preventing economic conditions even remotely satisfying the conditions of standard neoclassical price theory; and that private enterprises are also subject to political control and influence, albeit less directly than public enterprises. The second main area of difficulty is the absence of definitive information on actual private-public comparative enterprise performance. While we are concerned about substandard performance by state firms, and suspect this to be common, the evidence is mixed and inconclusive. What does seem to be established is that, within and across countries, there is a wide variation in levels of technical and managerial

STATE

DIVESTITURE

IN DEVELOPING

efficiency in public enterprises, which suggests that it may be more fruitful to enquire into the reasons for this, rather than pre-select ownership as the key determinant. A practical lesson to draw from the above is that the way in which divestiture is designed is likely to be of much importance, placing a large premium on doing it in ways which enhance freedom of entry and the intensity of competition in the markets in question. However, we have also pointed to a dilemma here. To make a public enterprise attractive to potential purchasers and to maximize the short-term receipts by the fist resulting from the sale, the government will be under pressure to grant the enterprise precisely the monopoly powers which reduce the microeconomic efficiency gains that it is intended to derive from the transfer of ownership. This dilemma has not yet been satisfactorily resolved by advocates of divestiture and is likely to be particularly acute in developing country circumstances. This then brings us to the question, what is the practical potential of divestiture for changing the structure of ownership and markets? In addition to the dilemma just mentioned, we have drawn attention to various rather severe constraints on the extent and pace of divestiture. These include the weakness or non-existence of capital markets through which it could be mediated, the political sensitivity of selling to non-nationals, and the often unattractive financial record of potential candidates for sale. Here is another dilemma: the desirability to the government of getting rid of a particular enterprise will be inversely correlated with its saleability. There is a related question about the timing of

1477

COUNTRIES

privatization. Engineering the transfer of an enterprise to private ownership during a period of recession, with its depressing effects on profitability and confidence, is likely to be particularly difficult. Thus, while privatization is often regarded as an integral part of a program of structural adjustment it may, in practice, become more feasible after extensive restructuring has been undertaken and the economy has responded. Given all these difficulties, we doubt whether divestiture will be widely employed and have noted that movement in this direction has so far been quite limited. The main point is that the conditions necessary for it to be both appropriate and successful are rather restrictive. Even its proponents acknowledge that divestiture is not “the most promising instrument and is, in any case, not a process amenable to outside pressure” (Berg, 1985, p. 63). We therefore see it as a specialized instrument for adoption in a rather narrow range of circumstances, not as a general solution to sub-standard public enterprise performance. This is not to suggest that the privatization movement is unimportant, however. One of its main effects has been to draw attention to the often grave economic deficiencies of many public enterprises and the importance of correcting these, whether or not divestiture is the best way of doing so.ls The problems and economic concerns which underlie the privatization movement have, moreover, nearly stopped the creation of new public enterprises dead in its tracks. That may, in the end, be their chief contribution to economic policy in developing countries.

NOTES 1.

Commander

(1987). See also Bardhan (1984) on

India. 2. For a highly critical assessment of their role see Bauer (1954). Also Hart (1982). 3.

Cited in Killick

4. Reported p. 2369.

7. See Kamien and Schwartz (1982, Chap. 3) for a valuable survey of the evidence; also Kamien (1987). 8. On the theory ef al. (1982).

of contestable

markets

see Baumol

(1986).

in West Africa,

10 November

1986,

to microeconomic policies, the 5. “In relation characteristic form [of the shift to conservatism] has been the move to privatisation” (Budd, 1987, p. 188). 6. In presenting this and related lines of criticism, have been particularly influenced by the arguments Nelson and Winter (1982).

we of

9. However, it is clear that in discussing the best means for awarding franchises in conditions of natural monopoly, the various bidding systems that have been devised will not invariably yield a more efficient outcome. Where investment in durable assets is significant and where technological and market uncertainties prevail, franchise bidding will tend to require an administrative apparatus little different from the prior, regulated system. In other words, market solutions will tend to require administrative support (Williamson, 1985).

1478

WORLD

DEVELOPMENT

10. See Furubotn and Pe;jovich (1972) for a survey of this literature, and Hanke (1986 and 1987b) for an application to the privatization issue. 11. A desire to reduce the public sector borrowing requirement has similarly been a feature of the privatization movement in the United Kingdom, although only achieved by the dubious expedient of treating the proceeds of asset sales as current revenue receipts. 12. “In identifying the key characteristics of conservative economics and the turnaround in thinking in the eighties modern conservatives drew two lessons from the past. First, the trade-off between efficiency and equity is less favourable than previously supposed. Greater equality costs more in inefficiency and lost output in the long run than was previously assumed. Hence, on average, new conservatives have backed off redistribution and state intervention in the search for greater efficiency; with a bigger cake to divide, eventually everyone may get more. Second, the mere fact that market failures can be identified does not automatically mean that government intervention can do better.” (The Editors in a special issue on The Conservative Revolution of Economic Policy, 5 October 1987, pp. 12-13.) 13.

An interesting

twist to this discussion

is given by

the case of Malaysia, where public enterprises have been used as a means of redressing the concentration of wealth in the Chinese population through a policy of divestiture to Malay entrepreneurs (Mallon, 1982). 14. See Jones point.

(1982,

15. For an extended see Heald (1985).

p. 5) for a discussion

discussion

of the transfer

of this

process

16. See Riely (1986) and Conesa (1986). A recent and undoubtedly underenu:nerated -estimate of bank and fiduciary deposits with Swiss banks suggests that around 25% of such total deposits originated from developing countries. At its most extreme, capital flight has in recent years assumed drastic proportions. In the case of Liberia, for example, bank and fiduciary deposits in Switzerland in 1984 amounted to $1,675 million when GDP appears not to have reached $1,000 million in that year (Christensen, 1986). 17. Shackleton (1987) similarly argues that divestiture is liable to fall foul of strong political resistances. 18. See Nellis (1986) for a discussion of alternative techniques for the improvement of public enterprise performance.

REFERENCES Abt Associates, “Senegal: Agricultural policy analysis,” Mimeo (Cambridge, MA, 1985). Bardhan, Pranab, The Political Economy of Development in India (Oxford: Basil Blackwell, 1984). Baumol, William J., “Entrepreneurship in economic theory,” American Economic Review, Vol. 58 (1968), pp. 64-71. Baumol, W. J., J. C. Panzar, and R. D. Willig, Contestable Markets and the Theory of Industry Structure (New York: Harcourt Brace Jovanovich, 1982). Bates. Robert H.. Markets and States in Tronical Africa: The Political Basis of Agricultural Pohcies (Berkeley: University of California Press, 1981). Bauer, P., West African Trade (Cambridge: Cambridge University Press, 1954). Berg, Elliott, “Divestiture of state-owned enterprises in LDCs,” Report prepared for the World Bank (Washington, DC: World Bank, November 1985). Berg, Elliott, “Privatization: Developing a pragmatic approach,” Economic Impact, Vol. 57, No. 1 (1987), pp. 611. Buchanan, James M., “Rights, efficiency and exchange: The irrelevance of transaction costs,” in Liberty, Market and State (Brighton: Wheatsheaf Books, 1986) pp. 92-107. Budd, Alan, “On the conservative revolution,” Economic Policy, No. 5 (October 1987), pp. 185-191. Christensen, B. U., “Switzerland’s role as an international financial center,” IMF Occasional Paper 45 (Washington, DC: IMF, 1986). Coase, R. H., “The problem of social cost,” Journal of Law and Economics, Vol. 3 (1960).

Commander, Simon, The State and Agricultural Development in Egypt Since 1973 (London: Ithaca Press, 1987). Commander, Simon, and Tony Killick, “Privatisation in developing countries: A survey of the issues,” in Cook and Kirkpatrick (1988). Conesa, E. R., Fuga de Capitales en America Latina, 197&1985 (Washington, DC: Inter American Development Bank, 1986). Cook, Paul, and Colin Kirkpatrick (Eds.), Privatisation in Less Develoued Countries (Brighton: Wheatsheaf \ Books, 1988).’ Furubotn, Eirik G., and Svetozar Pejovich, “Property rights and economic theory: A survey of recent literature,” Journal of Economic Literature, Vol. 10, No. 4 (December 1972), pp. 1137-1162. Hanke, Steve H., The privatisation option: An analysis,” Economic Impact, Vol. 3 (1986), pp. 14-20. Hanke, S. H., “Privatization,” in J. Eatwell et al., New Palgrave Dictionary of Economics (London: Macmillan Press, 1987a). Hanke, S. H. (Ed.), Prospecfs for Privatisafion (New York: Academy for Political Science, 1987b). Hart, Keith, The Political Economy of West African Agriculfure (Cambridge: Cambridge University Press, 1982). Heald, David, “Privatisation: Policies, methods and procedures,” Asian Development Review (1985). pp. 57-95. Heller, P. S., and A. Tait, “Government employment and pay: Some international comparisons,” IMF

STATE

DIVESTITURE

IN DEVELOPING

Occasional Paper 24 (Washington, DC: IMF, October 1983). Jackson, P. M., and A. J. Palmer, “The economics of internal organisation: The efficiency of parastatals in LDCs,” in Cook and Kirkpatrick (1988). Jones, Leroy P., Public Enterprise and Economic Develooment: The Korean Case (Seoul: Korea Development Institute, 1975). ’ Jones, Leroy P. (Ed.), Public Enterprise in Lessdeveloped Countries (Cambridge: Cambridge University Press, 1982). structure and innovaKamien, Morton I., “Market tion,” in J. Eatwell et al., New Palgrave Dictionary of Economics (London: Macmillan Press, 1987). Kamien, Morton I., and Nancy L. Schwartz, Market Structure and Innovation (Cambridge: Cambridge University Press, 1982). Kay, J. A., and D. J. Thompson, “Privatisation: A policy in search of a rationale,” Economic Journal, Vol. 96 (March 1986), pp. 18-32. Killick, Tonv. Policv Economics: A Textbook of Applied Economics on Developing Countries (London: Heinemann Educational Books, 1981). Killick, Tony, “The role of the public sector in the industrialisation of African countries,” Zndustry and Development, Vol. 7 (1983). Killick, Tony, “Twenty-five years in development: The rise and impending decline of market solutions,” Development Policy Review, Vol. 4, No. 2 (June 1986) -pp. 99-116: Mallon. Richard D.. “Public enterprise versus other methods of state intervention as instruments of redistribution policy: The Malaysian experience,” in Jones (1982). Mansoor, Ali, “The fiscal impact of privatisation,” in Cook and Kirkpatrick (1988). Matthews, Kent, and Patrick Minford, “Mrs Thatcher’s economic policies, 1979-87,” Economic Policy, Vol. 5 (October 1987), pp. 57-101. Millward. Robert. “Measured sources of inefficiency in the performance of private and public enterprises in LDCs.” in Cook and Kirkpatrick (1988). Millward, Robert, and D. Parker, “Public and private enterprise: Comparative behaviour and relative effiin R. Millward et al., Public Sector Ecociency,” nomics (London: Longman, 1983), pp. 9%274. Moose, Richard M., “Alternative sources of capital,” in C. Lancaster and J. Williamson (Eds), African Debt and Financing (Washington, DC: Institute for International Economics, 1986), pp. 147-161. Mosley, Paul, “Privatisation, policy based lending and

COUNTRIES

1479

in Cook and Kirkpatrick World Bank lending,” (1988). Nellis, John R., Public Enterprise in Sub-Saharan Africa (Washington, DC: World Bank, 1986). Nelson, Richard R., and Sidney G. Winter, An Evolutionary Theory of Economic Change (Cambridge, MA: Harvard University Press, 1982). Peacock, Alan, “Privatisabon in perspective,” Three Banks Review, Vol. 144 (December 1984), pp. 3-25. performance of public Pryke, R., “The comparative and private enterprise,” Fiscal Studies, Vol. 3, No. 2 (1982) pp. 68-81. Riely, Frank Z.. “Third World capital flight: Who who loses?” ODC Policy Focus, No. 5 gains, (Washington, DC: Overseas Development Council, 1986). Schumpeter, J. A., The Theory of Economic Development (Cambridge, MA: Harvard University Press, 1934). Shackleton, J. R., “Privatisation: The case examined,” National Westminster Bank Quarterly Review (May 1984) pp. 5%73. Shackleton, J. R., “The political economy of privatisation in less-developed countries,” Mimeo (London: Polytechnic of Central London, September 1987). Simon, H. A., “Theories of decision making in ecoAmerican Economic Review, Vol. 49 nomics,” (1959), pp. 253-283. Simon, H. A., Administrative Behaviour, Second edn. (New York: Free Press, 1965). Stren, Richard E., “Urban services in Africa: Public management or privatisation?“, in Cook and Kirkpatrick (1988). Tomlinson, J. D., “Ownership, organisation and efficiency,” Royal Bank of Scotland Review, Vol. 149 (March 1986), pp. 11-23. Williamson, 0. E., The Economic Institutions of Capitalism (New York: Free Press, 1985). World Bank, Annual Report 1986 (Washington, DC: World Bank, 1986a). World Bank, Financing Adjustment with Growth in Sub-Saharan Africa, 198690 (Washington. DC: World Bank, 1986b). World Bank, “Structural adjustment lending: A first review of experience,” Mimeo (Washington, DC: World Bank, 1986~). Yarrow, George, “Privatisation in theory and practice,” Economic Policy, No. 2 (1986). Zulu, J., and S. Nsouli, Adjustment Programs in Africa (Washington, DC: International Monetary Fund, April 1985).