Cape Verde
Recent Economic Developments
Author:
International Monetary Fund
Search for other papers by International Monetary Fund in
Current site
Google Scholar Close
Publication Date:
15 Oct 1996
eISBN:
9781451809237
Language:
English
Keywords:
ISCR; CR; state enterprise; U.S. dollar; exchange rate; foreign currency; cape Verdean authorities; substitution effect; base period; salary increase; special deposit
Download PDF (5.6 MB)

This paper describes economic developments in Cape Verde during 1990–96. After a noticeable slowdown during 1989–91, the Cape Verdean economy rebounded in 1992 and continued to expand in 1993–95, prompted by a sustained increase in public spending and the positive effects of a liberalization of economic activities. Real GDP growth averaged 4 percent during 1992–95, but domestic, economic, and financial developments were characterized by increasing imbalances. Total government expenditure increased rapidly, reaching on average 60.1 percent of GDP in 1994–95, far outpacing any growth in domestic revenues and external grants.

Abstract

This paper describes economic developments in Cape Verde during 1990–96. After a noticeable slowdown during 1989–91, the Cape Verdean economy rebounded in 1992 and continued to expand in 1993–95, prompted by a sustained increase in public spending and the positive effects of a liberalization of economic activities. Real GDP growth averaged 4 percent during 1992–95, but domestic, economic, and financial developments were characterized by increasing imbalances. Total government expenditure increased rapidly, reaching on average 60.1 percent of GDP in 1994–95, far outpacing any growth in domestic revenues and external grants.

Cape Verde: Basic Data, 1991–95

article image
Sources: Data provided by the Cape Verdean authorities; and Fund staff estimates.

Estimates after 1992.

Excluding public enterprise transfers.

Change as percent of broad money stock at end of previous period.

Includes credit to central and local government.

Ratio of GDP to broad money stock at end of period.

Including arrears.

I. Recent Economic Developments

1. Background

Cape Verde is an archipelago of 15 volcanic islands located some 620 kilometers off the westernmost coast of Africa. Nine of the islands are populated by some 390,000 residents; however, it is estimated that almost twice as many Cape Verdean emigrants live abroad, in the United States, the Netherlands, France, Portugal, and Italy. The emigrants’ close ties with the homeland are evidenced by significant remittances, which averaged some 20 percent of GDP during the period 1992-95.

Cape Verde faces severe development constraints: a Sahelian climate, few natural resources, and costly internal communications. Agriculture represents 11 percent of GDP, is highly vulnerable to the vagaries of the climate, and in the best years provides some 20 percent of the local population food requirements; most of the food deficit has been traditionally covered by aid from foreign donors. Cape Verde’s surrounding territorial waters are unsuitable for large-scale commercial fishing because of the absence of a significant continental shelf. Manufacturing production, primarily constituted by food processing and energy generation for domestic consumption and more recently by apparel processing for export, has risen to represent some 9.9 percent of GDP. Civil construction and services constitute over three fourths of GDP. The strategic location of the country, at the intersection of the routes between South Africa and Europe, South Africa and North America, South America and Europe, gives Cape Verde an advantage for international sea and air transport services, which, together with workers’ remittances, constitute the major source of export earnings.

At independence in 1975, the Government of Cape Verde adopted a centrally planned development strategy and followed mostly prudent policies that did not hamper external competitiveness. In 1991, the market-oriented Movement for Democracy won the first multiparty legislative and presidential elections, and the economic strategy changed markedly. The new Government’s 1992-95 economic program sought to raise the rate of economic growth by reducing the size of the public sector, encouraging the development of private sector activity, and opening up the economy to take advantage of the potential for rapid export growth. In the event, this policy resulted in widening financial imbalances. 1/

2. Overall developments

After a noticeable slowdown during 1989-91, the Cape Verdean economy rebounded in 1992 and continued to expand in 1993-95, prompted by a sustained increase in public spending and the positive effects of a liberalization of economic activities (Charts 1 and 2). Real GDP growth averaged 4.0 percent during 1992-95, but domestic, economic, and financial developments were characterized by increasing imbalances. Total government expenditure increased rapidly, reaching on average 60.1 percent of GDP in 1994-95, far outpacing any growth in domestic revenues and external grants. The overall public deficit increased from 8.9 percent of GDP in 1991 to an average 15.2 percent of GDP in 1994-95 and was financed with substantial domestic and external debt borrowing.

CAPE VERDE: REAL SECTOR DEVELOPMENTS, 1991–95

Chart 1.
Chart 1.

GDP GROWTH

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

CAPE VERDE: REAL SECTOR DEVELOPMENTS, 1991–95

Chart 2.
Chart 2.

DOMESTIC EXPENDITURE

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

The reform of the financial sector introduced elements of competitiveness in the banking sector, and with the introduction of treasury securities, improved the profitability of the banking system. Monetary policy was however primarily geared to meeting the financing requirements of the Government. The growth of public sector expenditure was much faster than that of overall GDP. Private consumption remained broadly constant as a share of GDP and increased somewhat per capita, but private investments declined markedly (Table 1). As a result of these developments, the external resource gap widened from 30.9 percent of GDP in 1991 to 48.2 percent in 1995.

Table 1.

Cape Verde: Sources and Uses of Resources, 1991–95

article image
Sources: Cape Verdean authorities; and Fund staff estimates.

Strong aggregate demand pressure and the liberalization of most merchandise transactions during 1991-92, resulted in a considerable deterioration of the external situation of the country. Given the small open economy nature of Cape Verde, the authorities chose to maintain a nominal exchange rate peg for controlling inflation. However, the combination of a pegged exchange rate and fiscal laxity resulted in a massive surge in import demand that far outpaced the increase in exports, service income, and transfers. The current account deficit widened from 2.7 percent of GDP in 1991 to 14.2 percent in 1994 and 1995. Sizable increases in project grants, net direct investments, and net loan disbursements were not sufficient to prevent a large drawdown of foreign exchange reserves. Gross domestic savings, which were already negative, declined from -2.0 percent of GDP in 1991 to -12.3 percent of GDP in 1995. National savings also declined, but remained strongly positive, reflecting the large and rising level of unrequited transfers.

3. Real sector developments

a. Gross domestic product

During the period 1992-95, the real GDP growth rate averaged some 4.0 percent, enhanced by a sustained increase in the public investment program and by positive effects of structural reforms initiated in 1992 (Tables 2-3, and Chart 2). 1/ Official data for 1992 and preliminary sectoral data for 1993-95 from primary production, industry and energy, commerce, construction, housing and public services point to a steady recovery of growth performance (Tables 4-7). In this period, the real growth rate was led primarily by a sustained increase in the public investment program, which promoted growth in transport and communication, as well as energy and construction. 2/ The liberalization of most imports in 1992 [see Section 3c(i)] contributed to an even faster growth in commerce, the largest sector of the economy. The reform of the Institute for Housing Development (Instituto de Fomento de Habitação - IFH) into a specialized financial agency prompted a housing boom in 1992-93, which was accompanied by a parallel demand for mortgages. Agriculture production, however, continued to fall: the negative trend of 1990-91 persisted in 1992, a partial recovery followed in 1993, but a severe drought in 1994 reduced most crop yields to their lowest recorded levels, and a partial recovery followed in 1995.

Table 2.

Cape Verde: Gross Domestic Product by Major Sector at Current Prices, 1991–95

article image
Sources: General Directorate of Statistics (DGE), Ministry of Economic Coordination; and Fund staff estimates.
Table 3.

Cape Verde: Gross Domestic Product by Major Sector at Constant 1980 Prices, 1991–95

article image
Sources: General Directorate of Statistics (DGE), Ministry of Economic Coordination; and Fund staff estimates.
Table 4.

Cape Verde: Production in Agriculture, Livestock, and Fishing, 1991–95

(In metric tons)

article image
Source: National Commission for the Assessment of the Agricultural Year.
Table 5.

Cape Verde: Industrial Production, 1991–95

article image
Sources: Boletím Trimestral de Estatística; General Directorate of Statistics (DGE), Ministry of Economic Coordination, and General Directorate of Fishing (DGP), Ministry of Fisheries, Agriculture, and Rural Development.
Table 6.

Cape Verde: Production and Consumption of Electricity and Water, 1991–95

article image
Source: Data provided by ELECTRA.
Table 7.

Cape Verde: Consumption of Petroleum Products, 1991–95

(In metric tons)

article image
Source: Data provided by the Cape Verdean authorities, ENACOL, and Shell Oil Company of Cape Verde.

CAPE VERDE: REAL SECTOR DEVELOPMENTS, 1991–95

Chart 3.
Chart 3.

GDP BY SECTOR

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

b. Prices

Inflation, measured by the consumer price index (CPI) of the Central Bank of Cape Verde (Banco de Cabo Verde - BCV) averaged some 7.2 percent in the period 1992-95, a relatively positive result in light of the poor agriculture outturns of the later years (Table 9). 3/ This is explained to a large extent by the trade liberalization of 1992-93 in the context of a broadly stable nominal exchange rate. 4/ After three years (1992-94) of moderate and steadily decelerating inflation, exogenous shocks and weak financial policies contributed to a rise in inflation from 3.5 percent in 1994 to 8.4 percent in 1995. The upsurge of CPI in 1995, measured by the General Directorate of Statistics (Direcção Geral de Estatistica - DGE) (Table 10), 1/ is primarily ascribed to a combination of the poor 1994 domestic crop and the higher international price of food imports. Prices increased further in 1996, when the administratively set prices of the most relevant food staples were revised (corn, rice, sugar, edible oil), the price of cement was liberalized, and import quotas on 26 products were introduced (Decrees 3-8/96 of March 1, 1996).

Table 8.

Cape Verde: Public Investment Expenditure, 1991–95

(In millions of Cape Verde escudos)

article image
Sources: General Directorate of Planning (DGP), Ministry of Economic Coordination; and Fund staff estimates.

Estimated as a residual.

Table 9.

Cape Verde: Consumer Price Flash Index, 1991–96

article image
Source: Bank of Cape Verde.

First five months of 1996 over corresponding 1995 period.

Accumulated in the first five months of 1996.

Table 10.

Cape Verde: Consumer Price Index, 1992–96 1/

article image
Source: General Directorate of Statistics (DGE), Ministry of Economic Coordination.

Monthly data available from January 1992.

First four months of 1996 over corresponding 1995 period.

Accumulated in the first four months of 1996.

c. Employment and wages

Unemployment remained stable, at about 25 percent of the active labor force during the period 1992-95. In the private sector, wages remained constant in real terms, but the differential with the much higher wages in the public sector continued to constitute an element of tension during contract renewals.

Based on the census of 1990, published by the Ministry of Labor, Youth and Social Promotion in February 1996, the economically active population was 122,000, or 53 percent of the resident population above the age of 10. Of these, 91,000 were employed, almost equally distributed between the primary, secondary, and tertiary sectors. The unemployed constituted just about 25 percent of the active labor force; more than two thirds of them were seekers of first-time employment.

Employment statistics for later years cover only the two islands (São Tiago and São Vicente) on which the main urban centers are located. In 1994, unemployment in Praia had reached 33 percent, compared with 22 percent in 1990, while in São Vicente it had reached 43 percent, compared with 35 percent four years earlier. 2/

As part of the appraisal of the Public Sector Reform and Capacity Building Project, in August 1993, the authorities documented 10,900 civil servants directly employed in Public Administration. However, the number of civil servants increased to 11,900 in 1995 (Table 11), as the authorities delayed the implementation of the civil service retrenchment program, which was a component of the public sector reform supported by a World Bank credit.

Table 11.

Cape Verde: Civil Service Employment by Administrative Unit, 1992–95

article image
Sources: General Directorate of Public Administration (DGAP), Ministry of Civil Service and Parliamentary Affairs.

Includes fixed-term, provisional, and other contractual staff.

August 1993 was the reference month for the civil service survey underlying the Voluntary Departure Program (PAV). Inconsistencies in the survey report data suggest that the total number of staff might be undereported by as many as 1,000 staff.

Until 1993, rural development was part of the Ministry of the Sea.

Includes the former Ministry of Finance and Planning.

Estimated as a residual.

According to the Bureau for the Restructuring of the State Enterprise Sector (Gabinete de Apoio à Reestruturação do Sector Empresarial do Estado GARSEE), the number of employees of public and mixed public enterprises was estimated to be about 6,000 in 1995, down from 6,400 in 1990.

Public sector salaries increased in real terms during the period 1991-94, before declining somewhat in 1995 (Table 12). Private sector salaries did not rise in tandem with public sector salaries and therefore the differential between salaries in the two sectors constituted an element of friction during private sector contract negotiations (Table 13). Based on the 1990 census, the average salary in the public administration and in public enterprises was some 20-30 percent higher than the average salary paid in private corporations. This disparity widened further in 1990-92 and this gap was maintained during 1992-94, when private sector salaries, particularly in the liberalized sectors, increased in line with inflation (Chart 4). Also notable was the widening wedge between top and bottom salary levels in the public administration, where the ratio between highly qualified and unskilled workers’ salaries grew from 3.5 in 1991 to 4.5 in 1995; comparable ratios in the private sector remained more or less unchanged in the tradable sector and decreased in the nontradable sector. 1/

Table 12.

Cape Verde: Basic Monthly Salaries of the Civil Service by Grade, 1991–95

(In Cape Verde escudos)

article image
Source: Data provided by the Cape Verdean authorities.

A wage increase of about 17 percent was granted in 1991 (Law Decree No. 101/III/90).

In 1992, a wage adjustment took place according to the civil service plan on positions, career, and salaries (Law Decree No. 86/92).

In 1993, salaries were increased to compensate for income taxation of civil servants.

As of September 1994; includes 5 percent salary increase (Law Decree No. 8/94).

As of January 1995; includes 5 percent salary increase (Law Decree No. 5/95).

Table 13.

Cape Verde: Average Daily Salary by Occupation, 1991–95

(In Cape Verde escudos)

article image
Sources: General Directorate of Labor and Employment (DGTE), Ministry of Justice and Labor.

CAPE VERDE: REAL SECTOR DEVELOPMENTS, 1991–95

Chart 4.
Chart 4.

REAL SALARIES BY SECTOR

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

4. Fiscal sector developments

a. Fiscal trends

During the period 1992-95, as a result of the implementation of the first stages of their tax reform directed to streamline the tax system and boost domestic revenues, a favorable result was achieved in increasing government revenue, which rose from 17.6 percent of GDP in 1991 to 26.1 percent in 1995, one of the highest ratios of domestic tax revenues to GDP in Africa (Table 14). However, this was more than offset by a substantial expansion of current outlays and public investments, and total budgetary expenditure rose from 37.5 percent of GDP in 1991 to an average of 60.1 percent in 1994-95. As a result, the overall deficit (commitment basis), which had averaged some 8 percent of GDP in 1992-93, doubled to 15.9 percent in 1994; it contracted somewhat in 1995, to 14.5 percent, reflecting a reduction in capital expenditure. The current balance, which had been positive between 1992 and 1994, turned into a deficit of 2.7 percent of GDP in 1995.

Table 14.

Cape Verde: Operation of the Central Government, 1991–96

article image
Sources: Ministry of Finance and Planning Ministry of Economic Coordination; Bank of Cape Verde; and Fund staff estimates.

Based on 1996 draft budget.

Includes the so–called Contes de Ordem, which are revenues, as budgeted, from the direct provision of services by government agencies offset by the same amounts of current expenditure for each of these agencies.

These are public enterprises’ shares of government investment costs, usually in infrastructure directly related to the activity of these enterprises.

Net of amortization.

Foreign financing averaged 4.4 percent of GDP during 1992-95 (Chart 5), and domestic financing 7.3 percent of GDP. Within domestic financing, privatization receipts and domestic nonbank financing covered a share of the deficit equivalent to 3.0 percent and 6.0 percent of GDP in 1994 and 1995, respectively; domestic bank financing made up the remainder (7.5 percent of GDP in 1994 and 4.3 percent in 1995), and primarily took the form of purchase of treasury bills. The stock of domestic debt rapidly escalated to 42.2 percent of GDP at the end of 1995. The growing domestic interest obligations introduced an additional element of inflexibility in the budget.

CAPE VERDE: FISCAL SECTOR DEVELOPMENTS, 1991–95

Chart 5.
Chart 5.

FINANCING

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

b. Domestic revenue and tax reform

The first stage of the tax reform, implemented in 1991, included measures to: (i) modernize the tax administration; (ii) simplify the tax structure and improve the revenue collection on imports; (iii) extend the coverage of the income tax; and (iv) increase the rate of some other taxes. 1/ The successful implementation of these measures allowed fiscal revenues to increase from the equivalent of 17.6 percent in 1991 to 23.7 percent of GDP in 1993 and further to 26.1 percent of GDP in 1995, thus reversing the decline in budgetary revenues previously recorded in the 1980s and the early 1990s (Tables 15 and 16, and Chart 6). The increase was primarily achieved by a strong performance of import tax revenues--which doubled between 1991 and 1995--and of tax revenues on business profits, which rose threefold, and of nontax revenues, which doubled.

Table 15.

Cape Verde: Central Government Revenue, 1991–96

(In millions of Cape Verde escudos)

article image
Sources: General Directorate of Contributions and Taxes (DGCI), Ministry of Economic Coordination.

Based on the 1996 draft budget.

In 1996 the unified income tax replaced the tax on business profits, the tax on salaries and self–employment tax, and the complementary tax. The revenues budgeted for these latter taxes for 1996 concer entitled credit arrears.

In 1993 the tax on the fuel distributor’s profits and capital distributions was replaced by the tax on petroleum products.

Collected by the customs departments on imports.

On government–guaranteed external borrowing.

These are the so–called Contas de Ordem, which are revenues, as budgeted from the direct provision of services by government agencies offset by the same amounts of current expenditure for each of these agencies.

Table 16.

Cape Verde: Selected Indicators of Central Government Revenue, 1991–96

article image
Sources: General Directorate of Contributions and Taxes (DGCI), Ministry of Economic Coordination.

Based on 1996 draft budget.

These are the so–called Contas de Ordem, which are revenues, as budgeted from the direct provision of services by government agencies offset by the same amounts of current expenditure for each of these agencies.

CAPE VERDE: FISCAL SECTOR DEVELOPMENTS, 1991–95

Chart 6.
Chart 6.

REVENUE

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

In the 1996 budget, the authorities began the implementation of the second stage of the tax reform, introducing the unified income tax (Decree Law 1/96), which replaced the tax on salaries and self-employment income, the tax on business profits, and the complementary tax; the expected effect on revenues should be neutral. Under the economic program for 1996-99 and with a view to improving the efficiency of the tax system, the authorities are committed to introducing the value-added tax, which will replace the consumption tax and, possibly, other indirect taxes.

A summary of Cape Verde’s tax system in 1995 is provided in Appendix I.

c. Expenditure developments

Government expenditure climbed from 37.5 percent of GDP in 1991 to 60.8 percent in 1994 and was then contained at 59.4 percent in 1995. The expansion of public capital expenditure was accommodated by a significant increase in both foreign and domestic financing.

(1) Current expenditure

Current expenditure rose by some 9.2 percent of GDP during 1992-95, with about half of this increase recorded in 1995, when it reached the equivalent of 28.8 percent of GDP (Tables 17 and 18, and Chart 7). Most of this growth is due to the 1992-93 reclassification of jobs and pay scales of civil servants, which brought about a structural increase in the wage bill of 2.3 percent of GDP. In addition, in 1994-95, the decentralization of several administrative functions to local authorities resulted in a structural increase in transfers of an additional 2.1 percent of GDP. Finally, in 1995, the reform of public sector financing and the increasing recourse to treasury bills pushed up interest payments on domestic public debt from negligible levels in 1994 to 2.9 percent of GDP in 1995.

Table 17.

Cape Verde: Economic Classification of Central Government Expenditure, 1991–96

article image
Sources: General Directorate of the Budget (DGO), Ministry of Economic Coordination.

Based on the 1996 draft budget.

These are the to called Contas de Ordem which are revenue, as budgeted, from the direct provision of services by government agencies offset by the same amounts in current expenditure for each of these agencies.

Table 18.

Cape Verde: Central Government Current Expenditure by Administrative Unit, 1991–96

article image
Sources: General Directorate of the Budget (DGO), Ministry of Economic Coordination; and Fund Staff estimates.

Based on the 1996 draft budget.

Until 1993, rural development was part of the Ministry of the Sea.

Includes the former Ministry of Finance and Planning.

CAPE VERDE: FISCAL SECTOR DEVELOPMENTS, 1991–95

Chart 7.
Chart 7.

CURRENT EXPENDITURE

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

Public Sector Reform

In January 1994, the authorities obtained an IDA credit to reform the public sector (Public Sector Reform and Capacity Building Project).

The principal objective has been to transform and modernize key public institutions. The project has five major components: (i) modernizing the civil service, including the financing of a public expenditure review (PER), and redeploying those civil servants who participate in the Voluntary Departure Program (Programe de Abandono Voluntario - PAV); (ii) strengthening municipal authorities and decentralizing public administration functions; (iii) strengthening economic management capacity, and creating the Institute of Statistics; (iv) modernizing the civil codes and reforming the judicial system; and (v) reforming public procurement.

The implementation of this project has been mixed: (i) the PER component suffered delays and no separations took place under the PAV; (ii) municipal strengthening efforts are under way; (iii) the creation of the Institute of Statistics is expected for 1997 but reducing the backlog in the processing of national accounts has been difficult; (iv) the reform of the civil codes and judiciary system are progressing rapidly; and (v) the public procurement system was updated.

In the context of the 1996-99 economic program, the authorities expressed their commitment to use this project to improve the effectiveness and efficiency of public administration. The authorities are pursuing the following objectives: improving the budget methodology; programming the development of the public debt; creating a secondary market for treasury bills as a precondition for the development of a capital market; consolidating the budgetary execution; developing a fiscal information system; improving and modernizing the Treasury payment system; streamlining controls of budgetary execution; improving the efficiency of the acquisition of goods and services; integrating budget and financial services information; making customs services more effective and efficient; and improving personnel management.

(i) Delays in the implementation of the civil service reform

Since 1993, the weight of the public administration wage and salary bill has remained stable, around 12.1 percent of GDP. Concurrent with the reclassification of jobs and pay scales of the civil servants, the authorities launched a Program of Voluntary Departure (Programe de Abandono Voluntário - PAV) directed toward reducing the number of civil servants by some 2,000 from 10,900: 400 through early retirement and the remaining 1,600 through voluntary departure. 1/ The PAV envisioned two incentive packages: early retirement for civil servants at 60 years of age or with 35 years of service; and compensatory separation payoffs.

The PAV was delayed because of the Government’s desire to achieve social consensus before implementing the more sensitive measures of the program. The early retirement of some 400 civil servants was achieved in 1994 but, to date, no separations have taken place under the PAV, although the overwhelming majority (77 percent) of the civil servants do not have a permanent contract (Table 10). Instead, by the end of 1995, the total staff of the public administration had increased by some 9 percent to 11,900 employees, notwithstanding a freeze on new hiring of nonqualified personnel. This was due, in part, to the delayed resignation of many civil servants, who are waiting for the separation package to be in effect, and, in part, to the hiring of new teachers to meet the education needs of the growing population.

(ii) Fiscal impact of the administrative decentralization

The process of administrative decentralization initiated by the authorities in 1992 brought an increase in transfers to the local authorities, from 2.2 percent of GDP in 1993 to 4.5 percent in 1995.

The areas of functional responsibility delegated to the 15 municipalities encompass: collecting municipal taxes; providing education services; issuing transport and trade licenses; and contracting municipal construction. To carry out these responsibilities, the municipalities received from the central authority the proceeds of four taxes (the estate and gifts tax, the tax on real estate transfers, the tax on rentals, and the tax on motor vehicles), equivalent to 4.5 percent of GDP in 1995. In addition, municipalities collected their own taxes and were allowed to borrow from institutions, including commercial banks, without treasury guarantee.

(2) Capital expenditure

The sharp decline in capital expenditure of the 1980s was reversed in 1991, and a further expansion took place in 1992-95. 1/ Capital expenditure rose from 17.9 percent of GDP in 1991 to 36.6 percent in 1994, and was somewhat contained at 30.5 percent of GDP in 1995 (Chart 8). External financing of public investment almost equally divided between loans and project grants, averaged slightly less than one third of total public investment expenditure. Furthermore, a relevant part of the domestic financing came from counterpart funds originating from external assistance in the form of food aid.

CAPE VERDE: FISCAL SECTOR DEVELOPMENTS, 1991–95

Chart 8.
Chart 8.

CAPITAL EXPENDITURE

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

During the 1990s, the PIP process presented significant weaknesses because of the lack of financial consistency among the planned increase in government investment, the availability of external project financing, and the inadequate provisions for recurrent costs. To correct these weaknesses, a rolling three-year PIP was developed in 1994 with the financial and technical assistance of the World Bank; this process however was not continued in 1995; and no consistent investment planning and management was thereafter pursued.

d. Deficit financing and public debt

The substantial increase in the overall public sector deficit (commitment basis) from 8.7 percent of GDP in 1992 to 14.5 percent in 1995 was covered by a rising recourse to domestic financing, which rose from 55.9 percent of total financing in 1992 (4.9 percent of GDP) to 71.2 percent in 1995 (10.3 percent of GDP).

In July 1993, a first issue of long-term public securities (Obrigações do Tesouro - OT; C.V. Esc 500 million) was placed at a 10 percent yearly interest rate. In 1994, OT amounting to C.V. Esc 6,927 million were issued, partially in connection with the clearing of a large portion (C.V. Esc 2,017 million) of nonperforming loans to public enterprises in the BCV’s portfolio. In November 1994, Decree Law 63/94 introduced three-month treasury bills and Decree Law 64/94 introduced new treasury bonds (Nova Série), medium- and long-term instruments with maturities longer than 18 months but not exceeding 30 years, which became the primary instruments of domestic government financing. Much of the old Government’s debt with the financial institutions was transformed into remunerated treasury securities. Since November 1994, the treasury bills have been sold through auctions, held at least once a month, with the participation of commercial banks and other financial institutions. The Central Bank, BCV, does not interfere with the functioning of the auctions.

By the end of 1995, the stock of outstanding public debt held by the financial system amounted to C.V. Esc 13,587 million, equivalent to 42.2 percent of GDP (Table 19). The BCV held 35.7 percent of the outstanding debt, exclusively long term and, for the most part, contracted at zero or below-market fixed interest rates. The commercial banks, Banco Commercial Atlántico and Caixa Económica de Cabo Verde, held an additional 53.0 percent of the public debt, and the long-term financial institutions, Instituto Nacional de Previdência Social (INPS) and GARANTIA, the remaining 11.3 percent. The government securities held by the two commercial banks carry an average interest rate of about 8.5 percent.

Table 19.

Cape Verde: Profile of the Domestic Public Debt of the Central Government, 1991–96

article image
Sources: Ministry of Economic Coordination, Bank of Cape Verde; and Fund staff estimates.

Istituto Nacional Previdencia Social (I.N.P.S.).

The debt of 1994 was contracted in connection with the clearing of non–performing public enterprises’ debt held by the Bank of Cape Verde.

The authorities have expressed concern that the structure of the budget is becoming increasingly inflexible as a result of the growing domestic debt burden. In 1995, domestic interest obligations amounted to C.V. Esc 934 million, equivalent to 2.8 percent of GDP. In 1996, interest obligations are projected to rise to 3.5 percent of GDP. In addition, an amount of outstanding debt equivalent to 6.0 percent of GDP will reach maturity.

5. Financial sector developments

a. Policy objective, monetary and credit trends

In view of the small open nature of the Cape Verdean economy, the presence of capital controls on private sector external transactions, and the choice of an exchange rate peg as the nominal anchor during the period 1992-95, the main objective of monetary policy was to control the external overall balance through the growth of net domestic assets of the banking system. Within the overall domestic asset target, the distribution between credit to the Government and to the private sector was a major element for economic policy instrument; credit to the private sector was controlled through loan ceilings for each commercial bank. 1/

During the period 1992-95, broad money grew at an average rate of 14.9 percent a year, well above the average 9.1 percent growth of nominal GDP (Table 21 and Chart 9). 2/ The shares of domestic and foreign currency deposits remained more or less constant, with the foreign currency component never accounting for more than 5.2 percent of broad money. Within the domestic component of deposits, the share of time deposits decreased, primarily on account of the minimum size for new deposits required by the largest commercial bank (BCA) in 1993 (Chart 10).

Table 20.

Cape Verde: Selected Financial Indicators of the State Enterprise Sector, 1991–95

(In millions of Cape Verde escudos)

article image
Source: Bureau for Support in the Restructuring of the State Enterprises Sector (GARSEE), Ministry of Economic Coordination.

Restructured, partially privatized and completing the privatization process.

Partially privatized to foreign partner.

Under restructuring.

Fully privatized.

Liquidated.

Public capital enterprises are enterprises whose ownership is not formally defined as public, private or mixed. The State’s claim on most of these enterprises arises from cross participations by other state or mixed enterprises.

Other public capital enterprises for which accounts are not available are CABETUR, the Cape Verdean Warehouse, Creola, Hotel Atlántico, Hotel 5 de Julho, and ULTRA.

Other mixed enterprises for which accounts are not available are SCC, ALUPAST, Armisticio, CENTROTEL, IBC, INTERCAP, INPHARMA Laboratories, and PETRONAVE.

Table 21.

Cape Verde: Summary of the Banking System, 1991–96 1/

article image
Sources: Bank of Cape Verde; and Fund staff estimates.

Includes Bank of Cape Verde, Banco Comercial do Atlántico and Caixs Económica da Cabo Verde.

Estimates.

Includes credit to central and local governments.

Since December 1989, data includes payments by the Bank of Cape Verde for external debt service due and not paid by some nonfinancial public enterprises.(NFPEs)

Includes mixed enterprises.

GDP/broad money stock.

CAPE VERDE: FINANCIAL SECTOR DEVELOPMENTS, 1991–95

Chart 9.
Chart 9.

BANKING ASSETS

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

CAPE VERDE: FINANCIAL SECTOR DEVELOPMENTS, 1991–95

Chart 10.
Chart 10.

BANKING LIABILITIES

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

CAPE VERDE: FINANCIAL SECTOR DEVELOPMENTS, 1991–95

Chart 11.
Chart 11.

DOMESTIC PUBLIC DEBT

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

CAPE VERDE: FINANCIAL SECTOR DEVELOPMENTS, 1991–95

Chart 12.
Chart 12.

PRIMARY SUBSCRIBERS

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

In 1992-93, broad money rose by 13.8 and 15.9 percent a year, respectively, while growth in net domestic assets was 6.9 percent and 12.1 percent, respectively, a year and net foreign assets increased. 1/ This trend was reversed in 1994 when broad money grew by less than domestic assets; claims on the Government rose sharply and claims on the economy declined, as about 80 percent of the banking system exposure to the parastatal sector was taken over by the Government. 2/

In 1995, broad money rose by 18.1 percent and the ratio of GDP to broad money (average velocity) fell to 1.3. The BCA exceeded its credit ceiling and, as a result, in 1995 credit to the private sector increased by 12.5 percent, after modest increases between 1992 and 1994, and net domestic assets of the banking system rose by 21.4 percent. In 1995, the net foreign asset position of the banking system registered a decline of C.V. Esc 728.7 million, 3.3 percent of broad money, equivalent to US$4.6 million. After falling to a historical low in November, gross reserves of the Central Bank increased by some C.V. Esc 1,550 million following the sale of 40 percent of the telecommunications company to foreign investors. As a result, over the year, the Central Bank external reserves rose over their previous year value by 31.3 percent in escudo terms; although they remained considerably below the peak reached in 1993 and amounted to only about 2.5 months of imports (Table 22).

Table 22.

Cape Verde: Balance sheet of the Bank of cape Verde, 1991–96.

article image
Source: Bank of Caps Verde.

Since December 1989, data include payments by the Bank, of Cape Verde for external debt services due and not paid by some nonfinancial public enterprises.

b. The structure of the financial system

After independence, the Cape Verdean financial system was composed of the Banco de Cabo Verde (BCV), which provided central and commercial banking, and the Instituto de Seguros e Previdência Social (ISPS), which provided social security and insurance services. In 1991, the ISPS was split into an autonomous social security agency (Instituto Nacional de Previdência Social - INPS) and a private insurance company (GARANTIA). Also in 1991, a joint venture between Portuguese and local investors created a second insurance company, the Companhia Caboverdiana de Seguros Impar (Impar). In September 1993, the BCV spun off its commercial activities to a new universal bank, the Banco Comercial do Atlântico (BCA), and the post office savings bank was transformed into a universal bank, the Caixa Económica de Cabo Verde (CECV). Therefore, in 1995, the financial system of Cape Verde consisted of: (i) the Central Bank, BCV; (ii) two commercial banks: BCA and CECV; (iii) the social security institute, INPS; and (iv) two insurance companies: Garantia and Impar. All the companies were publicly owned, with the exception of Impar. The main constraints these institutions faced were the lack of standardized accounting practices, the absence of an effective supervisory framework, and a heavy dependence on financing from the Treasury.

Capacity Building in the Financial Sector

In March 1996, the authorities obtained an IDA credit to support private sector institutions and strengthen financial sector reforms (Capacity-Building Project for Private Sector Promotion).

In the financial sector, the strategy includes the following components: (i) strengthening the role of the BCV; (ii) promoting a competitive banking industry; (iii) broadening private ownership of financial institutions; (iv) deepening financial intermediation; and (v) reforming the insurance company.

The project is slated to finance technical support to modernize the infrastructure for financial sector operators, including: the design and development of an effective payments system; the review and update of the legal framework for the financial sector; and assistance to the BCV to strengthen its supervisory function and redesign and implement an integrated economic and financial data base. The project will also finance technical assistance to BCA, CECV, INPS and ISCV (Instituto de Seguros de Cabo Verde), including training, audits, and management information systems. Assistance to born commercial banks will be provided to prepare them for privatization.

(1) Banco de Cabo Verde

In 1995, BCV was responsible for conducting monetary policy, supervising banks, and managing the foreign exchange reserves. 1/ The ceiling on Government borrowing from the BCV was limited by law to 10 percent of the previous year’s revenues.

The institutional framework regulating the BCV was established in 1990. 2/ In 1995, however, a draft proposal was put forward for discussion to allow the BCV to operate more independently of the Ministério de Coordination Económica (MCE) and potential political pressures. This proposal focused on strengthening: (i) the independence of formulating and executing monetary policy (Art. 13-16); (ii) the effectiveness of the supervision office (Art. 22); (iii) the financial autonomy of the BCV vis-à-vis the Government’s financial requirements (Art. 26); 1/ and (iv) the length of the mandate of the BCV Executive Board. 2/

(2) Banco Comercial do Atlântico

In 1995, BCA was the largest commercial bank in Cape Verde. Its assets amounted to C.V. Esc 22,840 million (70.8 percent of GDP) and its capital C.V. Esc 1,728 million. The BCA accounted for 86 percent of total deposits in the banking system.

The BCA has a relatively weak management information system, and reports financial data to the Central Bank with a three-month delay, a feature that hinders effective supervision from the BCV. 3/ In particular, the financial information provided by the BCA is not adequate to evaluate the share of the nonperforming assets in its portfolio.

Many of the financial problems faced by BCA were related to the unfavorable balance sheet structure--characterized by a reserve ratio of more than 47.5 percent--inherited from the BCV in September 1993 (Table 23). 4/ The BCA inherited some C.V. Esc 16 billion of deposit liabilities from the BCV and some C.V. Esc 17 billion of assets, of which C.V. Esc 9 billion was in nonremunerated excess reserves with the BCV. Of its liabilities, half was in interest-bearing deposits, remunerated at mandatory rates of 8.5 percent and 12.0 percent. The bank was subject to credit ceilings, allowing only one third of its deposits to be on-lent. As a result, the annual funding cost of the BCA exceeded the interest income on its loan portfolio. As any increase in liabilities would have further aggravated the excess liquidity position, between 1993 and the first half of 1995, the BCA discouraged the opening of new accounts and increased the minimum amount required to open a new demand deposit account.

Table 23.

Cape Verde: Balance Sheet of the Banco Comercial do Atlantico, 1993–96

(In millions of Cape Verde escudos: end of period)

article image
Sources: Banco Comercial do Atlántico (BCA); and Fund staff estimates.

To address this excess liquidity and promote the BCA’s recapitalization, the authorities issued more than C.V. Esc 6 billion of treasury instruments in 1994, enabling the BCA to reduce its unremunerated deposits with the BCV by some C.V. Esc 4 billion. As a result, in 1994, the BCA’s reserve ratio decreased to 18.8 percent, but remained in excess of the legal 12.5 percent reserve requirement.

During 1995, the BCA’s net claims on the government remained stable, but its claims on the private sector rose sharply, from C.V. Esc 5 billion to C.V. Esc 7.5 billion, 1/ and the reserve ratio decreased to 17.8 percent, just above the 17 percent mandatory reserve requirement. 2/ While the capital asset ratio increased from 4.8 percent in 1993 to 7.6 percent in 1995, the rate of profitability fell from 12.9 percent in 1993 to 4.4 percent in 1995. 3/

(3) Caixa Económica de Cabo Verde

In 1995, the CECV accounted for 14 percent of total deposits in the banking system. It provided monthly financial statements for BCV’s supervision in an accurate and timely fashion. The ratio of private sector loans to deposit was 62.5 percent at the end of 1995; some 12 percent of its total assets were past due. Its liquidity position was in line with the legal requirements, the cash to demand deposits ratio was 5.8 percent, and the capital to assets ratio was 7.6 percent (Table 24).

Table 24.

Cape Verde: Balance Sheet of the caíxa Económica, 1991–96

(In millions of Cape Verde escudos; end of period)

article image
Source: Caíxa Económica de Cabo Verde (CECV).

Bank of Cape Verde (BCV).

Banco Comercial do Atlántico (BCA).

(4) Instituto Nacional de Previdência Social

In 1995, the INPS was the country’s second largest financial institution with total assets of C.V. Esc 4,325 million, 13.4 percent of GDP. The INPS provided social security coverage (retirement benefits, disability pension, and complementary benefits) to some 23,000 private and public enterprises workers, who contributed 23 percent of their earnings. In 1995, the INPS was not subject to specific reserve, investment, insurance, or risk-management requirements nor did the Secretary of Labor, the supervisory authority, have the technical capability to provide effective supervision.

Given that the number of contributors exceeded the number of retirees by a ratio of 11:1, assets were equivalent to 810 percent of liabilities and the liquid disposable income was equivalent to 170 percent of total liabilities. 4/ Between 1983 and 1991, the INPS primarily invested in equity in public enterprises (such as CERIS, Hotelmar, and CVC); and from 1991, it invested primarily in deposit accounts, treasury bills, real estate, and in participations in financial companies (such as CECV and Garantia). With C.V. Esc 3,430 million in liquid assets (10.6 percent of GDP), the INPS has the potential for becoming a significant force in an expanded financial sector.

(5) Companhia Garantia de Seguros

In 1995, Garantia controlled about two-thirds of the insurance market. It was primarily involved in mandatory motor vehicle coverage (C.V. Esc 119 million in earnings) and maritime risk insurance (C.V. Esc 100 million). Its capital was C.V. Esc 254 million and its assets amounted to C.V. Esc 892 million, with a capital asset ratio of 28.5 percent. Assets were primarily held in commercial bank deposits (C.V. Esc 283 million), real estate (C.V. Esc 85 million), treasury bills (C.V. Esc 55 million), and equity participation (C.V. Esc 54 million).

(6) Companhia Caboverdiana de Seguros Impar

Founded in 1991 as a joint venture between Portuguese and local investors, Impar was the only private enterprise operating in the Cape Verdean financial system in 1995. Its main activities included optional motor vehicle insurance (C.V. Esc 47 million), commodity insurance (C.V. Esc 42 million), and mandatory motor vehicle insurance (C.V. Esc 26 million). Impar recorded its first year of profits in 1995; after accounting for its past losses, the capital of the enterprise represented C.V. Esc 175 million and its assets amounted to C.V. Esc 340 million, for a capital asset ratio of 51.5 percent.

c. Current regulatory and supervisory framework

In 1995, the Cape Verdean financial institutions were required to report to different entities: the commercial banks to the BCV, the social security agency to the Secretary of Labor, and the insurance companies to the ISCV. However, none of the three supervisory institutes had the technical capability to carry out its function effectively. In particular, the BCV had been unable to retain adequate personnel trained for this purpose and thus supervision was weak; for example, an overrun of the credit ceiling by the BCA in 1995 was detected only with substantial delay.

In April 1996, the authorities took measures to correct this situation with the appointment of a new Director for Supervision at the BCV, and with the centralization of all the supervisory functions at the Central Bank. 1/

d. Monetary policy instruments, interest rates, and treasury securities

In 1995, the BCV began phasing out the use of direct instruments of monetary policy--credit and interest rate ceilings--and began relying more on indirect instruments--reserve requirements, rediscount window, and primary market sale of treasury bills. This process was undertaken gradually and pragmatically by the authorities in light of the limits of the Cape Verdean financial system.

Bank-specific annual credit ceilings remain the primary instrument on which the BCV relies to implement monetary-policy. In January 1996, however, the BCV abolished sector-by-sector credit ceilings, prompting a reshuffle of credit among the sectors and the expansion of real estate credit (Table 25).

Table 25.

Cape Verde: Credit to the Economy by Sectors, 1994–95

(flow)

article image
Source: Bank of Cape Verde (BCV).

Banco Comercial do Atlántico (BCA).

Caíxa Económica de Cabo Verde (CECV).

In April 1996, the BCV reduced the last administratively set rate on six-month deposits from 8.5 percent to 8.0 percent (Table 26) and indicated that this rate would be liberalized shortly, prompting renewed interest by commercial banks in attracting deposits. 1/ Despite the liberalization of interest rates, competitiveness remains limited both on the deposit and on the lending side. On the deposit side, the BCA is the oligopolistic price setter; on the lending side, the existence of credit ceilings constituted the relevant factor that contributed to the limited competitiveness. The treasury bill market is segment in the financial market where competitive forces are more evident. Since November 1994 the interest rate at the time auctions has fluctuated between 8 and 9 percent; a development of the secondary market is also envisaged.

Table 26.

Cape Verde: Current Interest Rate Structure

(In percent)

article image
Source: Bank of Cape Verde.

Up to 180 days.

For deposits with a maturity in excess of one year.

These accounts will yield only 5.5 percent if an early withdrawl occurs.

Rates applicable to six–month and one–year deposits at the Banco Comercial do Atlántico (BCA) in the week of September 20–27, 1994.

In July 1995, the BCV increased the commercial banks’ reserve requirements on domestic and foreign assets from 12.5 percent to 17 percent. As of December 1995, neither commercial bank had had recourse to the rediscount window, owing to the persistence of excess liquidity. This excess liquidity was reduced with the increase of the reserve requirement to 20 percent in February 1996.

With the introduction of treasury securities, the share of central government borrowing from the banking system rose from 32.4 percent of total assets in 1993 to 52.1 percent in 1995 (Table 27). About one fourth of this increase was due to deficit financing, while the rest was related to the above-mentioned operation concerning the Government’s assumption of the BCV’s nonperforming loans with public enterprises.

Table 27.

Cape Verde: Credit by Maturity and Institutional Sector, 1991–95 1/

article image
Source: Bank of Cape Verde.

As of September 1993, data include Banco Comercial do Atlántico (BCA) and Caíxa Económica de Cabo Verde (CECV).

Includes local government.

Before 1992, data do not include payments by the Bank of Cape Verde for external debt service due and not paid by some nonfinancial public enterprises.

Includes credit to mixed enterprises.

6. Exchange rate policy and external sector developments

a. External sector trends

Following the trade liberalization of 1992 and 1993, the expansionary financial policies of 1992-95 led to four consecutive years of sharp increases in the trade deficit, from 38.8 percent of GDP in 1991 to 53.4 percent of GDP in 1995 (Table 28, and Chart 13). The viability of the external sector became increasingly dependent on emigrants’ remittances, which increased from 16.5 percent of GDP in 1991 to a record 22.7 percent of GDP in 1995, and on foreign grants, which fluctuated around an average 12.3 percent of GDP (Chart 14).

Table 28.

Cape Verde: Balance of Payments, 1991–95

article image
Sources: Bank of Cape Verde; and Fund staff estimates.

As from 1993, international transactions of the Banco Comercial do Atlântico are recorded above the line.

In 1991, arrears to OPEC and BADEA of US$2.58 millions and US$4.99 million, respectively, were rescheduled.

In 1991, arrears to BADEA of US$1.44 millions were refinanced.

CAPE VERDE: EXTERNAL SECTOR DEVELOPMENTS, 1991–95

Chart 13.
Chart 13.

REER AND TERMS OF TRADE

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

CAPE VERDE: EXTERNAL SECTOR DEVELOPMENTS, 1991–95

Chart 14.
Chart 14.

EXPORTS AND IMPORTS

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

Despite the substantial increases in project grants, in net direct foreign investment, and in net loan disbursements, the overall balance of payments turned negative in 1993, and reached a record US$20.4 millions deficit in 1994 (Table 29, and Chart 15). The overall balance returned to a surplus in 1995 of US$7.5 million, reflecting US$20 millions in privatization proceeds from the sale of 40 percent of the national telecommunications company to foreign interests.

Table 29.

Cape Verde: Balance of Payments, 1991–95

article image
Sources: Bank of Cape Verde; and Fund staff estimates.

As from 1993, international transactions of the Banco Comercial do Atlántico are recorded above the line.

In 1991, arrears to OPEC and BADEA of US$2.58 million and US$4.99 million, respectively, were rescheduled.

In 1991, arrears to BADEA of US$1.44 million were refinanced.

CAPE VERDE: EXTERNAL SECTOR DEVELOPMENTS, 1991–95

Chart 15.
Chart 15.

CURRENT ACCOUNT

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

Gross official reserves declined from the equivalent of 5.2 months of imports of goods and services in 1991 to 2.5 months of imports of goods and services in 1995; during the same period, and the external debt rose from US$147 million in 1991 (43.0 percent of GDP) to US$189 million in 1995 (45.5 percent of GDP), of which US$19 million represented external arrears.

b. Exchange rate policy

Since 1982, the authorities maintained a peg of the C.V. escudo to a basket of nine currencies of the countries (other than the United States) that are the most important sources of imports and emigrants’ remittances. 1/

This exchange rate arrangement was to provide a nominal anchor so as to control inflation, which fell from an average rate of 5.2 percent in 1992 to 3.5 percent in 1994. The supply shocks of 1994-95, however, led to an increase in inflation in 1995 to 8.4 percent. This brought about an appreciation of the real effective exchange rate in 1995 of 2.3 percent over its 1994 level (Table 30, and Chart 16; see also Chapter II).

Table 30.

Cape Verde: Effective Exchange Rates, 1991–96

article image
Source: IMF, International Financial Statistics and Information Notice System.

First four months of 1996 over corresponding 1995 period.

CAPE VERDE: EXTERNAL SECTOR DEVELOPMENTS, 1991–95

Chart 16.
Chart 16.

CAPITAL AND FINANCIAL ACCOUNTS

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

c. Current account

(1) Effects of the trade liberalization

After a deterioration of the terms of trade in 1992 (Table 31), exports of goods increased in value by 27.7 percent in 1994 and by an additional 62.5 percent in 1995 (Table 32), primarily as a result of the export promotion efforts and foreign investment policies. Given the small base, however, exports accounted for only 2 percent of GDP in 1995 (Chart 13). Footwear, apparel and frozen fish became Cape Verde’s main exports after the establishment of several light manufacturing factories in the Free Zone of Mindelo. 1/ Among the primary sector exports, lobsters accounted for 15.0 percent of the export value, but bananas, which constituted 39.5 percent of the export value in 1991, accounted for less than 1.0 percent of total export value in 1995, reflecting large cost increases.

Table 31.

Cape Verde: External Trade Indices, 1991–95

(1987 = 100)

article image
Sources: Bank of Cape Verde; and Fund staff estimates.

Does not include the item “Other exports.”

Calculated as Paasche index with moving weights.

Residually calculated from value and price indices.

Indices are based on WEO estimates of main import prices.

Table 32.

Cape Verde: Merchandise Exports, 1991–95

(Value in million of Cape Verde escudos; volume in metric tons; and unit value in Cape Verde escudos per metric ton)

article image
Source: Bank of Cape Verde.

Traditionally, Cape Verde has relied on food aid for some 80 percent of its food requirements, and imported all raw materials, intermediate goods, and capital goods. In order to promote export, in 1991 the authorities simplified the structure of taxes on imports, including import duties and consumption taxes on imported goods, and in 1992 and 1993 liberalized the imports of most commodities, with the notable exception of a few products received through food aid (corn, rice, sugar, milk, and edible oils) or produced domestically (fish, tomatoes, bananas, salt, beer, and soft drinks). 2/ In four years, imports of commodities (c.i.f.) doubled in value, rising from 45.6 percent of GDP in 1991 to 60.2 percent of GDP in 1995 (Table 33). During this period, food imports accounted for approximately one-third of the total import value; while equipment and capital products averaged about one-third of total import value, reflecting the expansion in investments. Faced with an anticipated further widening of the trade gap, on March 1 1996 the authorities introduced temporary import quotas on 26 products, which accounted for 15-20 percent of total 1995 import value. 1/ The authorities are committed to eliminating these quotas and to rationalizing the tariff system in late 1996, harmonizing it with that of the Economic Community of West African States (ECOWAS). 2/

Table 33.

Cape Verde: Merchandise Imports, 1991–95

(In millions of Cape Verde escudos)

article image
Source: Bank of Cape Verde.

Since manufacturing products became Cape Verde’s main export, the major share of exports has been directed toward Portugal, where most of the parent companies of recently established manufacturing enterprises reside (Table 34). The origin of Cape Verdean imports have remained less concentrated and are primarily associated with the many providers of bilateral aid.

Table 34.

Cape Verde: Direction of Trade, 1991–95

(In percent)

article image
Source: Bank of Cape Verde.
(2) Services

Cape Verde has traditionally maintained a surplus in nonfactor services. In 1992-95, this surplus averaged some C.V. Esc 1,600 million, or US$20 million, reflecting a substantial improvement in bunkering and communication service receipts (Table 35). Also, tourism receipts rose throughout the period, but tourism spending abroad, as well as embassies’ expenditures, grew at a faster rate, outweighing any increase in credits.

Table 35.

Cape Verde: Services, 1991–95

(In millions of Cape Verde escudos)

article image
Sources: Data provided by the Bank of Cape Verde, and transformed according to the fifth edition of the Balance of Payments Manual, IMF.

The overall factor income balance shifted from a surplus in 1992 to a deficit in 1993-95 (Table 36). Factor service receipts declined steadily, owing to decreases in investment income, salaries, and technical assistance receipts. Meanwhile, interest payments remained stable near 1.1 percent of GDP. The increase in interest payments was smaller than the increase in disbursements, reflecting the move towards more concessional financing. Technical assistance expenditure doubled in value terms declining from C.V. Esc 93 million in 1991 to C.V. Esc 226 million in 1995, reflecting requirements arising from new investments.

Table 36.

Cape Verde: Factor Income, 1991–95

(In millions of Cape Verde escudos)

article image
Sources: Data provided by the Bank of Cape Verde, and transformed according to the fifth edition of the Balance of Payments Manual, IMF.
(3) Transfers

Private and official transfers increased substantially from 28.0 percent of GDP in 1991 to 35.2 percent in 1995, and constituted an essential element for the viability of Cape Verde’s external position (Table 37).

Table 37.

Cape Verde: Private and Official Transfers, 1991–95

(In millions of Cape Verde escudos)

article image
Source: Bank of Cape Verde.

Includes unclassified food and equipment aid.

Favorable economic conditions in the host countries of many Cape Verdean emigrants and high interest rates at home prompted a substantial Favorable economic conditions in the host countries of many Cape Verdean emigrants and high interest rates at home prompted a substantial increase in emigrants’ remittances, which grew from 16.5 percent of GDP in 1991 to a record 22.7 percent of GDP in 1995 (see also Chapter III).

Official net transfers rose from 9.9 percent of GDP in 1991 to 11.5 percent of GDP in 1995. The principal component is financial aid which remained stable at 7 percent of GDP. The faster growing component was food aid, which increased from 3.0 percent of GDP in 1991 to 5.2 percent in 1995 (Table 38).

Table 38.

Cape Verde: International Food Aid, 1991–95

(Value in thousands of U.S. dollars; volume in metric tons; unit value in U.S. dollars per metric ton)

article image
Source: Data provided by the Cape Verdean authorities.

In 1993, the grant from the Netherlands was given in cash.

In liters.

d. Capital movements, financing, and external debt

During the period 1992-95, the capital and financial account balance improved substantially from C.V. Esc 502 million, 2.2 percent of GDP, in 1991 to C.V. Esc 5,313 million, 16.5 percent, in 1995 (Chart 15). Among the public sector components, project grants and loan disbursements grew by some 300 percent in value terms on account of the substantial increase in the public investment program; each constituted some 2.5 percent of GDP in 1991, and rose to some 5 percent of GDP in 1995. Amortization due remained stable in value terms, at about C.V. Esc 750 million, decreasing as a share of GDP from 3.2 percent in 1991 to 2.3 percent in 1995. Direct foreign investments increased from less than 1.0 percent of GDP in 1991 to about 2.4 percent in 1995 as a result of the effort to promote foreign investment, including through new fiscal incentives.

Balance of payments deficits of 1993-94 were financed drawing down gross official reserves, from US$75.8 million at end-1992, equivalent to 5.1 months of imports of goods and services, to US$42.9 million at end-1994, 2.2 months of imports of goods and services, and by some further accumulation of external arrears (Chart 17).

CAPE VERDE: EXTERNAL SECTOR DEVELOPMENTS, 1991–95

Chart 17.
Chart 17.

FINANCING

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

As Cape Verde increased its reliance on foreign loans to finance government capital expenditure, and continued to accumulate arrears, the stock of official external debt rose from US$147 million in 1991 to US$189 million in 1995 (Table 39). Most of the new debt was contracted on concessional terms and was primarily with multilateral institutions (Table 40, and Chart 18). The ratio of debt over GDP increased from 43.0 percent in 1991 to 45.5 percent in 1995; however, the ratio of debt service to export of goods and services fell from 28.7 percent in 1991 to 17.6 percent in 1995 (Chart 19). The stock of arrears increased from US$1.8 million in 1991 to US$19.1 million in 1995 (Table 41, and Chart 20).

Table 39.

Cape Verde: External Public Debt, 1991–95

article image
Sources: Bank of Cape Verde;, and Fund staff estimates.

All arrears are on the debt of the Central Government.

In 1991, arrears OPEC and BADEA of US$2.58 million and US$4.99 million, respectively, were rescheduled.

In 1991 arrears of US$1.144 million to BADEA were refinanced.

Includes interest arrears.

Table 40.

Cape Verde: External Public Debt by Creditor, 1993–95

(In millions of U.S. dollars)

article image
Source: Bank of Cape Verde.

CAPE VERDE: EXTERNAL SECTOR DEVELOPMENTS, 1991–95

Chart 18.
Chart 18.

FOREIGN EXCHANGE RESERVES

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

CAPE VERDE: EXTERNAL SECTOR DEVELOPMENTS, 1991–95

Chart 19.
Chart 19.

EXTERNAL DEBT

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

CAPE VERDE: EXTERNAL SECTOR DEVELOPMENTS, 1991–95

Chart 20.
Chart 20.

DEBT SERVICE RATIOS

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

7. Structural reforms

a. State enterprise reform and privatization

The development of a public enterprise sector dates back to the early post-independence period. In part, it constituted the practical solution to the nonexistence of private sector initiative and, in part, it was inspired emphasized creating an enabling environment for private economic initiative, and rationalizing the state enterprise sector, through restructuring, privatization and, where necessary, liquidation.

(1) The state enterprise sector in 1990

In 1990, a preliminary USAID-financed assessment of the sector identified 32 companies, 21 of which were wholly state owned and 11 were mixed economy enterprises. The sector employed more than 6,400 persons and generated about 25 percent of GDP. It accounted for 42 percent of the value added in manufacturing, 61 percent in transport and telecommunications, 87 percent in energy, 15 percent in commerce, 40 percent in fishing, and 45 percent in tourism. Although no public enterprise, except for the electricity and water utility, received direct subsides from the Government, a significant number of enterprises owed their viability to: (i) receiving credits from the BCV, about one-third of the total credit extended to the economy during the 1980s; (ii) obtaining grants and foreign loans, which accounted for more than 60 percent of external debt obligations serviced during the 1980s; and (iii) accumulating internal and external arrears.

(2) Strategy of the reform

In May 1992, the authorities obtained a US$4.2 million IDA privatization technical assistance to support the privatization of the public enterprise sector; the following month, a privatization law was approved. In January 1993, the Bureau for the Restructuring of the State Enterprise Sector (Gabinete de Apoio à Reestruturação do Sector Empresarial do Estado - GARSEE) became the core institution in charge of planning, steering, and supervising the restructuring and divestment of the state public enterprises. The GARSEE’s objectives included (i) preparing an action program and initiating its implementation; (ii) monitoring the transition process, improving the productivity of the public enterprise sector, and cushioning the retrenchment of redundant staff; (iii) restructuring selected public sector enterprises performing economically or socially strategic activities; and (iv) promoting private sector development in areas offering good prospects for growth and potential for exports.

(3) Status of the program

The reform program made considerable progress in 1994 and 1995, while, in 1993, there had been some difficulties because of the need to overcome legal hurdles related to ownership issues, and also because of labor unrest and lack of adequate corporate records.

In 1993, the GARSEE put together the divestiture program and initiated its implementation, managing the conversion of the 32 enterprises into 45 stock companies, with adequate accounting, market valuation, and terms of eventual sale. By April 1996, 28 of the 45 companies had been privatized or liquidated, and plans were in progress to privatize another 6 firms in 1996. Since 1994, the GARSEE has been able to make net transfers to the budget, which amounted to some 6 percent of GDP in 1995. The net transfers were not the only positive results provided by the reform. The restructuring of selected public sector enterprises performing economically or socially strategic activities progressed briskly: public enterprises aggregate earnings before taxes in the state enterprise sector turned positive in 1993 and remained positive thereafter (Table 20), notwithstanding that, since 1991-92, budgetary transfers have been drastically slashed (Table 17), loans from the domestic banking system decreased to minimal levels (Table 21), and foreign loans reduced to zero (Table 28). 1/

Private sector development was favored by the privatization law, which required the valuation of property by independent private experts, and by the fact that all the sales were well advertised and took place through open competitive bidding. The shares owned by the Government in mixed companies were usually purchased by the other shareholders, based on the right of first refusal given them by the statutes of these companies. The privatization of the small public enterprises generally resulted in management/employee buy-outs. The medium-size public enterprises were usually sold to a combination of insiders and outsiders, occasionally in joint ventures with foreign interests. For the large enterprises, strategies have been tailored to the enterprise; for example, for the telecommunications company, the Government adopted the strategy of offering 40 percent of the stock to a foreign strategic investor, 5 percent to employees, 5 percent to Cape Verdean emigrants abroad, and 15 percent to the general public, with the remaining 35 percent to be offered in stages within a few years.

(4) Problems and remedial actions

Three problems seemed to constrain the success of the privatization program: (i) the resistance of the labor unions and/or of the management of some public enterprises; (ii) the entrepreneurial inexperience of the domestic private sector; and (iii) the limited availability of fresh capital.

Public enterprises operating in socially sensitive sectors (electricity, and water and sewage), and those characterized by strong unions (port services and food distribution) or by powerful management (the airline) tended to resist the reform more successfully than others. The staff and the technical and legal advisors of GARSEE dealt successfully with the detailed tactical, legal, labor, and managerial issues of privatization, liquidation, and commercialization of state-owned property. The political commitment to the reform was a necessary ingredient to support GARSEE’s actions vis-à-vis the most resistant public enterprises. In the most sensitive cases, progressive private participation options, such as leasing or “build-own-operate,” constituted valuable alternatives to full-fledged privatization. To overcome the entrepreneurial inexperience of the domestic private sector, the authorities unequivocally expressed their intention and commitment to let the domestic and foreign private sector play a leading role in the economic development of the country. To translate this objective into actions, the Government took the following measures: the Law on Foreign Investments, the Law on Incentives to Export, and the Law on Export Processing Zones (Box 3).

Private Sector Promotion

The Law on Foreign Investment (89/IV/93) establishes that foreign investors are guaranteed the right to transfer outside the country all the profits (free of taxes) derived from the operations of the investment in Cape Verde and are authorized to maintain accounts in convertible currencies and to make all operations from these accounts. Moreover, foreigners are granted protection against expropriation and nationalization.

The Law to provide Incentives to Export (92/IV/93) stipulates that during the first five years of exports or re-exports, taxes on profits shall benefit from a percentage reduction equal to the percentage of foreign currency earning out of total earnings of the exporting entity. Subsequently, fiscal incentives are gradually reduced within a period of ten years.

The Law on Export Processing Zones (99/IV/93) grants special incentives to enterprises whose production is entirely destined to export. These incentives include (i) exemption from taxes on profits and dividends during the first ten years of activity; and (ii) complete exoneration of indirect taxes, including the stamp tax. After the initial period of exoneration, the tax rate on profits and dividends shall never exceed 15 percent of the annual profits and dividends. In addition, these enterprises are exempted from customs duties on construction materials, equipment, and fuels to be used in the facilities for their productive activities.

To attract foreign capital, GARSEE obtained the services of SIDA of Sweden to evaluate and finance the development of a capital market in Cape Verde and to constitute mutual funds with stakes in Cape Verde enterprises for sale domestically and abroad, primarily among Cape Verdean emigrants in the Netherlands, Portugal, and the United States.

b. Poverty alleviation policies

Poverty in Cape Verde is structural. In the past, high mortality, famines, and emigration acted to balance the population with the domestic resource base. Now, the pressure on domestic resources is increasing, owing to falling mortality rates but stable fertility rates, and declining opportunities for emigration related to restrictions in destination countries.

Unemployment in Cape Verde was about 25 percent in 1992-95, with much of the available employment concentrated in the service sector and in a still oversized public sector. Low agriculture potential resulted in rural to urban migration, which precipitated irregular construction, saturation of water and sanitation services, and excessive pressure on social service infrastructures.

Traditional safety nets, based on extended family and social solidarity, have broken down, owing largely to high population growth, declining incomes, and the “modern” way of life. Between 1992 and 1995, health and education represented on average 9 percent and 22 percent of public spending, respectively. There were concerns about the allocation of both budgets, however, given that more than 70 percent of both went to salaries. An important program was created to provide welfare assistance to the poor, as well as to foster asset creation through the labor-intensive public work projects (Frentes de Alta Intensidade de Mão de Obra - FAIMOs). This program was used to provide infrastructure such as roads, schools, hospitals, and to protect the environment through reforestation. The employment created amounted to an average 17,000 people a year and is an essential instrument of poverty alleviation (Box 4).

Food Aid and Domestic Financing of Public Investments

Selling the staple commodities obtained through food grants sold at prevailing international prices has proven particularly advantageous to Cape Verde. The profits generated by EMPA (the distribution company) are transferred to a department of the Treasury, Fundo Desenvolvimiento Nacional - FDN, which uses them to finance labor intensive development projects, mainly through highly labor-intensive public works projects (Frentes de Alta Intensidade de Mão de Obra - FAIMOs).

The management of food aid programs requires major improvements. EMPA continues to have a monopoly on the import of a few key consumer goods. However, as the number of products on which it has monopoly rights has declined, EMPA’s profits have been driven down and, as a result, EMPA has endeavored to reduce operating costs and privatize some parts of its operations. In July 1996, the authorities began a strategic evaluation of EMPA, with a view to eliminating its monopoly and transforming it into a competitive import and distribution company over the medium term.

Concerns were expressed about the quality of the infrastructure created by the FAIMOs. A more decentralized implementation has allowed local communities to identify their needs within the broader national development plan and ensure infrastructure creation in line with national needs and targets. At the local level, this translated into the implementation of projects directed toward producing the basic sanitation, schools, health facilities, and low cost housing needs of the rural communities. Ownership of the projects at the local level should also ensure proper maintenance in the future.

Between 1992 and 1995, the Government’s commitment to efficient market based development was accompanied by actions directed toward involving the weaker segments of the population in the transformation of the economy. These actions were primarily aimed at enhancing the productivity of the population and at improving the access to, and the quality of, social services. Actions to enhance the productivity of the population include (i) restructuring of the primary and secondary education systems; (ii) promoting the financing of low housing construction through market mechanisms; and (iii) financing the establishment of a rural credit development bank. 1/ Actions to improve the quality of social services include (i) improving access and efficient use of water resources; (ii) enhancing the provision and the cost recovery in health care; (iii) decentralizing the provision of social services and bringing the providers closer to the beneficiaries; and (iv) increasing the transparency of the use of profits generated through the sale of donated food and improving the effectiveness of the FAIMOs.

c. Environmental policies

The ecosystem of Cape Verde cannot adequately support a population of some 400,000: shortage of water is a chronic problem, and drought and deforestation have brought about extensive desertification.

In late 1994, the authorities took the commitment to pursue sustainable development and formulated a National Environment Action Plan (NEAP), which identified six areas of interventions: (i) poverty alleviation; (ii) management of natural resources; (iii) energy self-sufficiency; (iv) demographic control and environmental health; (v) water resources; and (vi) population awareness, information and education. The NEAP was accompanied by a detailed action plan, as well as programs and projects for donor financing. These plans, however, were not fully implemented in 1995, as the elections slowed decision making and donors did not respond with the flow of aid that the authorities had hoped for. The 1995 volcanic eruption on Fogo was dealt with primarily in an ad hoc manner, albeit with substantial donor support.

Table 41.

Cape Verde: External Debt Arrears by Creditor, 1993–95

(In U.S. dollars)

article image
Source: Bank of Cape Verde.

The change in arrears in this table does not equal change in arrears in the balance of payments table, owing to inconsistencies in the authorities’ data.

Table 42.

Cape Verde: Medium–Term Debt Service Projections on Outstanding Public External Debt at End–1995

(In millions of U.S. dollars)

article image
Sources: Bank of Cape Verde; and Fund staff estimates.

II. Economic Competitiveness

1. Introduction and overview

The advent of the new Government in 1991 led to a shift in economic strategy toward a more market-oriented approach to growth and export diversification, predicated on reducing the size of the public sector and creating an enabling environment for private initiative and investment in the productive sectors. In support of this change in strategy, the special office that had been created to promote foreign investment, the Center for Investment and Export Promotion (PROMEX), was strengthened, and recently has achieved some notable success. Between 1994 and 1995, 56 investment applications for a total value of US$93 million were submitted for approval. Of these, 16 enterprises are now currently active, representing total investment of US$16 million, while 10 additional investment projects valued at US$40 million are in the final phase of implementation, and are expected to become operational shortly. The investment, largely of Portuguese origin (but including investment from other countries, for example, Italy and China), has been concentrated in the light industry and tourism sectors.

As a result of the surge in foreign investment in light manufacturing industries over the past two years, total merchandise exports more than doubled from 1993 to 1995, albeit from a very small base. In relation to GDP, exports rose from an average of about 1.3 percent in 1991-94 to 2 percent of GDP in 1995. More significantly, the composition of exports has shifted from a reliance on primary commodities (mainly bananas, and fish and shellfish) to footwear and clothing, with the latter two products accounting for about three-fourths of total exports in 1995 (Chart 21).

CHART 21
CHART 21

CAPE VERDE COMPOSITION OF EXPORTS, 1991–95

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

Source: Cape Verde authorities.

The increase in foreign investment by private business groups may be seen as prima facie evidence of the relative attractiveness of investment opportunities and hence the current competitiveness of the Cape Verdean economy. The purpose of this analysis is to evaluate the available data to assess the trends in the external competitiveness of Cape Verde, and to compare developments in Cape Verde with those in a selected group of Countries that offer alternative potential investment opportunities. To the extent possible, this study will rely on industry-specific indicators where applicable; the industries chosen were footwear and apparel. It is organized as follows: Section 2 provides the analytical framework and describes various approaches to evaluating competitiveness; Section 3 applies the traditional indicators of the analysis of the competitiveness of Cape Verde; finally, Section 4 reviews the conclusions to be drawn from the foregoing analysis.

2. The analytical framework

There are two approaches to evaluating the competitiveness of the Cape Verde economy that will be pursued in the current analysis. The first is to compare recent trends (1990-95) in the relevant domestic costs of production in Cape Verde with those costs in a selected group of potential competitor countries. The factors of product ion that are often used to compare the relative attractiveness of alternative investment opportunities include labor, electricity (or other relevant energy sources), water, and land. This analysis will examine relative labor costs, as labor is the main domestic value-added component of production. The present study will examine the current level and recent trends in wages in Cape Verde, compare the current level of wages in Cape Verde with those in a selected group of competitor countries for which comparable data are available, and will survey how trends in U.S. dollar-denominated wages in Cape Verde since 1990 compare with wage developments in the same group of countries. 1/

Second, in order to complement such an approach, the analysis will also compare changes in the CPI-based real effective exchange rate (REER) of Cape Verde with those of selected groups of countries. 2/ The use of the comparison of changes in the REERs of different countries provides a gauge of shifts in relative competitiveness between countries. 3/ The period selected for the REER analysis is also the years 1990-95. In both instances, this time frame was chosen to provide insight into how Cape Verde’s external competitiveness evolved during the period beginning immediately prior to the shift in economic policy orientation.

3. Competitiveness indicators for Cape Verde

This section analyzes the competitiveness of Cape Verde from a microeconomic and macroeconomic perspective by comparing developments in Cape Verde with those in four groups of countries. The first group, comprised of Italy, Spain, and Portugal, the latter of which is the country of origin of most of the recent foreign direct investment in Cape Verde. The second group consists of two small island economies, Mauritius and Seychelles, and Morocco and Tunisia; the countries in this group are seen as possessing comparable actual or potential investment opportunities, particularly in the sectors that have experienced the recent surge in investment in Cape Verde, and/or have a similar proximity to European markets. The third group represents a group of potential African competitor countries, 1/ while the fourth group represents potential Asian competitor countries. 2/ The data used in this study are drawn from various UNIDO Industrial Statistics Yearbooks, the Fund World Economic Outlook (WEO) and Information Notice System (INS) databases, and from Fund documents (e.g., recent economic developments and statistical annexes) for the countries used.

a. Wage developments in Cape Verde, 1990–95

Private sector operators, who have recently invested in Cape Verde, indicated that labor costs represent the largest single domestic cost component for the industries considered. Moreover, they confirmed that the relative level of the cost of labor was the key economic factor contributing to their decision to invest in Cape Verde. 3/ This section examines the nominal level of wages in Cape Verde compared with the reference group of countries, and considers how recent wage trends in Cape Verde compare with those in the countries for which relevant data are available.

The average monthly wages for 27 occupational categories in Cape Verde stood at US$160 in 1995, whereas for the 6 lowest paid categories the average monthly wage was US$75 (Table 43). 4/ The average monthly wage level for the lowest paid categories in Cape Verde is considerably lower than the levels prevailing in the footwear and apparel industries of the countries in the European comparator group (Table 44). Thus, wages paid to unskilled workers in Cape Verde are approximately 14 percent of the estimated average monthly wages in the relevant industrial sectors in Portugal, and about 11 percent and 13 percent, respectively, of the average wages in the same industries in Italy and Spain.

Table 43.

Cape Verde: Average Wage Developments, 1985-95 1/

article image
Sources: Tables 47 and 48; and staff estimates.

Data conversion assumes 5.5 working days in a week, 8 hours a day, and 4.2 weeks in a month.

Deflated by the Consumer Price Index (CPI).

Table 44.

Estimated Nominal Wages in Footwear and Wearing Apparel Industrial for Selected Industrial Countries, 1985-95 1/

article image
Sources: UNIDO, 1995 Yearbook: OECD, ECU; and staff estimates.

Data conversion assumes 5.5 working days in a week, 8 hours a day, and 4.2 weeks in a month; the index of unit labor costs and the index of relative unit labor costs use standard OECD methodology.

Regarding the developing countries chosen, the UNIDO Yearbook provides data on the wages per employee in current dollars in the apparel and footwear industries only for Indonesia, Kenya, Malaysia, Mauritius, Morocco, Philippines, Senegal, and Thailand. Overall, wages for this group of countries averaged about US$163 a month in the latest years for which data are available (1989 to 1991) for the industries under consideration, while the wages in Cape Verde for the lowest paid categories ranged from US$52 to US$65 a month (Table 45). Thus, wage levels for the lowest paid categories in Cape Verde also compare favorably with those in the competitor developing countries for which data is available.

Table 45.

Cape Verde: Average Monthly Wages by Footwear and Wearing Apparel Industries in Selected Developing Countries

(In U.S. dollars)

article image
Source: UNIDO, 1995 Yearbook.

UNIDO presentation of data is annualized for thousands of U.S. dollars. Data presented in this table are UNIDO data divided by 12 times 1,000. Arithmetic average is being used.

Average wages in Cape Verde, expressed in U.S. dollars, increased by 40.6 percent from 1990 to 1995, while for the six lowest paid categories, wages increased by some 27.5 percent over the same period (Table 43). The average nominal increases in wages, expressed in the local currency, were 54.3 percent and 40 percent, respectively over this period. Most of the wage increases were concentrated in the 1990-92 period in response to labor militancy in the election period. Since 1992, average wage increases have been more moderate. Wages expressed in U.S. dollars fell from 1992 to 1993 (by 13.5 percent for the average of all categories, and by 7.7 percent for the lowest paid group), as the depreciation of the C.V. escudo vis-à-vis the U.S. dollar more than offset the increases in wages expressed in local currency. The relatively modest combined wage increases in local currency over the next two years, coupled with the cumulative appreciation of the C.V. escudo vis-à-vis the U.S. dollar, brought the average wages for all categories expressed in U.S. dollars back to their 1992 level, while the average for the lowest paid categories was about 11 percent above the 1992 level.

In contrast to developments in Cape Verde, between 1992 and 1995 the increases in unit labor costs and in nominal wages for the manufacturing sector in Portugal, expressed in local currency, were 29.5 percent and 30.7 percent, respectively. Adjusted for the impact of the depreciation of the Portuguese escudo vis-à-vis the U.S. dollar, unit labor costs and wages increased by about 16 percent and 17 percent, respectively, over the period as against the 11 percent increase for the lowest paid wage category in Cape Verde. Thus, the competitiveness of Cape Verde, judged by relative changes in U.S. dollar-denominated average wages, improved slightly when compared with Portugal over the 1992-95 period, although in the longer period 1990-95 this measure of competitiveness deteriorated. With regard to Italy and Spain, the available data suggest that productivity gains in Italy and Spain held the increase in unit labor costs to below the increase in nominal wages. 1/

b. Developments in the Real Effective Exchange Rate (REER)

Over the period 1990-95, the REER of Cape Verde remained relatively stable, fluctuating within a band of plus or minus 5 percent of the average REER in 1990. To evaluate the relative competitiveness of Cape Verde using this approach, the developments in the REER of Cape Verde (calculated on the basis of the regional exports and imports pattern, taking also into account competitors in third markets) were compared with developments in the four country groups described above (Charts 22-23).

CHART 22
CHART 22

CAPE VERDE AND SELECTED COUNTRIES REAL EFFECTIVE EXCHANGE RATES, 1990–95

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

Source: IMF, Information Notice System.
CHART 23
CHART 23

CAPE VERDE AND SELECTED COUNTRIES REAL EFFECTIVE EXCHANGE RATES, 1980–95

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

Source: IMF, Information Notice System.

With respect to the group of selected European countries, the REER of the Portuguese escudo appreciated vis-à-vis the Cape Verde escudo, whereas the REERs of the currencies of Italy and Spain depreciated vis-à-vis the Cape Verde escudo, implying that Cape Verde improved its competitiveness with Portugal, while experiencing a deterioration of relative competitiveness with respect to Italy and Spain. This reinforces the analysis based on changes in relative labor costs, which yielded the same result for this group of countries; this would reinforce any predisposition of Portuguese economic agents to invest in Cape Verde.

Compared with the countries in the “close competitor” group, the relative changes in the REERs, particularly since 1992, point to Cape Verde achieving a slight competitive edge. However, it must be stressed that the differences in performance are relatively minor, as the fluctuations remained within a narrow band (between 95 and 110), and hence may not represent a significant change in relative competitiveness. With respect to the third and fourth groups, an analysis of REER trends yields much more varied conclusions. In the case of the African countries chosen, the relative changes in the REERs imply that Cape Verde’s competitiveness improved over the period only vis-à-vis Kenya, while it lost ground when compared with Ghana, Madagascar, Senegal, and Togo. 1/ This phenomenon reflects the fact that these countries have all implemented adjustment programs recently, in which a key policy element was a devaluation of their exchange rates.

For the chosen group of Asian countries, only the Philippines appears to have suffered a meaningful loss in competitiveness vis-à-vis Cape Verde. China significantly improved its competitiveness vis-à-vis Cape Verde from 1990 to 1993; however, in 1995 in particular, Cape Verde’s competitiveness with China improved, which coincided with the recent investment by Chinese interest in Cape Verde. On the other hand, Cape Verde was broadly able to preserve its competitiveness vis-à-vis Indonesia, Malaysia, and Thailand.

Any analysis of relative competitiveness based on changes in the REER is sensitive to the base period chosen. Chart 24, which takes 1985 as the base year, shows that while the index of the REER of Cape Verde has remained relatively stable since 1985, fluctuating within a relatively narrow band (roughly between 95 and 100), the REERs of all the comparators have undergone significant shifts. The direction of the shifts indicates that the competitiveness of the Cape Verdean economy, as measured by relative changes in the REER, has been eroded when compared with the chosen developing country comparators in the period since 1985, but most of the loss occurred prior to 1990. This underscores the importance of judiciously choosing a base period for a competitiveness analysis.

CHART 24
CHART 24

CAPE VERDE AND SELECTED COUNTRIES REAL EFFECTIVE EXCHANGE RATES, 1985–95

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

Source: IMF, Information Notice System.

4. Conclusion

In conclusion, the analysis shows that, as expected, Cape Verde has a strong comparative advantage in terms of wages with developed countries such as Portugal, Italy, and Spain. In relation to developing countries, it has improved its competitiveness since 1990 toward close competitors such as Mauritius, Morocco, and Tunisia, while maintaining competitiveness when compared with key Asian producers. With regard to African countries, such as Ghana, Madagascar, Senegal, and Togo, the relative changes on the REERs since 1990 suggest that a clear loss of competitiveness has taken place. In this context, a more exhaustive comparison of the level of wages is necessary to reach a determination regarding Cape Verde’s current competitiveness. In addition to the increase in investment in Cape Verde, these data would seem to indicate that the exchange rate policy pursued by the authorities has been broadly able to maintain competitiveness, especially when taking into account the productivity of the Cape Verdean worker.

Table 46.

Cape Verde: Comparison of Relative Unit Labor Costs in Italy and Spain, 1991–94

(1990=100)

article image
Sources: OECD; and staff estimates.
Table 47.

Cape Verde: Level of Monthly Wages by Occupation, 1985–95

(In Cape Verde escudos)

article image
Source: Bank of Cape Verde.
Table 48.

Cape Verde: Level of Monthly Wages by Occupation, 1985-95 1/

(In U.S. dollars)

article image
Sources: Bank of Cape Verde; and IMF, International Financial Statistics.

Data conversion assumes 5.5 working days per week, 8 hours a day, and 4.2 weeks in a month.

Table 49.

Cape Verde: Nominal Changes in Monthly Wages by Occupation, 1990-95 1/

(In Cape Verde escudos)

article image
Table 50.

Cape Verde: Real changes in Monthly Wages by Occupation, 1990–95

(In percent change, CPI based)

article image
Source: Bank of Cape Verde.
Table 51.

Cape Verde: Nominal changes in Monthly Wages by Occupation, 1990–95

(Percent change of U.S. dollar figures)

article image
Source: Bank of Cape Verde.

III. Determinants of Emigrants’ Remittances

1. Introduction

Emigrants’ remittances have historically constituted a fundamental component of Cape Verde’s income; they have accounted for more than 12 percent of GDP since 1978. They increased steadily by 12.4 percent on average from 1984 to 1995, with the average growth rate increasing from 11.5 percent in the 1980s to 13.2 percent the 1990s. 1/

The United States has been the major source of these remittances, accounting for some 25 percent of the total on average, followed by the Netherlands, responsible for approximately 20 percent, Portugal and France with 12 percent each, and Italy and Germany with some 8 percent each. The sources of the remittances have been mostly stable, with these six major countries covering more than 80 percent of total remittances. While the total share of remittances from these six countries has been constant, the composition has changed over time (Chart 25); notably, the share of remittances from the United States has declined while that from Europe has increased (Chart 26). Among the five European countries, the relative shares from France and Italy have increased as opposed to a declining share coming from the Netherlands (Chart 27). An interesting case is Portugal, whose share declined sharply until the mid-1980s, followed by a significant recovery after 1986.

CAPE VERDE: REMITTANCES, 1978–95

Chart 25.
Chart 25.

REMITTANCES

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

Chart 26.
Chart 26.

REMITTANCE SHARE BY COUNTRY

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

Chart 27.
Chart 27.

REMITTANCES

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

Sources: Bank of Cape Vendor; and Fund staff estimates.1/ 6 OECD countries Include U.S.A., Portugal, Italy, Netherlands.

Changes in the time pattern of remittances from the major source countries are highly correlated with Cape Verdean monetary policies. Interest rates in Cape Verde remained unchanged from 1974, when the country gained its independence, until January 1, 1985; the interest rate for time deposits was kept at 6.5 percent. However, with inflation reaching some 20 percent in the early 1980s, real interest rates had become strongly negative, adversely affecting both domestic savings and emigrants’ remittances. In response, the Council of Economic Ministers approved an increase in interest rates and the creation of three special deposit accounts for emigrants’ remittances.

The first type of account is for deposits in foreign exchange, open to both emigrants and local holders of foreign exchange. The deposits can be in U.S. dollars, Deutsche marks, Dutch guilders, or French francs. The interest rates are linked to Euromarket rates augmented by a small margin. Under this arrangement, the principal can be drawn in the original currency while interest can be drawn only in local currency.

The second type of account targets emigrants exclusively and aims to attract remittances deposits in local currency after conversion from the foreign currency. Such deposits have accounted for more than 70 percent of total emigrants’ remittances, with their share increasing over time (exceeding 80 percent in March 1996). In February 1996, the Banco de Cabo Verde freed the rate on emigrants’ C.V. escudo deposits, which has been administratively set at a fixed 12 percent interest since January 1985, a 2 percentage point premium above what is paid to local depositors. Under the free market mechanism, the Government decreased the percentage point premium to 1 percent. As a trade-off for this interest rate premium, this account does not permit any conversion back to foreign currency.

The third type of account applies to both emigrants and local residents and provides for saving and credit in C.V. escudos. It pays a 6 percent interest rate for six-month deposits and 10 percent for one year. The account allows borrowing up to twice the deposited balance. For such loans, local residents pay a 12.5 percent interest rate, while emigrants are entitled to an 8 percent interest rate.

The natural question is whether these policies have been effective in attracting remittances from abroad. Is the special emigrant account the main reason for this steady inflow and change in country composition share? Or is it mainly due to some other factors such as positive income effects abroad, or is it a tighter family links between emigrants and their relatives in Cape Verde? This study attempts to analyze the main forces driving these inflows into Cape Verde and examines to what extent the country’s policies have been effective. The next section introduces a two-period single good model that provides some clear indications of what affects the emigrants’ remittances. Given the predictions from the model, Section 3 attempts to apply the theory to remittances data for major source countries from 1979 onward.

2. Model

This section introduces a two-period model to characterize the emigrants’ remittances behavior. It focuses on the saving behavior of an emigrant abroad who lives apart from his family in Cape Verde. Remittances can be used either for immediate consumption by family members or/and saved for future consumption. 1/ The nature of remittances depends mostly upon whether emigrants plan to stay abroad permanently or temporarily. While permanent emigrants are joined by the rest of the family abroad and do not return to the home country, temporary emigrants earn and save for their family’s consumption in Cape Verde. Under the arrangements of the high interest special emigrant accounts, only temporary emigrants would take advantage of the accounts. Saving in Cape Verde is a close substitute to saving abroad; however, the nonconvertibility back to the original foreign currency for deposits in local currency would restrict those who benefit from saving in Cape Verde; this is a sunk cost. Permanent emigrants would not place their savings in the local currency to avoid this sunk feature. Instead, they can save in foreign exchange accounts, which would guarantee the convertibility, but still be partly bound by the provision of no interest withdrawal in foreign currency. Hence, remittances from permanent emigrants will be merely for consumption purposes. Temporary emigrants, on the other hand, will have two saving options: save abroad and bring the money back to the home country or save in the home country.

The likelihood of returning home is not discrete ex ante, and thus the classification of permanent or temporary needs to be generalized. The most likely case would be emigrants with some expectations of returning to the home country. Nevertheless, the sunk feature of the emigrant special account will have a similar impact on the emigrant’s saving behavior, depending upon how likely an emigrant is to return home. The more likely it is that one remains abroad, the less the amount that will be transferred to Cape Verde savings.

More formally, the model considers a saving decision of an emigrant abroad who earns in the first period and retires in the second period in a simple two-period single good model (see Annex I for details of the model). At retirement, the emigrant can either remain abroad or return to the home country. However, the family gets together in the second period independent of resident location and no consumption transfers are made in the second period. Define μ ϵ [0,1] as the likelihood of the emigrant returning to his home country in the second period; emigrants with μ=0 would be permanent while μ=1 implies temporary emigration. Those in between, i.e., μ ϵ (0,1), are those still uncertain at the beginning of period 1.

Assume for simplicity an additive and separable utility function with respect to own consumption abroad in the first period, C1*, family’s consumption in Cape Verde in the first period, Ch, and the whole family’s future consumption in the second period, C2 if resident in Cape Verde, or C2* if abroad. Given (i) his likelihood of returning, μ, (ii) wages earned in the first period w, (iii) good prices for two periods, (Pi*, Pi), i=(1,2), (iv) the interest rates (ri, ri*) i=(1,2) (where a superscript * denotes prices abroad), and (v) the exchange rates (e1, e2) i=(1, 2), the emigrant’s problem is to decide how much to save in the foreign and home markets, Sf(μ) and Sh(μ), and how much to send to his family, T.

Saving in the home country essentially depends on the likelihood of his returning to the country, owing to the sunk feature of the emigrant special accounts. Intuitively, the high interest rate in Cape Verde will affect the emigrant’s saving behavior through his likelihood of returning. Even if the interest rate is relatively high in Cape Verde, if his likelihood of returning is low, then he will try to avoid the sunk deposits in Cape Verde and instead save abroad. Conversely, the more likely a person is to return to his home country, the more attractive the high interest in emigrants’ special accounts.

A Review of the Literature on International Workers’ Remittances

Two main approaches to modelling international workers’ remittances (IWR) prevail in the economic literature. One approach treats IWR as an endogenous variable in the decision making of migration and remittances within the family, therefore insensitive to home country economic policy. The other approach treats IWR as a transfer of saving from one region to another, therefore very sensitive to portfolio considerations and home country economic policy.

The endogenous migration approach treats remittances as an intertemporal contractual agreement between the migrant and his family. Lucas and Stark (1984) explore altruism and self-interest as elements in the remittance contract, and Chaney (1986) models remittances as intra-family resource transfers in the context of a family production function. One of the major implications of this approach is that there is a minimum level of “required”, remittances determined by the terms of the contract between the migrant and his family. This literature suggests an empirical framework that includes as major determinants of remittances the composition of the family at home and abroad, the level of disposable income (wage) in the host country, the income differential, and the anticipated length of stay. Empirical evidence in general corroborates the predictions of this framework (e.g., Oberai and Singh (1980), and Serageldin et al. (1981)] and show that the fraction of remittances is not likely to be affected by incentive policies in the labor-sending country such as preferential exchange rates and interest rates, the permission to maintain foreign exchange accounts, and land purchase arrangements [e.g., Swamy (1981), Bhat (1981), Straubhaaar (1986), and Glytsos (1988)].

The portfolio approach treats remittances as a result of a broader process of portfolio allocation of his savings by the migrant worker. This approach focuses on relative rates of return, relative prices and uncertainty as primary determinants in the decision to remit or to maintain savings in the host country [e.g., Glytsos and Katseli (1986), Miranda (1988) and Straubhaar (1986)]. This literature suggests an empirical framework that includes as major determinants of remittances the deviation of the official exchange rate from the one defined by a purchasing power parity equilibrium between the sending- and the host-country, the covered interest parity differential, the stock of emigrant workers abroad and their income.

Comparing the empirical results obtained by models based on the two different approaches, it appears that, while the impact of income and demographic variables seem to be corroborated by most studies, the evidence on the impact of relative rates of return is more mixed.

The model introduced in this chapter synthesizes these two approaches to modelling IWR to analyze the determinants of Cape Verde experience with emigrants’ remittances. In this model, remittances are on one hand determined by the family’s intertemporal consumption decision based on the contractual agreement between the emigrant and his/her family; an altruistic parameter summarizes how much commitment is expected in the family agreement (although it is treated as a fixed parameter in the model, it could be added as another control variable of the maximization problem in order to encompass the different degree of commitment). On the other hand, remittances are also determined by portfolio considerations. The empirical results confirm the findings prevailing in the economic literature. Also in the experience of Cape Verde, the income discrepancy between Cape Verde and the emigrants’ host country is highly significant, while the covered interest parity differential is insignificant.

(1) Permanent emigrant (μ=0)

Because of the sunk feature of the home saving, permanent emigrants will only send consumption-related transfers to Cape Verde and save the remaining income abroad. Hence, the level of the interest rate in the emigrant accounts does not have any impact on the emigrants’ saving decision. Instead, the wage rate and the interest rate in the working country would determine the amount of the transfers through the income effect channel. Since consumption in both period 1 and 2 are equally desirable, under the current preference assumptions, a higher lifetime income will induce emigrant to increase consumption in both periods. 1/ 2/

(2) Temporary emigrant (μ=1)

The temporary emigrants can take advantage of the high interest rate policy of the home country if saving at home is relatively more profitable than saving abroad. Remittances to the home country would be for both consumption and portfolio purposes.

Portfolio: Two saving alternatives are possible for temporary emigrants: save abroad (earn r1*) and convert it at the expected C.V. escudo rate in the second period (at E(e2)), or convert into C.V. escudos in the first period (at e1) and earn home country’s interest (r1). In this case, two factors, interest rates and the exchange rates, play a key role in determining the emigrant’s portfolio.

A higher interest rate either at home or abroad would imply that less saving is necessary to ensure consumption in the second period (income effect). 3/ In order to assess the relative profitability of the two saving assets, however, comparing the level of two interest rates is not sufficient. It depends critically on each emigrant’s expectations concerning the exchange rate in the next period, since this will affect the timing of the conversion of the remittances to the home currency. If a depreciation is expected, given the two countries’ interest rates, it is more profitable to hold the savings abroad and bring it back in the second period. On the other hand, if an appreciation is expected, then it is better to send money in the first period to avoid conversion losses. Hence, the amount of saving in the home country will depend on the relative profitability, taking into consideration both interest rates and the expected exchange rates. If saving at home is more profitable then all the money will be saved at home.

Consumption purposes: As in the case of a permanent emigrant, remittances for consumption purposes depend mainly upon the income level. A higher lifetime income will imply higher transfers for consumption uses. The substitution effect also applies between consumption in two periods in the same fashion as in the permanent case, but the income effects dominate under the model’s specification.

(3) Uncertain emigrant (0<μ<1)

The case of an uncertain emigrant who has not decided whether to return home or not upon his retirement can be treated as an extension of the temporary case. In both cases, the relative profitability between savings abroad and home will determine where to save, i.e., how much portfolio-related money to send to the home country. The difference is the uncertainty, which adds a subjective evaluation in comparing the relative profitability. The emigrant may not save in the home country, even if the interest rate is relatively high, unless his likelihood of returning is sufficiently high. Thus the assessment of relative profitability is transmitted to each emigrant through his “personal filter” by attributing different weights to the benefit of holding a unit of the currency home or abroad. Given the objective relative profitability of two savings, the lower the likelihood of returning home, the lower the saving will be in the home country.

3. Data and estimation

Emigration in Cape Verde dates from the nineteenth century when whaling ships recruited sailors from the islands. It is estimated today that there are over 600,000 Cape Verdeans abroad, by far exceeding the current resident population in Cape Verde. 1/ Most of the emigration to the United States occurred at the beginning of the twentieth century, which implies that current United States residents are fifth generation. Although there is still some flow of emigrants to the United States, the number is rather limited (about 650 every year). On the other hand, emigration to western Europe began in 1946 and became significant in the 1960s and 1970s, owing to a high demand for labor, which persisted until the western European countries, notably France and Germany, began taking restrictive measures against immigration and began offering incentives to induce emigrants to return to their home country. Given this historical relation, emigration to the United States could be broadly classified as permanent and that to Europe, predominantly temporary. According to the model introduced in the previous section, this classification would imply that the trend of remittances from the United States should be mainly affected by income trends there. On the other hand, not only the income effects but also the substitution effect induced by differences in the two savings assets should be also considered in the case of Europe.

Chart 28 shows relative income trends among the major source countries. GDP per capita is used as a proxy to denote income and its dispersion is expressed in terms of the ratio of per capita income in each country to that in Cape Verde’s. While the income trends for France, the Netherlands, and Germany move together on a slight decline, the trend is increasing for Portugal and is almost stable for Italy. Relative income in the United States also shows an overall decreasing trend after the 1980s. The peak period in mid-1980s partly overlaps the period when the United States share of remittances also peaked, confirming the role of income effect in the United States in assessing remittances.

Chart 28.
Chart 28.

INCOME DISCREPANCY WITH CAPE VERDE

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

Chart 29 depicts the trend of real interest rates in these countries net of Cape Verde real interest rates. 1/ It clearly shows that holding foreign currencies was more profitable before the mid-1980s, explained by a large gap in the profitability between the C.V. escudo and other foreign currencies. The gap substantially decreased and even became negative after 1985, implying a higher return to savings in Cape Verde. While the high interest gap is a common feature among source countries until the mid-1980s, only remittances from Portugal (and the Netherlands, to a small extent) seem more responsive to this interest gap. In fact, remittances declined sharply until 1985, when the interest parity was restored. According to the model, emigrants must have substituted some of the savings previously made in the home country to more profitable saving in Portugal. After 1985, on the other hand, both effects from the high interest rate in Cape Verde as well as the higher relative income in Portugal likely contributed to a sharp increase in the remittances level.

Chart 29.
Chart 29.

REAL INTEREST RATE DIFFERENTIAL

Citation: IMF Staff Country Reports 1996, 103; 10.5089/9781451809237.002.A001

A simple OLS estimation was carried out to investigate the impact of income and the relative rate of return among countries in analyzing the remittances trend. The model predicts that the remittances, which now represent a sum of both consumption and portfolio transfers, are a function of income, the relative rates of return, and expected exchange rates. Rational expectations are assumed. All remittances data are pooled to estimate the following: 2/

I ⁢ n ⁢ Re ⁢ m t = α 0 + α 1 ⁢ Y t + α 2 ⁢ ( Δ ⁢ R t ) + α 3 ⁢ I ⁢ n ⁢ e t + 1 + α 4 ⁢ D 85 + ϵ t , ( 1 )

where Remt is remittances in period t, Yt is the income dispersion ratio in t defined as GDPi/GDPcv in per capita terms, ΔRt is the difference in real interest rates between country i and Cape Verde, and et+1 is the next period realized exchange rate, which is equivalent to time t’s predicted value under the rational expectations. Finally, a dummy variable D85 is introduced to differentiate those remittances prior to the creation of emigrants’ special deposit accounts.

The estimation results show positive and significant coefficients for income effect and the dummy variable for special accounts, and a negative and significant coefficient on the expected exchange rate (Annex 2). The positive sign for income dispersion confirms that the transfers will be larger given higher relative income abroad. The introduction of the emigrant special account appears to have been successful; the impact is positive and highly significant. The negative and significant sign in the expected exchange rate verifies that, as predicted by the model, the evolution of the exchange rate has an effect through a possible conversion gain or loss of the foreign exchange depending on the timing of its conversion. The interest rates are, on the other hand, statistically insignificant. This might be partly caused by a possible structural change in 1985 induced by the introduction of the high interest policies, which is proxied by the positive coefficient of the dummy variable. The coefficient on the intercept is highly significant, which implies that there are some other factors not captured by the model. Also, this might incorporate the impact of likelihood of returning home discussed in the model which could not be explicitly introduced in the estimation, owing to the lack of more extensive data on stock and flow of emigrants abroad. The altruistic parameter of the emigrant, if measurable, would also contribute to higher remittances.

ANNEX I

The Model

The maximization problem of an emigrant characterized by likelihood μ∊[0,1] to return to his home country after retirement becomes as follows:

Max s ⁢ . ⁢ t. ⁢     ln C 1 * ⁢ +  α ⁢ ln C h ⁢ +  β ⁢ [ ( 1 − μ ) ⁢   ln C 2 * ⁢ +  μ ⁢ ln C 2 ] C 1 * ⁢ + T + S t ⁢ ( μ ) ⁢  + S h ⁢ ( μ ) = w ( 1 )
T =  P 1 ⁢ C h ⁢ e 1 ( 2 )
P 2 * ⁢ C 2 * ⁢   =  S t ⁢ ( μ ) ⁢ R 1 * , and ( 3 )
P 2 ⁢ C 2 ⁢ E ( e 2 ) ⁢   =  S t ⁢ ( μ ) ⁢ R 1 * ⁢ E ( e 2 ) ⁢  +  S h ⁢ ( μ ) ⁢ e 1 ⁢ R 1 , ( 4 )

where

  • Ci (i=1,2): consumption in period 1 and 2.

  • α > 0: altruistic parameter of the emigrant.

  • Ch: consumption of the remaining family in the first period

  • T: transfer for consumption purposes

  • Si (i=f, h): savings abroad (f) and home country (h)

  • Ri=1+r1 where ri is interest rate in period i.

  • e1, E(e2): exchange rate (C.V. Escudos per unit of foreign currency) in period 1 and expected value in second period.

  • Pi: price of consumption good in period i.

Asterisks stand for foreign denomination. Finally,

P 1 * = 1. ( 5 )

The following equations are derived by solving the first order conditions:

1 C 1 * = λ ( 6 )
α C h = λ P 1 e 1 ( 7 )
β ⁢ ( 1 − μ ) C 2 * ⁢ = λ ⁢ P 2 * ( 8 )
β μ C 2 = λ P 2 E ( e 2 ) ( 9 )

From (3) and (8), we obtain

β ⁢ ( 1 − μ ) λ ⁢ R 1 * ⁢ = S f ⁢ ( μ ) ( 10 )

Substitute (10) and (9) into (4) and get

S h ( μ ) = β [ μ − ( 1 − μ ) E ( e 2 ) ] e 1 R 1 λ ( 11 )

Define b > 0 such that bR1* = e1R1, i.e., the parameter accounts for profitability of the two savings in either currency depending on the expectations concerning the next period exchange rate. The two are perfectly substitutable when b = E(e2) assuming no transaction cost. Thus, if b > E(e2), this implies that it is more profitable to hold a unit of currency until the second period rather than converting in the first period. Conversely, if b < E(e2), then converting to the home currency in the first period is more profitable. Using (5) (7) (10) (11) and (1), we get a function for λ:

λ = ( 1 + α ) ⁢ b ⁢ R 1 * + β ⁢ [ ( 1 − μ ) ⁢ ( b − E ⁢ ( e 2 ) ) + μ ] b ⁢ R 1 * ⁢ w ( 12 )

Thus we obtain the savings function in the home country as follows:

S h ( μ ) = β [ μ − ( 1 − μ ) E ( e 2 ) ] w ( 1 + α ) b R 1 * + β [ ( 1 − μ ) ( b − E ( e 2 ) ) + μ ) ] ( 13 )

Given λ > 0, the denominator of (12) is positive, thus Sh⁢(μ)<>⁢0 if

μ < > ⁢ ( 1 − μ ) ⁢ E ⁢ ( e 2 ) ⁢ . ( 14 )

Thus the lower the probability of returning to the home country, the lower the saving in the home country. Equation (14) compares the profitability of holding a unit of home currency in the second period. The left-hand side denotes the value of holding a unit of home currency if the emigrant returns to the home country at probability μ and the right-hand side denotes the value also in home currency if the emigrant remains abroad at probability 1 - μ.

Given the above, the home country saving function is defined as a function of income abroad (w), the expected price of the C.V. escudo (E(e2)), the interest parity parameter (b), the probability of returning to the home country (μ), and the foreign interest rate (R1*). By taking partial derivatives, the function can be characterized as

∂ S h ∂ w > 0 ⁢ ∂ S h b < 0 ⁢ ∂ S h ∂ μ > 0 ⁢ ∂ S h ∂ R 1 * < 0 ⁢   a ⁢ n ⁢ d ⁢ ∂ S h ∂ E ⁢ ( e 2 ) < 0 ⁢ .

Transfers for consumption purposes, on the other hand, can be described as:

T = T ( b , R 1 * , w , E ( e 2 ) ) = α b R 1 * w ( 1 + α ) b R 1 * + β [ ( 1 − μ ) ( b − E ( e 2 ) ) + μ ) ] . ( 15 )

Equation (15) gives ∂T/∂R1*>0, ∂T/∂w > 0 and ∂T/∂E(e2) > 0. With respect to b, the sign depends on the size of μ.

∂ T ∂ b < > ⁢ 0   ⁢ i ⁢ f ⁢ μ < > ⁢ ( 1 − μ ) ⁢ E ⁢ ( e 2 )

The condition above is directly linked to Sh<>⁢0; therefore, the above implies that

∂ T/ ∂ b ≥ 0 if S h ≥ 0, and ∂ T/ ∂ b < 0 if S h < 0 .

Case I. μ = 0: (Permanent) Given μ = 0, Sh = 0. Hence,

C 1 * = R 1 * ⁢ w β + R 1 * ⁢ ( 1 + α ) ( 16 )
T = α ⁢ R 1 * ⁢ w β + R 1 * ⁢ ( 1 + α ) ( 17 )

Thus ∂T/∂R1*>0, ∂T/∂w > 0.

Case II. μ = 1: (Temporary) Given Sh > 0, we get

S h = β ⁢ w ( 1 + α ) ⁢ b ⁢ R 1 * + β
T = α ⁢ b ⁢ R 1 * ⁢ w ( 1 + α ) ⁢ b ⁢ R 1 * + β ( 18 )

Equation (18) gives ∂T/∂R1*>0 and ∂T/∂w > 0; ∂Sh/∂w > 0, ∂Sh/∂b < 0 and ∂Sh/∂R1* < 0.

ANNEX II

Estimation Results

ln Remt = α0 + α1Yt + α2(ΔRt) + α3et+1 + α4(D85) + єt

  • Rem: Remittances (millions of C.V. Escudos) in log

  • Y: Income dispersion measure

  • ΔRt Interest rate differential

  • et+1 Next period exchange rate

  • D85 Dummy variable to distinguish the introduction of special depost accounts.

article image
article image

APPENDIX Cape Verde: Summary of Tax System, 1995 1/

article image
article image
article image
article image
article image
article image
article image
article image
article image
article image
article image
article image
Source: Ministry of Finance, Department of Contributions and Taxation.

Effective January 1, 1996, the Unified Income Tax (annexed) replaced the Tax on Business Profits (1.1), Tax on Salaries and Income Self-Employment (1.2) and the Complementary Tax (1.3).

Capa Verde: Summary of Tax System, 1996

article image
article image
article image
article image

References

  • Chaney, Rick “Regional Emigration and Remittances in Developing Countries: The Portuguese Experience,” Prager Publisher, 1986.

  • Glytsos, Nicholas, P., “Remittances in Temporary Migration: A theoretic Model and Its Testing with the Greek-German Experience,” Weltwirtschaftliches Archiv, 1988.

    • Search Google Scholar
    • Export Citation

  • Glytsos, Nicholas, P., and Louka Katseli, “Theoretical and Empirical Determinants of International Labor Mobility: A Greek-German Perspective,”- Center for Economic Policy Research, Discussion Paper Series No. 148, October 1986.

    • Search Google Scholar
    • Export Citation

  • Lucas, Robert E., and Oded Stark, “Motivations to Remit,” Migration and Development Program Discussion Paper Series No. 10, 1984.

  • Miranda, Kenneth “Workers’ Remittances, Commodity Aid Utilization and Exchange Rate Unification,” IMF, January 1988.

  • Oberai, A.S., and H. Singh, “Migration, Remittances and Rural Development: Findings of a Case Study in the Indian Punjab,” International Labor Review, Vol. 119, No. 2, 1980.

    • Search Google Scholar
    • Export Citation

  • Serageldin, I., Socknat, S. Birks, B. Li and C. Sinclair, “Manpower and International Labor Migration in the Middle East and North Africa,” Research Project on International Labor Migration and Manpower in the Middle East and North Africa, World Bank, 1981.

    • Search Google Scholar
    • Export Citation

  • Straubhaar, Thomas, “The Determinants of Workers’ Remittances: The Case of Turkey,” Weltwirtschaftliches Archiv 122, October 1986.

    • Search Google Scholar
    • Export Citation

  • Swamy, Gurushri, “International Migrant Workers’ Remittances: Issues and Prospects,” World Bank Staff Working Papers, No. 481, August 1981.

    • Search Google Scholar
    • Export Citation
1/

Between December 1995 and February 1996, parliamentary, municipal, and presidential elections confirmed the Movement for Democracy for a second term in power.

1/

Official estimates indicate that the real GDP growth rate averaged more than 5 percent during the 1980s, but fell below the average 2.7 percent population growth rate in 1990-91, owing mainly to two consecutive years of decline in the agriculture and commerce.

2/

During the 1992-95 period, the public investment program (PIP) rapidly rose from 22.3 percent of GDP in 1991 to well over 30.5 percent in 1994-95 (Table 8). The authorities directed the PIP to provide the basic social services--energy, water, and sewerage education and health services--necessary to meet the rising needs of the population; and to integrate the Cape Verdean economy into the world economy--providing transport and telecommunications infrastructure, as well as air and maritime transport. Investments in basic social services--schools, generation plants, and desalinization facilities--rose from 4.9 percent of GDP in 1991 to some 9.2 percent in 1995. The expansion of the ports of Praia, São Vincente and Porto Grande, and the technological upgrading of the telecommunications company (CTT) accounted for the expansion of transport and communication investment from 2.4 percent of GDP in 1991 to an average 5.5 percent in 1992-94.

3/

Average inflation exceeded 11 percent in the 1980s and some 9 percent in 1990-91.

4/

Some agricultural imports were restricted and their prices administratively set. Other agricultural products were subject to seasonal quotas (see Section 5c.).

1/

Since 1992, the DGE has been producing a CPI based on the 1989 consumption basket, while the CPI produced by the BCV is based on the 1983 consumption basket and constitutes the only available database for completing historical price series.

2/

These data, however, do not seem to adequately represent the strong increase in self-employment and employment in the informal sector, which has took place in the urban centers after the trade liberalization of 1992. A more comprehensive definition of employment used by the Institute of Employment and Vocational Education estimated, for early 1996, an unemployment rate of 24 percent in Praia and 30 percent in São Vicente.

1/

Section III.3a contains a discussion on wage developments in the tradeable sector in 1990-95.

1/

In 1990, technical assistance by the Fiscal Affairs Department of the Fund provided the analytical framework of this tax reform: “Cabo Verde: Medidas para melhorar a Eficiência Económica e a Efectividade Administrativa do Sistema Fiscal.”

1/

These figures are based on the civil servant census of August 1993.

1/

During the period 1981-90, capital expenditure sharply declined from its historical high of more than 40 percent of GDP of 1983 to a minimum of some 12 percent of GDP in 1990. This decline followed a sharp decrease in external financing, particularly grants. Concessional loans accounted on average for 25 percent of foreign financing, and domestic financing of capital expenditure never exceeded 1 percent of GDP.

1/

Until 1995, the Central Bank also set sectoral credit ceilings. These were discontinued in 1996.

2/

The ratio of GDP to broad money stock has been steadily falling in Cape Verde since the mid-1980s. Since the authorities projected money demand growth to be equal to the expected growth rate of the economy, the decrease in money velocity provided room for a modestly higher-than-targeted domestic assets outturn, as well as the expected net foreign assets results.

1/

All the percentage growth rates presented in this section are in terms of beginning-of-period broad money, unless otherwise specified.

2/

The nonperforming loans to public enterprises in the BCV’s portfolio, amounting to C.V. Esc 2,017 million, were taken over by the Government which issued treasury securities as a counterpart. At the same time, additional treasury securities were subscribed by the BCA as an investment of its liquid resources, bringing total Ots subscribed in the operation to C.V. Esc 5,027 million.

1/

The only commercial banking function remaining at the BCV was a small recovery unit responsible for collecting residual delinquent private sector commercial loans that were not transferred to the BCA.

2/

Decree Law 52-D/90 of July 4, 1990.

1/

The draft proposed reducing the ceiling of MCE borrowing from the BCV to 5 percent, with borrowing permitted only in exceptional circumstances.

2/

Draft “Lei Organica do Banco de Cabo Verde.”

3/

The balance sheet of the BCA presents several deficiencies in accounting and internal controls. These deficiencies are being addressed with the assistance of an international auditing agency in the context of the ongoing World Bank Capacity Building Project for Private Sector Promotion (Box 2).

4/

Reserves with the Central Bank as a ratio to domestic and foreign liabilities.

1/

When in early 1996 the BCV detected this increase in credit, it claimed that the BCA had overrun its credit ceiling. A dispute arose, with the BCA measuring net credit flows and the BCV measuring on a gross basis.

2/

In July 1995, the mandatory reserve ratio was increased from 12.5 percent to 17 percent.

3/

Profits as a ratio of capital and reserves.

4/

The balance sheet of the INPS did not conform to actuarial accounting practices and the figures presented in this section could be subject to relevant methodological revisions.

1/

As of May 1996, the ISCV had transferred its responsibilities to the BCV but it remained to be decided whether or not the Ministry of Labor should have also relinquished its supervisory responsibilities of the INPS.

1/

In 1995, the interest rate on emigrants’ deposits denominated in local currency remained administratively set at 12 percent (with a 2 percent subsidy provided by the Government to the bank). Foreign currency deposits by nonresidents were paid interest on the basis of rates applicable for such deposits in the countries issuing the respective currencies, augmented by a small margin. In February 1996, the interest rate on six-month deposits and the interest rate on emigrants’ deposits in local currency were liberalized; however, the Government still provides a 1 percent subsidy on the latter.

1/

The choice of currencies as well as each currency’s respective weight in the basket is revised annually, using a formula that gives a two-thirds weight to imports and a one-third weight to remittances. In 1995, the basket was composed of the following currencies: the Portuguese escudo (with a weight of 40.330 percent), the Dutch guilder (15.041 percent), the Swiss franc (8.027 percent), the Swedish krona (6.616 percent), the French franc (6.066 percent), the Deutsche mark (1.852 percent), the British pound. (1.603 percent), the Japanese yen (1.458 percent), and the Italian lira (1.339 percent).

1/

Foreign entrepreneurs, who recently established their activities in the Free Trade Zone of Mindelo, cited the political stability of the country, the discipline of the labor force, and its ability to absorb training as the primary reasons for having chosen to invest in Cape Verde. Wages were seen as being somewhat higher than elsewhere in the region, but in line with the productivity of the Cape Verdean labor force.

2/

In August 1991, the import tariff structure was simplified from 79 rates, ranging from 5 to 100 percent, to 11 ad valorem rates, ranging from 5 to 50 percent (with the majority of products following in the 10-20 percent range). Furthermore, the system of consumption taxes on imports was simplified by reducing the number of rates from 16 to 4, with a range from 10 to 60 percent.

1/

The quotas were enforced through foreign exchange rationing. For each product subject to quota, a maximum value of imports was established and foreign exchange correspondingly restricted.

2/

Cape Verde will however maintain its quantitative restrictions on some agriculture commodities.

1/

Foreign loans to public enterprises returned to positive levels in 1994 in relation to the acquisition of two airplanes by TACV (airline services).

1/

In response to the difficulties of agricultural and fishing companies in obtaining credit, and in an effort to stimulate rural development, the Government authorized the establishment of the Caixa de Crédito Rural (CCR). This has an authorized capital of C.V. Esc 180 million to be totally subscribed by the Government. The capital of the CCR has yet to be paid to allow it to become operational and no definite date for starting operations has been established.

1/

The reliance on data on the level of nominal wages as the basis of comparison fails to take into account variations in labor productivity in the competitor countries, and therefore does not provide a basis for comparing industry-specific indicators of unit labor costs. However, this approach is dictated by the lack of current and comparable data for the countries considered. Nevertheless, it may be considered justified, given that in discussions with managers of the firms that have recently started operations in Cape Verde, they stated that the relative productivity of the Cape Verde labor force in the industries considered compares favorably with the productivity of workers in the home countries of the investors and other potential investment locations.

2/

The CPI-based real effective exchange rate is the principal index used for most developing countries, chiefly because of data availability and comprehensiveness of coverage, not because of its inherent superiority as an indicator of export competitiveness.

3/

An alternative approach, which seeks to determine the absolute competitiveness of the Cape Verdean economy, would require a judgment as to the equilibrium REER. The equilibrium REER concept, defined as the value that is consistent with achieving and sustaining internal and external balance over the medium term, is not measurable with currently available data due to the availability of labor unit cost index for Cape Verde, and thus will not be used in this analysis.

1/

Ghana, Kenya, Madagascar, Senegal, and Togo.

2/

China, Indonesia, Malaysia, Philippines, and Thailand.

3/

The main noneconomic factors that reportedly entered into the decision included political stability, and relative social tranquility.

4/

Wage data for the lowest paid categories are more relevant for the present analysis because the newly established industries mainly employ unskilled workers, who constitute some 30-40 percent of the active work force in Cape Verde.

1/

While beyond the scope of the present study, it is significant to note that, according to data available from the OECD, relative unit labor costs indices for Italy and Spain (defined as the ratio of unit labor costs in the subject country divided by the unit labor costs of main partner countries weighted by manufacturing trade) fell by about 24 percent and 17 percent, respectively, from 1991 to 1994 (Table 46). This would suggest that, by taking wage developments in Cape Verde as a proxy for unit labor costs (admittedly a very strong assumption), the relative competitiveness of Cape Verde vis-à-vis Italy and Spain fell over the period.

1/

A determination of the relative competitiveness between different economies requires a certain degree of comparability. The economies of the African comparators are different from that of Cape Verde in that these countries rely heavily on the export of primary commodities. The implicit justification for the comparison with these countries is that each of these countries will continue to pursue policies aimed at diversifying their export base, and in so doing will seek to attract investments in the same type of labor-intensive light industries as Cape Verde.

1/

The definition of emigrants’ remittances was changed in 1981 when transfers in currency and goods began to be treated separately. Therefore, the data in the early 1980s should be analyzed with caution.

1/

Saving here addresses the emigrant and his family’s future consumption only. Remittances can be also used as saving for other Cape Verdeans who could receive the remittances rather than emigrate themselves; however, saving in this study is limited only to future consumption of the emigrant himself and his family.

1/

To be more precise, there are substitution effects with respect to interest rates since a higher interest rate implies that a unit of currency today is relatively more valuable than next period’s which induces less consumption in the first period. Under this model’s specification of the log preference, however, the share of each type of consumption remains unchanged so that only the positive income effect will affect the decision.

2/

If the amount of the consumption transfer is fixed in C.V. escudos and therefore independent of any emigrant maximizing problem, then the exchange rate plays a crucial rule. The lower the C.V. escudo value (higher value of e1) the lower will be the amount of transfers in foreign currency. However, in C.V. escudos this is neutral.

3/

Again the intertemporal substitution between present and future consumption need not to be considered under this model’s specification. The share of today’s and tomorrow’s consumption would remain constant, thus the substitution effect is null.

1/

No accurate data on the number of emigrants are available. A survey conducted by the authorities in the late 1980s shows that more than 50 percent of the emigrants are in the United States, followed by Portugal, Italy, France, and the Netherlands. Also, among the African countries, Senegal and São Tomé e Principe account for a significant share.

1/

Deposit rates for six months or longer are used.

2/

A separate estimation on remittances from the United States could not be carried out, owing to the limited number of degrees of freedom.

Same Series

Other IMF Content

Other Publishers

  • Collapse
  • Expand
Cover IMF Staff Country Reports
Cape Verde: Recent Economic Developments
Author:
International Monetary Fund
Volume/Issue:
Volume 1996: Issue 103
Publisher:
International Monetary Fund
ISBN:
9781451809237
ISSN:
1934-7685
Pages:
141
DOI:
https://doi.org/10.5089/9781451809237.002
Chart 1.

GDP GROWTH
Chart 2.

DOMESTIC EXPENDITURE
Chart 3.

GDP BY SECTOR
Chart 4.

REAL SALARIES BY SECTOR
Chart 5.

FINANCING
Chart 6.

REVENUE
Chart 7.

CURRENT EXPENDITURE
Chart 8.

CAPITAL EXPENDITURE
Chart 9.

BANKING ASSETS
Chart 10.

BANKING LIABILITIES
Chart 11.

DOMESTIC PUBLIC DEBT
Chart 12.

PRIMARY SUBSCRIBERS
Chart 13.

REER AND TERMS OF TRADE
Chart 14.

EXPORTS AND IMPORTS
Chart 15.

CURRENT ACCOUNT
Chart 16.

CAPITAL AND FINANCIAL ACCOUNTS
Chart 17.

FINANCING
Chart 18.

FOREIGN EXCHANGE RESERVES
Chart 19.

EXTERNAL DEBT
Chart 20.

DEBT SERVICE RATIOS
CHART 21

CAPE VERDE COMPOSITION OF EXPORTS, 1991–95
CHART 22

CAPE VERDE AND SELECTED COUNTRIES REAL EFFECTIVE EXCHANGE RATES, 1990–95
CHART 23

CAPE VERDE AND SELECTED COUNTRIES REAL EFFECTIVE EXCHANGE RATES, 1980–95
CHART 24

CAPE VERDE AND SELECTED COUNTRIES REAL EFFECTIVE EXCHANGE RATES, 1985–95
Chart 25.

REMITTANCES
Chart 26.

REMITTANCE SHARE BY COUNTRY
Chart 27.

REMITTANCES
Chart 28.

INCOME DISCREPANCY WITH CAPE VERDE
Chart 29.

REAL INTEREST RATE DIFFERENTIAL