C EPA CENTRE FOR POLICY ANALYSIS

GHANA SELECTED ECONOMIC ISSUES No. 20

2010

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY

Copyright © 2010 Centre for Policy Analysis No. 4 Prempeh II Street GIMPA Campus, Greenhill Accra Mailing Address: P. O. Box AN 19010 Accra-North Ghana Tel: (233-302) 420054, 450055, 420057, 420058 Fax: (233-302) 420056 ISSN: 0855-7144 The Centre for Policy Analysis is an independent, non-governmental think-tank, which provides rigorous analysis and perspectives on economic policy issues. Our objective is to: promote a non-partisan informed debate on macroeconomic, growth and poverty alleviation issues which are pertinent to the Ghanaian economy, to enhance the capacities of institutions in Ghana through training and finally to disseminate and publish economic information in order to raise public awareness of economic and developmental issues. The Issues Paper Series is part of CEPA’s publications and presents analysis of current economic issues relevant to Ghana’s developmental agenda. The Centre for Policy analysis is funded by the African Capacity Building Foundation (ACBF), Harare, Zimbabwe; and the Ghana Research and Advocacy Programme (G-RAP), North Ridge, Accra. For enquiries on publications email: [email protected] CEPA Website: www.cepa.org.gh

Table of Contents Introduction.....................................................................................................1 Dutch Disease.................................................................................................1 How does the Dutch Disease Occur?.............................................................2 Dutch Disease in Ghana.................................................................................8 Policy Proposals...........................................................................................17 References.....................................................................................................20

Table of Contents

Page i

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY Introduction As Ghana enters the oil and gas era there are high expectations about the stimulus that the exploitation of oil may provide for the rest of the economy. It provides some fiscal space while offering the resources and opportunity to address some of the structural constraints to investment and our development. Paradoxically, for a number of nonOECD countries such discoveries have not resulted in the relevant stimulus. As indicated elsewhere, the growth performance of such economies has not been shown to be significantly higher than the nonresource developing economies. One of the explanations given for this poor overall performance is the Dutch Disease.

Dutch Disease Dutch Disease refers to the adverse consequences of a large increase in a country’s wealth. It is the process by which a boom in a natural resource sector of an economy leads to the shrinkage of the nonresource tradable sector. The process may lead to the country specializing in the resource and non-tradable sectors. This makes the economy more dependent and vulnerable to the resource-specific shocks. “Although the disease is generally associated with a natural resource discovery, it can occur from any development that results in a large

Page 1

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY inflow of foreign currency, including a sharp surge in natural resource prices, foreign assistance, and foreign direct investment.”1

The Dutch Disease is part of the processes that are collectively referred to as the “resource curse” by which the discovery of a natural resource may lead to a decline in the performance of the economy.

How does the Dutch Disease Occur? There are two processes at work which may produce the Dutch Disease. Let us use the discovery of oil — but this may be applicable to an increase in natural resource prices in the world market, or a decrease in production costs, or a substantial increase in foreign aid — that makes the country a net oil exporter. The wealth and income of the country increase as a result of the oil discovery. More revenue is generated for government. As a result of the increased wealth and revenues, spending on non-tradables or services increases. Since the supply of services is not perfectly elastic, this spending effect results in a rise in the prices of services. Meanwhile, in a small open economy because the supply of tradables may be perfectly elastic their 1

Christine Ebrahim-Zadeh (2003): “Dutch Disease. Too much wealth managed unwisely”, Finance and Development, Vol. 40, No. 1 March

Page 2

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY prices may not change much. This effectively means that the real exchange rate, which is the relative price of non-tradables to tradables, appreciates.

The real exchange rate appreciation reduces the competitiveness of the country’s exports and domestic production of import-competing products. Thus, the real exchange rate appreciation will not only affect non-resource exports but also domestic production of importables. As the exchange rate appreciates, imports become relatively cheaper so that domestic producers are disadvantaged. This may lead to a drop in production in that sector as well as in non-oil exports.

The exchange rate is also an important economy-wide relative price signaling relative profitability of and the need for inter-sectoral resource transfers and factor movements. As a result of the change in relative price mentioned above, factors of production move from the non-resource sectors to resource- and non-tradable sectors which lead to an expansion in non-tradables and services, and to the shrinkage in tradables such as manufacturing and agriculture. At the same time, this resource allocation effect will bid up the price of labour while the price of capital remains the same. This may increase capital-intensity

Page 3

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY of production as employers substitute capital for labour, creating unemployment. The Dutch Disease occurs both as a result of the “spending effect” and the “resource allocation effect”. Similar analysis may be applied to a surge in foreign aid. In the case of aid, there may be a slight complication because the spending unit may be different from the unit responsible for the use of foreign exchange. In a framework developed by the Fund2, a distinction may be made between spending the resources from foreign aid and the absorption of the resources by the economy.

When foreign financial assistance is given to Ghana the foreign exchange is surrendered or lodged with the Bank of Ghana. Bank of Ghana then credits Ghana Government with the cedi-equivalent of the foreign aid. The government may then spend the cedis.

The foreign exchange can be accumulated by the Bank of Ghana as part of its international reserves or it can sell the foreign exchange to the public through the deposit money banks (DMBs) or the forex

2

International Monetary Fund (2005): “The macroeconomics of managing Increased aid inflows — Experiences of low-income countries and policy implications”, Policy Development and Review Department, IMF, Washington D.C.

Page 4

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY bureaux for their normal foreign exchange transactions. If the Bank of Ghana keeps the foreign exchange as reserves, the foreign exchange and aid is not absorbed by the economy even though the government would have spent the cedis. If in addition to the government spending, the foreign exchange is sold to the public, then the non-aid current account deficit will widen and the foreign aid is said to have been absorbed. Thus, the foreign aid is spent and absorbed. As the Bank of Ghana sells the foreign exchange, the real effective exchange rate will appreciate with adverse consequences for exports and producers of import-competing goods — the typical symptoms of the Dutch Disease.

The Dutch Disease framework can be applied to situations where there is substantial increase in world market prices for a country’s exports, decrease in costs of production, etc.

The Dutch Disease can exert a negative impact on growth, not only through the possible slump in the traditional export sector but also by discouraging the diversification of exports. The non-traditional exports are credited with higher income elasticities of demand, less volatile terms of trade (TOT) and higher prospects of dynamic productivity gains. The Dutch Disease can also negatively affect growth Page 5

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY when there are infant industries and learning-by-doing external to the firm in the shrinking tradables sector.

This phenomenon is called Dutch Disease because it was first observed in the Netherlands. The Netherlands discovered huge deposits of gas in the North Sea in the 1960. As it exploited and exported the gas and oil, the Dutch guilder appreciated, making Dutch non-oil exports less competitive, leading to de-industrialization of the Dutch economy.

This phenomenon generated interest in other countries where the framework was applied. In Australia the appreciation of the exchange rate accompanying the gold boom had the effect of making other sectors less competitive, especially the tradables goods sector. This phenomenon is referred to as the Gregory effect. A similar framework was applied in the United Kingdom (UK) where it is referred to as the Lawson effect. Some even refer to the Dutch Disease as the “resource curse”, even though the resource curse is a wider concept including the deterioration in governance, and the social and political governance that may accompany discoveries of large natural resource deposits.

Page 6

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY Dutch Disease problems were experienced in several oil exporting developing countries in the 1970s. Similar experiences were recorded for coffee exporters following high export prices also in the 1970s. Some economists3 have argued that Dutch Disease effects may not be a major explanation for the poor performance of resource exporting non-OECD countries.

The Dutch Disease is a resource re-allocation process away from less profitable tradables to non-tradables sectors, and by so doing it is efficient and welfare-improving. Where the expected wealth or inflows are perceived to be permanent, then the adjustment may reflect a change to the new structure of the economy. However, even in this situation there are transitional costs to be borne to assist the lagging sectors, and employees adjust to the new system and increased vulnerability of the economy to the booming sector.

3

Paul Collier (2007): “Managing Commodity Booms: Lessons of International Experience”, Paper prepared for the African Economic Research Consortium (AERC), January 2007, Centre for the Study of African Economies, Department of Economics, Oxford University

Page 7

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY Dutch Disease in Ghana In Ghana, the application of the Dutch Disease phenomenon was first made in the 1990s. In “Aid and the Dutch Disease: Macroeconomic Management when Everybody Loves You”, Stephen Younger argued that the effects of a booming public sector and conditionalities were to raise interest rates and create a certain type of Dutch Disease through crowding-out of the private sector. Other discussions of the Dutch Disease in Ghana relate to the aid surge after 2000. Macroeconomic management of substantial increases in aid flows could generate Dutch Disease effects. Increased aid and private remittances represent a resource boom and like all resource booms could lead to concerns about resource curse. The fundamental question is whether the aid upsurge experienced in Ghana had Dutch Disease effects, appreciating the real exchange rate and undermining the growth of exports — especially non-traditional exports.

Three recent studies have analyzed the Ghanaian situation with respect to aid, aid management and the possible consequences in terms of the Dutch Disease. All the studies agree that 2001-2003 was a period where Ghana experienced an aid surge. All the studies also agree that for the period 2001-2003 increased aid to Ghana was neither spent nor Page 8

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY absorbed. However, for 2004-2007, aid was both spent and absorbed. Elbadawi and Kaltani4 observed that “the high aid flows did not lead to real exchange rate appreciation until 2005 and only since 2006 do we see some mild overvaluation cropping up”. CEPA and TIPCEE5 conclude that over the past five years “the management of aid and exchange rate policies has given room for concern about Dutch Disease effects. This may arise in part from the composition of expenditure from aid and the use of the exchange rate as a nominal anchor within the context of disinflation policy and now inflation targeting”.

It must be observed that for the first aid-surge period the aid was neither absorbed nor spent. In part this must have been the result of the need to accumulate reserves as buffer against external shocks as well as to avoid Dutch Disease problems. The inflow of aid was perceived as transitory. Moreover, the recent shift in the use of aid

4

Elbadawi, Ibrahim and Linda Kaltani (2007): “Scaling-up aid for Ghana: Maintaining Competitiveness, Avoiding the Dutch Disease, and Accelerating Growth in Ghana”, CEM Technical Review Growth Workshop on Meeting the Challenge of Accelerated and Shared Growth, May 2-3, 2007 5 CEPA and TIPCEE (2007) Study on the Appropriate Exchange Rate Regime for a Competitive Export-led Growth Strategy for Ghana, USAID/TIPCEE and Centre for Policy Analysis, Accra

Page 9

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY raises concerns. There is an increasing preoccupation with using aid directly for poverty reduction and meeting the Millennium Development Goals (MDGs) of the United Nations. As indicated earlier, using aid for direct productive investment activities promotes growth while the humanitarian spending, while important, may not enhance growth. One implication of the shift is that non-tradables became the more favoured output, increasing the danger of the Dutch Disease effects.

Moreover, of the four countries (Ghana, Ethiopia, Tanzania and Mozambique) studied by the IMF6 and Foster and Killick,7 there was no evidence of aid being channeled to the private sector, or it being utilized within the public sector for the explicit purpose of promoting private sector development.

The challenge posed by the Dutch Disease is how to migrate from a position of low productivity and limited innovation in an environment in which strong “natural resource” sector and accompanying services

6

Ibid, IMF (2005): “The Macroeconomics of Managing Increased Aid Inflows: Experiences of Low-income Countries and Policy Implications”, Policy Development and Review Department, IMF, Washington D.C. 7 Foster, Mick and Tony Killick (2006): “What Would Doubling Aid do for Macroeconomic Management in Africa?” ODI Working Paper, Vol. 264, April, Overseas Development Institute, London

Page 10

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY establish a high real exchange rate (RER) and an effective floor on wages. The dominant exports — cocoa, gold and timber — have not lent themselves to substantive downstream value-added activity. They have supported a relatively high real exchange rate and, with associated service sectors, contribute to the establishment of a floor for wage levels, thus hindering the development of the manufacturing sector. The economy has changed very little in the last two decades with the dominant traditional exports — cocoa, gold and timber — also unchanged and contributing about the same proportion of total exports earnings. To that extent they have provided Ghana with a foretaste of the Dutch Disease. The main sector that may be affected will be manufacturing and agriculture.

Manufacturing The evidence is that Ghanaian enterprises may not be competitive in international markets. Data from the Ghana Enterprise Survey suggest that only a small number of manufacturing firms have succeeded in developing new exports. Ghana is ranked 30th in 42 countries of subSaharan Africa (SSA) in respect of rate of growth of total exports over the 2000-2006 period. Page 11

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY

In the 2009 Competitiveness Report of the World Economic Forum (WEF), Ghana was shown as relatively weak in related areas of technology, innovation, education and labour market efficiency. These findings suggest that skills and productivity lie at the heart of the competitiveness challenge facing Ghana.

Ghanaian enterprises are judged to be less productive than those of comparator countries in SSA mainly because they are less capital and skill intensive and relatively small in size. Ghanaian enterprises are also primarily oriented toward the domestic market. Productivity, however, is not a differentiating characteristic since those enterprises that export do not have above average productivity rates.

There is limited scope for gradualism. Support for agriculture and rural enterprises will be important to sustain employment. Ghana’s poor infrastructure emerged as by far the dominant perceived barrier to development in the most recent Enterprise Survey — some 49 percent of enterprises unreliable electricity supply as the bigger obstacle to growth. Access to water is also frequently cited as an important constraint. Page 12

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY

Ghanaian enterprises have less machinery and equipment per worker than those in comparator countries. The median Ghanaian enterprise has about US$ 1200 worth of machinery and equipment per worker. The same comparator values are Nigeria (US$2,600), South Africa (US$16,000) and Malaysia (US$26,000).

Ghanaian enterprises are less likely to export less than a quarter of Ghanaian manufacturing enterprises export compared to more than half for comparator countries. Since the international market represents the “frontier in efficiency and productivity” this suggests that labor costs are not low enough to compensate for low labour productivity: The value-added per worker in China, Thailand, Kenya, and Swaziland is between six and seven times that for Ghana. Malaysian worker produces about 14 times as much and the South African over 25 times as much. Low costs will not compensate for low productivity in Ghana. Consequently, an accommodating business regulatory environment and free market with factors of production — land, labour, etc. — will not be sufficient to lift Ghana to the higher productivity platform necessary to be competitive as a resource-rich economy. Moreover, the wage floor established by the services sector Page 13

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY will continue to be central on the urban areas encouraging migration. Ultimately, Ghana must migrate to a position of higher value-added production (and service delivery) through investment in education and technology.

Labour productivity generally tends to be low in small-sized enterprises and Ghanaian enterprises are small. The median Ghanaian manufacturing enterprise has only 10 employees compared to Kenya (50) and Swaziland (60). Teal et al8 suggests that there has been a dramatic shift in Ghana over the past decade with small and potentially informal enterprises being increasingly important.

Labour market policies, such as minimum wages, statutory redundancy pay, etc, have an obvious bearing on the labour market, setting the rules of the game. The fact that hiring workers formally results in high costs make many Ghanaian employers hire causal labour. Ghana has not had the competitiveness to attract internationally mobile labourintensive industries such as garments, footwear or electronic assembly.

8

Teal et al (2006):

Page 14

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY With such low levels of productivity, an appreciating exchange rate with a wage floor could worsen the situation.

Agricultural Competitiveness The international experience shows that agricultural sector competitiveness is often hurt by the discovery of a natural resource. And yet the performance of the agricultural sector will largely determine the national benefit from the discovery of oil in Ghana.

An estimated two-thirds of the enterprises in Ghanaian manufacturing reportedly depend on agricultural inputs. Consequently, an uncompetitive and stagnating agricultural sector would undermine the competitiveness of the manufacturing sector.

Moreover, as the World Development Report (2008) of the World Bank emphasized, inclusive agricultural growth is particularly effective in reducing poverty particularly in situations like Ghana’s where large numbers of the poor are in farming. Finally, Ghana’s oil boom is, on current estimates of reserves and production schedules, likely to be short-lived. Subsequent future growth will once more be dependent on agriculture. The international Page 15

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY experience points to the fact that once market share is lost it can be extremely difficult to regain. This may be due to the loss of commodity-specific capital — both physical and human (such as scientific knowledge and technical skills). In export markets where supply chains are often complex and difficult to establish this is a common occurrence. In the Ghanaian context this problem can be seen in the continuing struggle to recover market share in the European pineapple market. Slow conversion to new varieties demanded by European supermarkets caused the loss of market share to the Latin American producers.

The agricultural sector could suffer the consequence of the Dutch Disease if the there is a loss in competitiveness of the sector on account of any of the following:  The ‘spending effect’ — an increase in the price of non-tradable goods and services due to an expansion of demand fuelled by oil revenues. This can be on account of public and/or private demand expansion;  The ‘resource movement effect’ — upward pressure on domestic production costs as factors of production (such as labour) are attracted by higher remuneration into the booming oil sector; and

Page 16

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY  An appreciation of the exchange rate — both the real exchange rate (the relative price of non-tradable to tradable goods) and the nominal exchange rate — because of increased oil revenues leading to a fall in the price that domestic producers receive for exports and for products competing with imports. [Competing imports become cheaper resulting in loss of domestic market share; a loss of competitiveness in international trade would lead to a decline in Ghana’s exports. On the positive side, increased price of urbanization that would come with the oil boom and increasing incomes can be expected to increase urban preferences for processed foods and foods with greater domestic value-added. So, provided the tradable sector remains competitive, this would be an opportunity for both agriculture and the food processing manufacturing sector].

Policy Proposals In theory, there are two proposals for dealing with and/or avoiding Dutch Disease effects. These are fiscal and structural.

On the fiscal side, mitigating the Dutch Disease effects involves examining the composition of expenditures and decreasing the degree of spending on non-tradable services. The easiest way is to reduce

Page 17

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY spending from the windfall income and invest the savings abroad, or alternatively, direct spending towards import-intensive expenditures.

In view of the initial conditions prevailing in Ghana with respect to the levels of poverty, infrastructural gaps, and lagging productivities in agriculture and manufacturing, and fiscal gaps including arrears of payments, such an approach may be inappropriate.

Instead, investments of savings in reducing the structural constraints to growth may be more appropriate. Investment in infrastructure combined with short-term arrears clearance, and tax incentives may help provide support for the private sector in terms of reducing their transactions costs, access to credit facilities, and retained profits.

Depending on the size of the windfall, there may be room to restructure taxes to eliminate distortionary taxation. This feeds into the second proposal dealing with structural-side policies.

The workings of markets are important in terms of the extent to which the resource allocation effect will impact on factor prices. Policies to open up and increase competition in factor markets — labour and

Page 18

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY capital — may help offset some of the impact of resource price shocks. As discussed in a another paper9, Ghana may need programmes and projects to support agriculture and manufacturing against the Dutch Disease, A spatial development equalization programme may also be desirable to stem the possible migratory movements and inequality that may occur with Dutch Disease effects. Modeling results for Ghana “suggest that typical Dutch Disease effects of the oil boom negatively affect export agriculture and rural households. In all scenarios, agriculture and rural areas either get hurt or tend to have smaller positive effects relative to the non-agricultural sectors and urban areas. Smoothing in oil revenue allocations into the fiscal budget by saving part of the inflow in an oil fund is not only preferable for growth and stability, but also for equality, since the negative short-run effects on agriculture are significantly moderated. In addition, using oil revenues to increase agricultural productivity and raise

9

See CEPA (2010): Ghana: Managing An Oil Economy, Centre for Policy Analysis, August 2010

Page 19

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY competitiveness has the potential to offset the negative impacts from the Dutch Disease”10.

References CEPA (2010): “Managing an Oil Economy: Ghana”, Centre for Policy Analysis, Accra, Ghana CEPA and TIPCEE (2007): Study on the Appropriate Exchange Rate Regime for a Competitive Export-led Growth Strategy for Ghana, USAID/TIPCEE and Centre for Policy Analysis, Accra Clemens Breisinger, Xinshen Diao, Rainer Schweickert, and Manfred Wiebelt (2009): “Managing future oil revenues in Ghana – An assessment of alternative allocation options”, Kiel Working Papers, No. 1518, May, Kiel Institute for the World Economy Collier, Paul (2007): “Managing Commodity Booms: Lessons of International Experience”, Paper prepared for the African Economic Research Consortium (AERC), January 2007, Centre for the Study of African Economies, Department of Economics, Oxford University Ebrahim-Zadeh, Christine (2003): “Dutch Disease: Too Much Wealth Managed Unwisely”, Finance and Development, Vol. 40, No.1 (March) Elbadawi, Ibrahim and Linda Kaltani (2007) Scaling-up Aid for Ghana: Maintaining Competitiveness, Avoiding the Dutch Disease, 10

Clemens Breisinger, Xinshen Diao, Rainer Schweickert, and Manfred Wiebelt (2009): “Managing future oil revenues in Ghana – An assessment of alternative allocation options”, Kiel Working Papers, No. 1518, May, Kiel Institute for the World Economy

Page 20

GHANA: THE DUTCH DISEASE IN AN EMERGING OIL ECONOMY and Accelerating Growth, Ghana CEM Technical Review Growth Workshop on Meeting the Challenge of Accelerated and Shared Growth, May 2-3 Foster, Mick and Tony Killick (2006): “What Would Doubling Aid do for Macroeconomic Management in Africa? ODI Working Paper 264 (April), Overseas Development Institute, London International Monetary Fund (2005): “The Macroeconomics of Managing Increased Aid Inflows: Experiences of Low-Income Countries and Policy Implications”, Policy Development and Review Department, IMF, Washington D.C. Younger, Stephen D. (1992): Aid and the Dutch Disease: Macroeconomic management when everybody loves you, Cornell University, Ithaca, New York, USA

Page 21