Welfare economics, political economy, and policy reform in Ghana

____________(AJ PS 1381 POLICY RESEARCH WORKING PAPER 1381 Ecnomic policy reform wras W elfare Economics, ~~~~~~~~~~~~not invented in AThica in the Political Economy, and 98 It -hais a=ways been' Polic Reform in Ghana -CEconomic -olicy reform is ~aot gainers and losers, and - vIYosIto stop it. The wnner are not as likelv to- puh oreform with the - smevgo. ut therecomes . .= -. 0 - a ... o. - ch thei Welfare Economics, ,-'g'0Economicpoliynr!terests becom oh~lmingan pposition The World Bank Western Africa Deparnment Ghana Resident Mission 44 November 1994  POLICY RESEARCH WORKING PAPER 1381 Summary findings Welfare economics develops the logic of how the gains imports - dominated for more than a decade in the of the gainers a. I the losses of the losers should be nineteenth century. It eventually split the Tory party. weighed against each other, in a specific ethical Britain's transformation from an agrarian nation to a framework. Political economy develops the logic of how manufacturing one spcllcd the decline of the power of they wilt be weighed against each other, in the context of the landed aristocracy and the ascendance of sociopolitical institutions. manufacturing. IIL the end, the Corn Laws were repealed Kanbur applies the disciplines of both welfare because of the growing power of the urban masses and economics and political economy in this evaluation of their employers, and the debate soon turned to policy reform in Ghana. When considerations from both protection against imports from fast-industrializing disciplines are aligned, he explains, policy r'form not France and Germany. only should be enacted but is also likely to be enacted. Such episodes from history help us understand the Often, though, there is no such alignment, so reforms protection of rice in Japan today, for-example, and - that might improve social welfare do not succeed. the focus of this paper - the past decade of reform in Anal, says Kanbur, should consider past reforms Ghana, and thie decade that awaits. from both perspectives, and should leam from history, in Kanbur argues that the political economy of policy evaluating proposed reform - so they can assess both reform in Ghana is likely to prove tougher in the second the desirability and the feasibility of reform. decade than in the first, for three reasons: (1) the Policymakers, on the other hand, should work toward economic situation in the second decade is no longer one organizing and mobilizing the gainers from reform that of absolute disaster, with only one way to go; (2) in the would advance social welfare, so that resiscance to such second decade, policy reform will have to coincide with reform by the losers can be overcome. the transition from military to constitutional rule; and In an example from history, he explains that Britain's (3) the nature of the reforms to be undertaken in the debate over the Corn Laws - basically a device for second decade is different from tbat of those undertaken protecting domestic production of grain from cheap in the first This paper -a product of the Ghana Resident Mission, Western Africa Department- is the text of the FirstAnnual Social Sciences Lecture of the Ghana Academy of Arts and Sciences, delivered in Accra in May 1993. Copies of this paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact Michele Youssef, room JS-250, extension 34614 (25 pages). November 1994. The Policy Rzscarch WZorig Paper bS y daXrathe the fsiear of wDrssem prin to Cenwgter c of tems abow developnrt izuesAn objeciv of dse sai is toga thefdos out qucky, even if thepresettiow are kss han Jkl@ypofbe& 7be pap caryte mama of the aswbors and Am&l be ged and caedaccordsolyg. TOh fouirgV4 inepiarions and candusiae the Iasd)' owmn0dsbould beat crd to tc NMBcnk ds EuimBaedof ha orany ofiAs membecows; Produced by the Policy Research Dissemination Center  Welfare Economics, Political Economy and Policy Reform in Ghana by Ravi Kanbur The World Bank * Test or the Frt Annua Socda Sciue Leture of the Ghan Acadmy of Arts aud ces, elivered i Acm i May i993.  1 Welfare Economics, Political Economy and Policy Reform in Ghana by Ravi Kanbur The World Bank 1. Introduction: Economic Policy Reform The subject of this paper is as old as the subject of economics, but one which is in the forefront of African policy making today. Some of the classical writings in economics arise from the need to analyze various economic policy reforms; yet the moming newspaper headlines are also full of the latest twists and turns in the reform saga. And, what is more, the issue of economic policy reform is global--it is not one specific to Africa. Tr, read some recent writings, one might think that economic policy reform was a-peculiarly African phenomenon. Yet there has been hardly an epoch in history where the issue was not prominent, and there is hardly a region in the world where the major debates of the time do not centre around economic policy reform. Even if we take as distant a time as the -middle of the nineteenth century, the debate in Britain over the Corn Laws dominated the political landscape for over a decade; it eventually split the Tory party, and the final outcome was a reflection of the emerging socio-economic transformation of the nation from an agrarian past to a dominant future as the workshop of the world. It also spelt the beginning of the end of the power of the landed aristocracy, and the start of the dominance of manufcturing. And it  2 was to help elucidate and analyze the issues involved that David Ricardo developed what we now call the "Ricardian" model of the distribution of income between classes. The issues in the Corn Laws debate were simple. Stripped of their institutional garb, they were a device for protecting domestic production of com from cheap imports. Sounds familiar? It should, because the same question comes up in our newspapers, in Ghana and elsewhere, every day of the year. Cheap imports would, of course, help producers at the expense of consumers. All of the usual arguments about the "national interest", *food security', etc., were made at the time. The battle lines were drawn up very clearly. The Economist magazine was founded in 1843 specifically to argue the case against the Corn Laws, and it has maintained its Free trade position ever since. At the end of the day, it was not necessarily the elegant economic modelling of Ricardo, nor the elegant and stylish economic journalism of The Economist, which won the day. The Corn Laws were in fact repealed, but it was the growing numbers and the growing power of the urban masses, and their employers, which spelled the doom of that particular piece of protection. It goes without saying that very soon the debate turned to the protection of manufacturing against the imports from fast industrialising Germany and France. Not surprisingly, the urban population was now in favor of protection-citing, yet again, the national interest. I am sure that te irony of this was not lost on The Economist, which steadfastly maintained its Free Trade position.  3 But why go to nineteenth century Britain? Why not look in our own times at one of the most economically powerful nations on earth-Japan? Let us look, in fact, at rice policy in Japan, which bears an uncanny resemblance to the Corn Laws in Britain a century and a half ago. Through various means and devices, the importation of rice into has been hindered when not directly banned or deterred by punitive tariffs. One does not need fancy economics to surmise that the main beneficiaries of this policy are the rice farmers, and the main losers must be rice consumers. Whatever the true national interest, it is a fact that farmers' interests have, so far, been protected. The political economy of this is in some ways simple, in other ways, complex. It is recognised that the method of election to the Japanese Parliament favors the rural constituencies in which rice farmers live. Also, although urban rice consumers pay more for rice as a result of the rice policy, the links between urban dwellers and their rural folk have, until now, been strong. Moreover, it is difficult to underestimate the cultural significance of rice in Japan. No wonder, then, that protection of rice in Japan has proved difficult to dislodge from the outside. However, careful observers of the Japanese scene were already saying some years ago what others have started saying in recent years, in the wake of Japanese-American trade negotiations. This is that as the socio-economic and demographic transfornations underway in Japan gradually work their way through, support for rice protection will become more and more of an anachronism. As a generation grows up in Japan that had even less to do with the countryside than its parents and grandparents, and as  4 electoral reform evens out the distribution of power, the internal forces supporting rice protection will begin to crumble. In fact, this is already happening before our very eyes. I want to draw two main conclusions from the above discussion of economic policy reform. The first is that the issue of economic policy reforn was not invented in the 1980's in Africa-it has always been with us, everywhere. The second is that economic policy reform is about gainers and losers. There are bound to be losers from policy reform-and these will fight, perhaps viciously, to stop it. But there are also likely to be winners-history has shown that these are not as likely to push for reform with the same vigor, but there comes a point when their interests become overwhelming and opposition crumbles. Thus economic policy reform is to do with gainers and losers. This is why economic policy reform touches upon two aspects of the subject of economics-welfare economics on the one hand, and political economy on the other. The former develops the logic of how the gains of the gainers and the losses of the losers should be weighed against each other, in the context of particular ethical frameworks; the latter analyses how the gains of the gainers and the losses of the losers will be weighed against each other, in the context of the institutions of a particular socio-political system. 2. Welfare Economics and Political Economy  5 By many accounts, the term Welfare Economics was coined by the great economist A.C. Pigou. In modem economics, the term has come to characterise that part of the subject that develops the logic of choosing between different policy options, by reference to their consequences for the welfare, or the well being, of individuals in a society. As seen from the discussion in the last section, every comparison of policy options will almost certainly lead to a comparison of gainers versus losers. Thus it is incumbent upon welfare economics to provide us with a framework for making such comparisons. The history of welfare economics in the first half of this century was, more or less, to avoid this issue. The Utilitarians of the nineteenth century, and their twentieth century followers, had no difficulty in comparing and, in fact, in 'Tadding up" pleasure and pain, gains and losses, across individuals. But some in the following century, most notably Lionel Robbins and his followers, felt that economics did not have any special competence to pronounce on such matters-which required "value judgements". The only case that Robbins was willing to allow for comparison was the case named after Vilfiedo Pareto-this was the case of a Pareto improvement, when a policy change made nobody worse off and some people better off. Of course, such a restriction meant that economists could never pronounce, or help to pronounce, on any policy reform that was actually likely to be undertaken, and the first half of this century saw many a case of intellectual gymnastics designed to get the economist, Houdini-like, out of the clutches of this straight jacket. Thus Kaldor and Hicks  6 proposed "compensation criteria". A policy reform could be pronounced a social improvement, even though there were gainers and losers, if the gainers could compensate the losers to leave them as well off as they were before, and still have something left over for themselves. The crucial point is that the compensation need only be paid hypothetically, not actually. The logical conundrum should now be clear. If the compensation was actually paid, we could appeal straightforwardly to the uncontroversial Pareto criterion. But if the compensation was not paid, then there could be no argument whatsoever for ignoring the losses of the losers relative to the gains of the gainers. In the second half of the twentieth century-actually, more like in its third quarter-a group of economists, led by Amartya Sen, Anthony Atkinson and Joseph Stiglitz-accepted that gains and losses were the natural business of policy reform, that these could and should be compared and, moreover, the comparison should be related to the measurement of inequality and poverty in society. Thus, it was argued, analysis of policy cannot be done without value judgements. In fact, it is naive to think that such a value free analysis can be accomplished. ,What is important, however, is that the value judgements be made explicit and above board-so that we ali know what somebody means when they invoke the "national interest". The job of the welfare economist is to develop frameworks in which value judgements can be rationally discussed and made precise and, by the same token, to expose and to lay bare the value judgements underlying policy claims.  To my mind, the most appealing criterion for evaluating policy reform is its impact on poverty-in other words on the well being of those at the bottom of society as measured by their income, their consumption, or their fulfillment of basic needs and access to basic social services. This is my value judgement, but I believe it to be widely shared. However, stating the value judgement in general terms is only the first. step. The real difficulties start when we try to operationalise the concept of poverty and to taclde the tough questions of tradeoffs between the short term and the long term. This is not the place to discuss in detail the massive literature on poverty measurement. Suffice it to say that, despite the huge conceptual. and empirical problems, we now have a basis on which we can discuss coherently the imDlications of policy reform for the poor. Every reform .creates gainers and losers. The identity of gainers and losers will be determined by the nature of the reform and by the nature of the economic structure being reformed. Thus, for example, reducing a tariff will benefit consumers of the commodity and hurt the producers. Now, there are rich and poor consumers and rich and poor producers (in the latter group I am including the workers who work in the factories that produce the good in question). So some poor will get hurt, and some poor will benefit. How to aggregate these. disparate effects? And aggregate we must, because if we say no to a policy reform even if a singlepoor person gets hurt, there wMl never be any policy reform. One way to pose the question is as follows: Are the gainers on average richer or poorer than the losers? If the. answer is that the gainers are on average poorer than the losers, that is an argument for supporting the reform. The same applies to any reform at all, not just trade policy reform. If  8 there are to be cuts in public expenditure, we have to tell the full story. What would happen without the cuts? How would the expenditure be financed? By higher taxes? Who would bear the burden of this? By increased fiscal deficit and thus higher inflation? Who would bear the burden of this? And so on. The welfare economics approach to evaluation policy reform is thus very clear. For each reform, or for the reform package as a whole, we first describe the consequences for gainers and losers, and then ascertain whether, on average, the gainers are richer or poorer than the losers. The calculus of welfare economics may not be of much use in assessing the political economy of policy reform. Here what matters is how gainers and losers form coalitions and use the institutions of politics to advance their cause. The outcome depends not only on the nature of the reform, but also on the nature of the political system. One feature - of many refonns, particularly in the trade area, is that the gainers from reforms are numerous but each gains relatively small amount, while the losers are small in number but each loses a large amount. The political outcome thus depends critically on the threshold at which a gain or a loss becomes sufficiently significant for an individual or a group for them to begin organising themselves. Typically, removal of tariffs benefits each of millions of consumers by a small amount (shirts are cheaper, say), but hurts each of a small number of producers by a lot (eg textile workers lose their jobs). There is then a built in bias against reform, because the intensity of the losers in support of their cause will be far greater. The political system in place is also important. Lester Thurow wrote a book about America some years ago in which he pointed out that the American political system essentially permitted  9 well organised interest groups, no matter how small, to block change. The conclusion he drew from this was that there had to be a national consensus on distribution on the basis of which change could be advanced--otherwise America would fall further and further behind. Thus the rules under which a blocking interest group can be overcome are also important in the political economy of the reform process. -3. Economic Policy Reform in Ghana, 1983-1993 With the above framework in mind, let us now tum to the Ghana story. The story is well known, and has been told many times before. But it may benefit by a retelling from even a slightly different perspective. The basic features of the Economic Recovery Program (;RP), launched in 1983, are by now well known. The effects on the macroeconomic performance, as measured by real GDP growth, are also well known. After a decade of real GDP declining at 1% per annum, in the decade following 1983 real GDP rose at 5% per annum. The first question to ask is-was it all simply due to the pouring in of aid? The answer is no. During this period Ghana suffered severe terms of trade losses, and the aid flows barely managed to compensate for this. The growth came from somewhere else. Another way to look at this is to do the following back of the envelope calculation. External assistance is around 8% of GDP. Even if as much as 50% of this went into investment (investment is around 10-15% of GDP in Ghana), and even if we take a relatively generous  10 value for the incremental capital output ratio of 2, this inflow would, on its own, account for only about 2 percentage points of the growth rate. Something else was clearly going on. That something else was a surge in public investment, and this seems to me to be the real story of the ERP decade. In 1983, government revenue (excluding aid) had fallen to 5% of GDP. Clearly, this is not enough to provide the basic infrastructure necessary in an economy (roads, electricity, water, telephones, etc), on which private investment relies. Not surprisingly, private investnent had itseli almost disappeared. In addition, investment in social infrastructure had also gpound to a halt. With no revenue of its own, the government couldn't provide the basic public goods. Jonathan Frimpong-Ansah has recently analysed the anatomy of this. decline. To use his. colorful phrase "the Vampire State" had sucked so much blood from its prey that there was no more to be had. To use another image, the state was lling the goose that laid the golden eggs. To use more prosaic economic language, the regime of controls ensured that most activity was in illegal, and therefore non-taxable channels. The most important of these controls were on the foreign exchange marketL When the black market rate is thirty times the official rate, as was the case in 1983, you don't need to be a genius to know that the govemment is losing huge amounts of revenue through customs duties. When there is an import licensing regime instead of a tariff regime, you know that revenue that should come to the government is going into the hands of the  officials and traders who are dividing up the scarcity value of the commodity between themselves. And so on. As the external sector was libiralized, government revenue shot up, as activity came back into taxable channels. The szrengthening of the revenue raising agencies also helped. By 1991, government revenue (excluding aid) was running at 15% of GDP. This is a remarkable transformation by any standards. Government expenditure, particularly investment in economic infrastructure, rose parl passu. Expenditure on the social sectors also increased. And all this at a time when inflation fel from three digits to 10% per annum. TO my mind, the increase in government expenditure on basic infrastructure is the key feature of the decade of the ERP. But it was made possible by other aspects of the Policy mix, in parficular, the external sector liberalization. There were also, of course, other features of the reform process, such as the move to ensure that cocoa farmers got a higher share of the world price of cocoa, which were also successful. There were yet other aspects which did not get as far,- such as the divestiture prgram or the program to restructure the public service. We will come to these presently. As noted in the previous section, there were bound to be gainers and losers out of the reform process. And there would be poor people amongst the gainers and the losers. What, taldng the welfare economics question first, would be the consequences for the poor of the ERP?. The first point to make is that this is not an easy question to answer with  12 certainty. We simply do not have the "before and after" data to provide a clean analysis. What we do have is a poverty profile of the country in the middle of this period on which to anchor corroborative evidence from different sources. The second point to make is that even if we had the best data in the world, we have a basic conceptual problem to do with the counterfactual. What is the ERP period being compared against? Against what would have happened if things would have gone on as in the previous decade? Or against some alternative policy mix that, it is claimed, would have been better? There is no simple answer to these questions. The basic fact to hang on to is that in the decade previous to the ERP real GDP in Ghana fell by 1% per annum. It would take a real stretch of the imagination to argue that this could have been good for the poor had it gone on. During the decade of the ERP, real GDP grew at 5% per annum. This is clearly better, but it is likely that this figure does not fully reflect what happened to the money in individual pockets because a large part of the increase went into pubic investment. In otiher words, national income may have gone up at 5% per annum, but national consumption did not, because public investment went up at a faster rate. Even so, it must be better than growth at minus 1 % per annum! Who were gainers and the losers from the foreign exchange liberalization? The main gainer was the govemment, which got revenue that was previously going out as rents to the black market. The main losers were clearly those who had privileged access to foreign exchange at the subsidized official rate. Are these likely to be the poorest of the poor?  13 Hardly. The gainers from government expenditure were those who benefit from basic public infrastructure such as better roads, more regular electricity and water supply, better telephone communications, as well as more expenditure in the social sectors. This would be a mixed bunch, consisting of the poor and not so poor. A word about the removal of subsidies, or rather, the introduction of user charges, which was another feature of the ERP. Who gains and who loses from a subsidy? Obviously, those who manage to have access to the subsidy gain, and the losers are those on whom the incidence of taxation to raise the money for this subsidy falls. The point about access to the subsidy, whether it be for fertliser or for drugs, is important. A general subsidy will reach the poor provided there is sufficient budget to back up the access for all those who demand it. When the budget isn't there, access will be rationed, and who gets a look in depends on the rationing mechanism that is adopted, implicitly or explictly. We can be fairly certain that the very poor will not get a look in. They do not have the contacts or whatever is needed to ensure access to that subsidized bag of fertiliser or those prescription drugs. The rich will always be alright-they can use their contacts or bribe their way into the rationing queue. Then there will be some groups of people such as the lower rungs of civil servants who may get access because of institutional features such as free drugs for the emnployees of the health ministry, etc. These are the groups who will lose as a result of the removal of the subsidy-they are not the very rich, but they are not very poor either. And they are organised; but we will come to the political economy presently. The macroeconomc data show that by 1983 there must have been widespread rationing of the so called  14 subsidies-since government revenue and hence expenditure had fallen to all time lows. In this context, recognising that the budget simply cannot support the range of subsidies that exist is an important first step. One way to fill the gap is cost recovery, unless expenditure is cut elsewhere, or revenue is raised. Even if revenue is raised, as it was in Ghana, the question still remains as to whether to spend this on subsidies or to invest it in basic economic infrastructure. These are difficult questions, but they must not be complicated further by the image of a 'golden age" where subsidies were indeed available to everybody, including the poor. Even if this was the case immediately after independence it was not sustainable, and in any case by 1983 it had ceased to be true. Despite the conceptual and data difficulties inherent in answering the question, the welfare economics evaluation of the ERP would, in my opinion, be pos.tive. A positive growth rate is better for the poor than a negative growth rate-at least, it would be a strange situation if the reverse was the case. And when we look at the components of the reform package, there are many elements that can be characterised as either being directly beneficial to the poor, or at least not harming them greatly. This is not to say that no poor people lost out, nor is it to say that everything is fine for the future. Rather we must learn from the lessons of the past decade to fashion the reform program of the next. What of the political economy of the reform process'? It would be too easy to simply attribute everthing to the fact there was a military govemment in Ghana during this period. As Jonathan Frimpong-Ansah shows in his analysis of the political economy of  15 decline from independence to 1983, and as Clark LCi'Lh and others have shown, a similar play of forces continued through military and democratic regimes. The basic feature was that the state kept squeezing the productive sectors to the point where they could not or would not yield up any more resources to grease the wheels of political compromise. It was this more than anything else, according to Frimpong-Ansah and others, that led to the turnaround. The economy had hit rock bottom, and there was only one way to go. This was the defining feature of 1983. 4.- The Second Decade of Reform in Ghana The first decade of the ERP, 1983-1993, was an undoubted success. But the nature of the success, and its limitations must also be borne in mind. As argued above, the ERP is actually a brilliant example of a public sector led recovery which has gone a long way towards restoring the basic infrastructure without which private investment cannot take off. However, the time is now ripe for the baton to be handed over to the private sector. The public sector cannot continue to grow at the rate at which it has grown during the ERP. Key reforms such as divestiture of state owned enterprises have not been undertaken. Moreover, the Public Service itself needs restructring and downsizing to focus on the core areas that the state should have responsibility for. Despite the fact that the growth rate during the ERP period averaged around 5% per annum, which was far superior to the minus 1% in the decade before, ftis level is  16 not good enough. With population growth running at 3% per annum, this will produce a maximum increase in consumption per capita of 2% per annum. At this rate, Ghana will take up to half a century to reach the ranks of the middle income countries. This is the sense in which the ERP growth performance, although brilliant, has to be bettered. A double digit growth rate is what is needed to make a significant dent in poverty in our lifetimes. Even if we take a more modest growth rate of 8% per annum as the target, this will require an increase in the investment ratio of 60%. In 1991 the ratio of investment to GDP was 16%. In other words, in order to achieve a growth target of 8% per annum investment will have to increase by 10 percentage points of GDP to 26%. Where will this increase come from? We can think of the sources as being foreign or domestic, and private or public. Starting with domestic/private, it has to be said that there is not much scope for dramatic increases in the household saving ratio, given the low level of incomes in Ghana. Typically, increases in household savings come after increases in household incomes, not before. Turning now to the foreign/public category, in other words aid, there is not much s.cope in the m,edium term for an increase in this-and it can be argued that aid dependence should in fact be reduced. This leaves the two remaining categories-domestic/public and foreign/private. The former is simply public sector savings, while the latter is foreign direct investment What is needed, therefore, is for the public sector to run a surplus of 5% of GDP, and for foreign direct investment to rise from its present derisory figure to 5% of GDP.  17 Both these are problematic. The 5 % public sector surplus will require more revenue raising but also cuts in public expenditure, at least as a share of GDP. The iatter cannot come from basic economic or social infrastructure. The selling off of loss malking enterprises can have a good budgetary as well as a good signalling effect. Downsizing and contracting out of many public services, and a hard look at many of Ghana's 150 or so "subvented organisations' is also called for. The rise in foreign direct investment pre-supposes that there exist the incentives to attract such investment, and that many of the features which previously frightened off such investment have been removed. Ghana has come a long way in this area, but much remains to be done. Divestiture will also help-there is nothing like the. suspicion created by a lingering state owned enterprise in a sector to hold back fresh investment, whether domestic or foreign. It must also be admitted that the political economy of a surge in foreign investment will have to be managed. A public education campaign and leadership will be needed to overcome public attitudes that owe much to the colonial legacy. Will there be demand for the products of the new investment? The key here is that Ghana, a small economy, has no option but to hook itself into the global market place by staying outward oriented and developing an aggressive strategy of promoting exports. The trade libalisaton of the ERP decade was important, as has been noted, in removing controls that otherwise drove economic activity onto non-taxable channels. But it was also important in giving Ghana an outward orientation. This should not now be reversed in the face of populist demands. However, much more attention should be devoted than was in the  18 ERP decade to the promotion of non-traditional exports. Institutional and other non-discriminatory support should be forthcoming. Growth by itself will not have as dramatic an effect on poverty reduction as if the growth was equitably distributed. The experience of fast growing countries that had a dramatic reduction in poverty, compared to equally fast growing economies that did not, suggests two differences. The former (primarily countries in East Asia) have kept an outward orientation and flexible labor markets that have allowed the fruits of growth to benefit the factor of production that the poor possess in abundance--their labor. Moreover, they have maintained a broad based pattern of public expenditure on basic health, basic education and basic infrastructure. Such a reorientation of public expenditure must be a high priority for Ghana in an era of restraint on government expenditure. -Using a broad sweep, we might perhaps characterise the sorts of reforms that have been or are in progress in Africa into two categories. Type I or Phase- I reforms include the following: - restoration of macroeconomic balanoe - establishing of realistic exchange rates - removal of quantitative controls, particularly in trade - - reduction of taxes on agriculture - rehabilitation of basic infIastructure Type It-or Phase II reforms include the following: private sector development;export promotion divestiture of state owned enterprises public sector restructuring and downsizing in some areas  19 financial sector rehabilitation and liberalisation reallocation of public investment and expenditure towards basic health, education and infrastructure Now the actual sequence followed in any one country may include a mix of these types of reforms. Moreover, the reforms are linked-we have already noted how exchange rate reform provided the revenue increases that underpinned the increased expenditure on rehabilitating economic infrastructure. However, it would be true to say that in a country like Ghana, which is perhaps the furthest along on the reform road, Type II reforms are still on the agenda, and they can therefore be described as Phase II reforms. Type I reforms have certain features worth noting. They are in many cases 'stroke of the pen' reforms. Changing the official exchange rate is a good example of a policy reform that can be done by the stroke of a pen, as is the lowering of taxes on agriculture, particularly when this is done through giving higher producer prices. Another feature of Type I reforms is that their benefits come through very quickly (there are costs too, of course, but my point here focuses on the benefits). Equalization of official and black market exchange rates leads to an overnight jump in government revenues. Raising producer prices affects cross-border smuggling immediately. More public expenditure on roads inproves their quality discernably straight away. Removing trade controls puts goods on shelves rapidly. And so on.  20 Type II reforms, on the other hand, are more institutional in nature. Many of them, such as divestiture, face legal and other hurdles. They take time. Reorienting public expenditure requires first that an adequate expenditure monitoring and control mechanism be put in place-otherwise budgets and actuals bear no relation to each other. At the same time, the costs of many of the Type II reforms come through almost immediately, while the benefits remain in some unquantified future. The costs of public sector downsizing are palpable-they will probably be demonstrating outside the door of the responsible minister the day aftr (o- the day before) the announcement. The same is true of layoffs brought about as a result of divestiture. Yet the benefits of divestiture in terms of the expected surge in investment and so on are always in the future, and difficult to quantify. This is very different from the revenue effects of exchange rate liberalisation- which can be forecast accurately and which come through quickly. I have referred to Type I and Type II reforms also as Phase I and Phase II reforms because, by and large, this is the sequence that African countries in general, and Ghana in particular, seem to be following. I have also argued that this is, broadly speaking, the right sequence to follow. The Welfare Economics evaluation of Phase I reforms for Ghana suggests strongly that, overall and on average, they have been a success. But it also suggests that Phase II reforms are needed because without them we will not get the growth with equity needed to make a significant dent in poverty within a lifetime. However, it should be clear that the political economy of Phase II reforms is different from that of Phase I reforms. Many of the costs of the former are upfront, while the benefits do not necessarily  - 21 come through quickly. The losers are thus likely to organise against the reforms, and the potential gainers are likely to be skeptical. Pushing for these reforms will look like a leap of a faith. Putting all this together, in the case of Ghana it is clear that the political economy of the reform process is likely to be very different in the second decade after the launch of ERP than it was in the first decade, for three reasons. First, 1983 was a rock bottom year; in 1993, thanks to the success of the first decade, the economy is no longer at rock bottom. It has been argued that coalition formation for reform would be much easier in the first set of circumstances. Second, in the decade starting in 1993 there is a regime switch from a military government to a constitutional govemment. The mere fact of such a regime switch will make the process of reform different-for example, until the institutions of the Constitution have settled down relative to each other, the passage of reforms will have to traverse uncharted waters. Third, as noted above, the nature of the reforms still left to be done is very different from the reforms undertaken in Phase I. Phase n reforms are such that the distribution and timing of costs and benefits are more likely to give an upper hand to the likely losers than to the likely winners. What all this means is that if there is to be an alignment of welfare economics and political economy in the second decade of reforms in Ghana,as there was in the first decade, the reformers will have to pay close attention to marshalling the forces of those liEly to benefit from the reforms in the long run, to convince them of this argument and  22 bring them into the national debate. Such public education is essendal because only with the implementation of the second phase of reforms can there be the surge of equitable growth that is needed to eradicate poverty within a generation. 5. Conclusion This essay has considered the sub-disciplines of Welfare Economics and Political Economy as two frameworks for evaluating policy reform and applied them to the case of Ghana. When the two sets of considerations are aligned, policy reform not only should be enacted, it is also likely to be enacted. Quite often, though, we do not have such an alignment, so that reforms which might improve social welfare do not get through. The agenda for analysts is to consider past reforms from these two perspectives in order to leam from history, and to evaluate proposed reforms in the same way, so that the feasibility and desirability of refonns can be assessed. The agenda for policy makers is to work towards oranizing and mobilizing the gainers from reforms which advance social wefare, so that resistance to such refonns from the losers can be overcome. The first decade of reform in Ghana presents the results of such alignment. The second decade of reform now awaits.  23 aBublio- n Alderman, H. (1991) Downturn and Economic Recovery in Ghana: Impacts on the Poor, Cornell Food and Nutrition Policy Program Atidnson & Stiglitz.(1980) Lectures on Public Economics, New York: McGraw Hill Bates, Robert H. (1981) Markets and States in Tdropical Aica: The Political Bias of Agricultural Policies, University of California Press Bates, Robert H. and Krueger, Anne 0. (eds) (1993) Political and Economic Interactions in Economic Policy Reform, Basil Blackwell Boateng, E. et-al (1990) "A Poverty Profile of Ghana, 1987-88, "Journal of African -economics, 199Z2- Bravennan, A. and Kanbur, R. (1987) "The Political Economy of Agricultural Price Reform,' World Development Findlay, R. (1990) "The New Political Economy: Its Explanatory Power for LDCs", Fcnomics and-Politics Frimpong-Ansah, J.H. (1991) The Vampire State in Africa: The Political Economy of Decline in Ghana, Africa. World Press Hicks, J.R. (1939) "The Foundations of Welfare Economics," Economic Journal Hilton, B. (1987) "Corn Laws" in J. Eatwell, M. Milgate and P.Newman (eds), The New Palgrave: A Dictionary of Economics, Macmillan Hollander, S. (1970) The Economics of David Ricardo  24 I.M.F. (1991) "Ghana: Adjustment and Growth, 1983-91," Occasional Paper. No. 86 Jeffries, R. (1989) "Ghana: The Political Economy of Personal Rule," in D.B.C. O'Brien. J. Dunn and R. Rathbone (eds) Contemporary West African States, Cambridge University Press. Kaldor, N. (1939) "Welfare Propositions in Economics and Interpersonal Comparisons of Utility," Economic Journal Kanbur, R. and Myles, G. (1990) 'The Politics of Policy Reform," European Journal of Political Economy Kanbur, R. (1987) "Measurement and Alleviation of Poverty" I.M.F. Staff Papers Kanbur, R. and Mink, S (1992) "Povorty and Accelerated Growth in Ghana: Policy Issues in Light of Global Experience' forthcoming in Bulletin of the Economic Society of Ghana Leith, J. Chlrk and Lofchie, Michael F. (1993); "The Political Economy of Structural Adjustment in Ghana,' in Bates and Krueger (eds) Pigou, A.C. (1920) The Economics of Welfte, London: Macmillan Ravallion, M. (1992) Poverty Comparisons: A Guide to Concepts and Methods," The World Bank, LSMS Working Paper Number 88 Robbins, L. (1938) "Interpersonal Comparisons of Utility: A Comment, 'Economic Joumal Sarris, Alexander H. (1993) "Household Welfare During Crisis and Adjustment in Ghana", Journal of African Economies Sen, A.K. (1984) Resources.Values and Development, Oxford: Basil Blackwell  25 Thurow, Lester (1980) The Z:er Sum Sociely, Basic Books Tullock, G. (1987) "Public Choice" in J. Eatwell, M. Milgate and P. Newman (eds) The New Paigrave: A Dictionary of Economics, Macmillan World Bank (1994) Adjustment in Africa: Reforms. Results and the Road Ahead, Oxford University Press World Bank (1993) Ghana 2000: Accelerated Growth with Povet Reduction World Bank (1990) World Development Report: Poverty World Bank (1991) Symposium Issue on Poverty and Adjustment: The World Bank Economic Review. May. 1991  Policy Research Working Paper Series Contact Title Author Date for paper WPS1358 Pattems of Behavior in Andrew Melrck September 1994 A. Maranon Biodiversity Preservation Martin L Weltzman 39074 WPS1 359 When Method Matters: Toward a Martin Ravalilort Seplember 1994 P. Cook Resolution of the Debate about Binayak Sen 33902 Bangladesh's Poverty Measures WPS1360 Are Portfoiro Flows to Emerging Sudarshan Gooptu September 1994 R. Vo Markets Complementary or 31047 Competitive? WPS1361 Extenmal Shocks and Performance F. Desmond McCarthy September 1994 M. Divino Responses during Systemic Chandrashekar Pant 33739 Transition: The Case of Ukraine Kangbin Zheng Giovanni Zanalda WPS1362 Regulation, institutions, and PabloT. Spiller October1994 B. Moore Commitment The Jamaican Ceziey l. Sampson 38526 Telecommunications Sector WPS1363 Brazil's Sugarcane Sector. A Case Brent Borrell October 1994 M. Bale of Lost Opportunity Jose R. Bianco 31913 Malcolm D. Bale WPS1364 Why Do Some Economies Adjust Ishrat Husain October 1994 J. Schwartz More Successfully Than Others? 32250 Lessons from Seven African Countries WPSI 365 The Macroeconomics of Adjustnent Ishrat Husain, October 1994 J. Schwartz in Sub-Saharan Afrcan Countries: 32250 Results and Lessons WPS1366 Distributive Concems When Salvador Valdes-Preto October 1994 E. Khine Replacing a PayAs-You-Go System 37471 with a Fully Funded System WPS1367 The Econormics of Cash Shortage Patrick Conway October 1994 L Suki 33974 WPS136B Sustained Inflation in Response to Patrick Conway October 1994 L Suki Price Lberalization 33974 WPS1369 Economic Policy Reforn, Philip L Brock October 1994 P. Sintim- Govemment Debt Guarantees, and Aboagye Financial Bailouts 38526 WPS1370 Is East Asia Less Open than North SumanaDhar October 1994 J. Ngaine America and the European Economic Arvind Panaganya 37959 Commumity? No  Policy Research Working Paper Serles Contact TItle Author Date for paper WPS1371 The Evolution of Trade Treaties and Sarath Rajapatirana October 1994 J. Troncoso Trade Creation: Lessons for Latin America 37826 WPS1372 Administrative Charges In Pensions Salvador Vald6s-Prieto October 1994 E. hline in Chile, Malaysia, Zambia, and the 37471 Unitrd States WPS1373 Finr 3ehavior and the Labor Market Simon Commander October 1994 B. Keller in t. H Hungarian Tmnsition Janos Kollo 35195 Cecilia Ugaz WPS1374 Infi -Iructure Finance: Issues, Anand Chandavarkar November 1994 M. Geller Ins-:.utions, and Polides 31393 WPS1375 Policy Lessons from a Simple Shantayanan Devarajan November1994 C. Jones Open-Economy Model Delfin S. Go 37699 Jeffrey D. Lewis Sherman Robinson Pekka Sinko WPS1376 How Trade, Aid, and Remittances Maurice Schiff November1994 J. Ngaine Affect Internaional Migration 37947 WPS1377 Macroeconomic Adjustment to Vittoro Corbo Novemnber1994 R.Vo Capital Inflows: Lafin American Style Leonardo HemAndez 31047 versus East Asian Style WPS1378 Mexico after the Debt Cris: Daniel Oks November1994 S. Harbi Is Growth Sustainable? Sweder van sijnbergen 37143 WPS1379 Financing Infrastructure in Developing Barry Eichengreen November'1994 WDR Countries: Lessons_from the Railway 31393 Age WPS1380 Transfers and the Transiion from Kathie Krumm November1994 K Krumm Socialism: Key Tradeoffs Branko Milanovic 34263 Michael Walton WPS1 381 Welfare Economics, Political Ravi Kanbur November1994 M. Youssef Economy and Policy Reform in Ghana 34614