Colombia - Economic development and policy under changing conditions (Vol. 1 of 2) : Main Report

jLE CO,V Report No. 4444-CO Colombia: Economic Development and Policy Under Chlanging Conditions (In Two Volumes) Volume 1: Main Report August 5, 1983 Latin America and the Caribbean Regional Office FOR OFFICIAL. USE ONLY Document of the V*wbd Bank This document has a resticted distribution and mnay be used by recpients only in the performarce of their official duties. Its contents may not otherwise be disdosed without World Bank authorization.  CURRENCY EQUIVALENTS Currency Unit = Colombian Peso (Col.$) Exchange Rate As Of July 6, 1983 US$1 = Col$78.61 Col.$1 = US$0.0127 Average Exchange Rate 1980 1981 1982 US$1 = Col$47.280 US$1 = Col$54.491 US$1 = Col$64.102 Col$1 = US$0.0212 Col$1 = US$0.0184 Col$ = US$0.0156 WEIGHTS AND MEASURES Metric System GOVERNMENT OF COLOMBIA FISCAL YEAR January 1 - December 31 ft  FOR OFFICIAL USE ONLY This report is based on the findings of an economic mission which visited Colomb:ia during June/July 1982. The mission was comprised of Messrs. Jose B. Sokol (Chief of Mission), Vinod Thomas (Deputy Claief), James Hanson, Ricardo Moran, Desmond McCarthy, Charles McClure, Jr. (consultant), William Dougan (consultant), Oswaldo Schenone (consultant), Ms. 'Isabel Guerrero and Ms. Consuelo Cruz. The mission also benel.ited from consolidated public sector accounts prepared by A. Antonaya (IMF), and from background papers in agriczulture provided by M. Blanc. This document has a restricted distribution and may be used by recipients only in the perof their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  COLOMBIA: ECONOMIC DEVELOPMENT AND POLICY UNDER CHANGING CONDITIONS TABLE OF CONTENTS Page No. VOLUME I - THE MAIN REPORT COUNTRY DATA SUMMARY AND CONCLUSIONS ..................................... i-xxii CHAPTER I. FEATURES AND SOURCES OF GROWTH 1 A. Long-term Trends .................... 1 Overall Performance ...................1............... Growth Trends ......................................... 2 B. Stractural Change ....................... 4 Supply Side: Sectoral Shares ..... .............. 4 Demand Side: Changes in Expenditure .................. 5 C. Sources of Past Growth ...... ................ 6 Factors of Production ...... ................ 6 D. Sumnary and Conclusions ...... .......................... 10 CHAPTER II. DEMOGRAPHIC TRENDS AND EMPLOYMENT 12 A. Demographic Change .................... ................. 12 Population Growth Patterns ............................ 12 The Fertility Decline ............... .. ............... 13 Family Planning Services .............. ... ............ 15 Urbanization and Migration ............. ....15 ......... B. Trends in Labor Supply and Employment . .................. 17 An Overview ............................. ............ 17 Labor Use Indices ....................... 18 Sex Differences ............................... 20 Trends in Metropolitan Areas . . 21 C. Sectoral Sources of the 1973-78 Employment Boom 24 D. Sumaary and Conclusions ................... .......... 26 CHAPTER III. WAGES, INCOMES AND POVERTY 29 A. Spaltial and Occupational Comparisons .................. 29 Rural Wages .......... . 29 Urban Wages .................................. 30 Relative Wages Rural/Urban and Occupational . .......... 32 B. Differences Among Income Classes ..................... 33 C. Rich and Poor Families in Cali, 1970-80 ......... 34 Family Income ..................................... 34 Other Welfare Indicators ................ ............... 35 A View of Income Distribution ..................... ... 36  -2- Page No. D. Evidence from Household Surveys ........................ 36 E. A Summary of Trends in Poverty and Income Distribution . 37 F. Policy Framework for Redistribution: Some Implications. 38 Macroeconomic Issues ................................. .38 Sectoral Issues ......... ............................. :38 CHAPTER IV. ECONOMIC GROWTH AND PRICE STABILIZATION 40 A. The Historical Experience with Macroeconomic Policies 40 A Background: 1962-66 ............... .. .............. 40 The 1966-70 Development Plan ......................... 42 The 1970-74 Development Plan ......................... 43 The 1974-78 Development Plan ......................... 44 The 1978-82 Development Plan ......................... 47 B. Trade-off Between Output Growth and Inflation .......... 51 C. Replicating the 1967-74 Growth Experience: Policy Implications ................................ 53 Exports .............................................. .53 Imports .............................................. 54 Trade Incentives ..................................... 54 Financial Policy ..................................... 55 Fiscal and Monetary Considerations .... ............... 56 Policy Coordination and Direction ..... ............... 56 CHAPTER V. FINANCIAL POLICIES FOR DEVELOPMENT 58 Introduction .58 A. Government Policies and the Financial System, 1967-82 59 Gradual Liberalization, 1967-72 .59 The Financial Reform of 1974-75 .60 The 1977 Stabilization Program. 151 Recent Developments, 1978-82 .53 B. Interest Rates and Government Policy ................... 154 Interest Rates on Deposits ........................... 64 Interest Rates on Credit ............................. 66 C. Resource Mobilization .................................. 67 Saving and Investment ................................ 67 Resource Mobilization Through Financial Institutions.. 70 D. Credit and Its Allocation ....... ....................... 71 Institutions, Terms and Sectors ...................... 71 The Central Bank and PROEXPO ......................... 72 E. Conclusions and Recommendations ........................ 73 CHAPTER VI. PUBLIC EXPENDITURE AND RESOURCE MOBILIZATION 76 A. Public Sector Expenditure ............... .. ............. 77 Overview of the Problem ............... .. ............. 77 Public Sector Budgeting ............... .. ............. 78 Fiscal Federalism ....... .............................. 80  -3- Page No. B. Public Sector Investment ................................ 82 Education . ........................................... 83 Health ........................................... 83 Electrical Power ..................................... 84 C. Revenues: Indirect Taxes .............................. 85 Sales Tax ............................................ 87 Gasoline Tax ......................................... 88 Coffee Tax ........................................... 89 D. Revenues: Income and Complementary Taxes .... .......... 89 Introduction ......................................... 89 Major Reforms Since 1968 ............................. 91 Analysis of Taxation of Individuals .................. 92 Analysis of Taxation of Business Income .... .......... 93 Evasion and the Need for Administrative Reform ....... 96 Revenue Effects ...................................... 98 Conclusions and Recommendations ...................... 99 CHAPTER VII. AGRICULTURAL DEVELOPMENT AND STRATEGY 101 A. Sectoral Performance and Policy ........................ 102 Composition of Production and Trends ........ .. ....... 102 Elements of Growth and Problems ........... .. ......... 103 Outlook and Strategy ................ .. ............... 106 B. Input Provision, Investment and Technology ...... ....... 106 Water Development and Management .......... .. ......... 107 Fertilizer ....................... ..................... 107 Research and Extension ............... .. .............. 107 Marketing ........................ ..................... 108 Agricultural Credit ................. .. ............... 109 C. Competitiveness and Incentives ............ .. ........... 110 Demand Considerations .......................... ... 110 Input Pricing: Fertilizer ........................... 113 Trade Policy ...................... .............. 114 The Special Case of Coffee ............................ 114 D. Institutional Constraints .......................... -... 116 CHAPTER VIII. ISSUES AND POLICIES IN MANUFACTURING INDUSTRY 118 A. Past Trends and Recent Developments ................. .... 118 Production and Exports ................................. 118 Composition of Output and Change ...................... 119 Current Problems ...................................... 120 B. Foreign Trade Strategy and Industry ............... 121 Import Substitution and Export Promotion: The Experience ....................................... 121 Trade Incentives and Policy Options ... ............... 122 C. Sector Developments and Specific Issues ................. 126 Investment, Productivity, Wages and International Competitiveness ..................................... 126 Textiles ....... ...................... 127 The Metal-Mechanic Industries ........................ 129  -4- Page No. CHAPTER IX. RECENT DEVELOPMENTS AND OUTLOOK 132 A. Recent Economic Developments .............. 132 B. Development Issues and Strategy ........... .. ........... 132 C. Output and Growth ..................... 136 Macroeconomic Objectives .............. .. ............. 136 D. Public Sector Investment Program ........... ............ 137 E. Balance of Payments ................... ................. 140 F. External Capital Requirements ................... 145 G. Debt Management and Creditworthiness ................ 145 MAP  Page 1 of 2 WUINiY DATA - C0LaXBIA AREA POPULATION ISLiSrt 1,138,900 sq.km. (total) 26.9 million (mid-1982) 23.6 per sq.km. 232,000 sq.km. (arablea) Rate of growth: 2.C% (frcs 19J5 to 1982) 116.2 per sqlam. of ara'le land POPULATION CHAACTERISTICS I/ IEAL- 1/ Crude birth rate (per 1,000) 30 Population per physician 1,967 Crude death rate (per 1,000) 8 Population per hospital bed 619 Infant mortality (per 1,000 live births) 2/ 56 IUflf DISTRIBUTION (1978) DISM=IfIONXT OF IAND 02TCIP (1971) % of national incane, highest 1Cr 40 % owned by top 10% of owners 80.0 lowiest 2G% 5 % owne by smallest 10% of owners 0.2 ACES SAFE WMER 2/ AcaSS TO ]IRICITY (1973) % of population - urban 73 % of pciiultion - urban 87.3 - rural 46 - rural 13.2 - total 64 NURITION EXCATION Caloric intake as % of requirements 98 Adult literacy rate (1973) 80.8 Per capita protein intake (grams per day) 49 Primary sdaol enrollment ratio 2/ 128.0 GNP PER CAPITA IN 1981 3,' US$1,380 GROSS NAIIONAL PR(DUCT IN 1982 4/ ANNJAL RAlE OF G1CLWI (%, Constant Prices) US$ MLn. % 1960{,5 1965-70 1970-7'5 1975-80 1982 GNP at Market Prices 38,969 100.0 4.5 5..5 6.6 6.0 1.4 Gross Domestic Investmenl: 10,224 26.2 1.8 8.L 1.3 10.3 - 3.1 Gross National Saving 7,981 20.5 2.1 10.8 6.5 10.5 - 2.5 Current Account Balance - 2,243 - 5.8 Exports of Goods, NFS 4,679 12.0 2.1 4.[ 4.4 7.3 - 6.0 Imports of Goods, NFS 6,444 16.5 1.9 10.2 0.8 11.9 - 9.0 urPTr, LAB(FO FORCE AND PRDODCrIVrrY IN 1980 Value Added Iabor Force 2/ V.A. per Worker US$ Mln. % WML. Z US$ % Agriculture 8,181 27.6 2.3237 25.8 3,521 106.9 Industry 9,108 30.7 1.909(l 21.2 4,770 144.8 Other 12,381 41.7 4.773; 53.0 2 ,594 78.7 Total Weighted Average 29,670 100.0 9.0068 100.0 3,294 100.0 GWE1RET FINACE Public Sector Central Govermant (Col$ Mln) % of GDP (Col$ Mln) % of GDP 1981 1981 1976-1980 1981 19131 1976-1980 Current Revme 561,634 28.0 24.5 261,499 13.0 14.3 Current Expenditure 398,919 19.9 17.0 98 474 4.9 7.8 Current Account Surplus 162,715 8.1 7.5 163,025 8.1 6.5 Capital Expenditure 148,933 7.4 7.1 90,914 4.5 3.9 External Borrowing (net) 35,091 1.7 1.3 1/ Between 1978 and 1982. 2/ Between 1978 and 1980. 3/ The per capita CNP estimate is calculated by the conversion technique in the World Bank Atlas. Other conversions to dollars in this table are at the average exlange rate prevailing durirg the period covered. 4/ Preliminary. Not applicable. N.Lt available.  Page 2 of 2 CCuM DM - CX4IA ?NEY, CREDIT AND PEICES 1976 1977 1978 1979 1980 1981 1/ (Eiiiona of ColS outstadling ed priod) Mouey ard Qiasi MDsy 105.1 140.7 181.7 224.6 325.3 440.8 Bank Credit to the Public Sector 15.5 19.2 16.1 - 5.66 -15.41 - 6.75 Bank Credit to tbe Private Sector 109.2 142.3 149.3 181.6 269.3 361.1 (Percentages or index tnbers) Mony and Qlasi !'rv ars % of GDP 19.7 19.6 19.8 18.8 20.5 22.0 Consumer Price Index (December 1978 100) 69.5 85.7 100.0 124.9 158.9 203.6 Amual percentage dcanges in: Consumer Price Irlex 19.9 34.7 16.7 24.9 27.2 28.1 Dbcey and Quasi Dbney 33.9 33.9 29.1 23.6 44.8 35.5 Bank Credit to the PRblic Sector - 2.3 23.9 -16.1 -35.2 272.3 -56.2 Bank Credit to the Private Sector 23.3 30.3 4.9 21.6 48.3 34.1 BA"ANCE CF PAMIEWS (Million US$) NMtANDISE EXCRIS (Average 1978-81) 1979 1980 1981 1982 1/ TS$ ba % Exports of Goods, NFS 4,658 5,677 4,606 4,679 Coffee .1,929 51.0 ILports of Goods, NFS 3,939 5,494 6,078 6,444 Mjor Non-Coffee Agriculture 517 13.7 Resource Gap (Deficit = -) 719 183 -1,472 -1,765 Major Mamufactured Goods 610 16.1 Petroleun Derivatives 217 5.7 Factor Service Incoe (net) - 255 - 210 - 428 - 701 All Other Goods 510 13.5 Receipts (267) (494) (647) (496) Paym-ts (522) (704) (1,075) (1,197) Total 3,783 100.0 Net Transfers 98 164 242 223 Balxnce cn zrrent Account 562 137 -1,658 -2,243 EXRAL EBT, DECEXBE 31, 1981 1, Direct Foreign Inestlment 104 52 228 268 Net MLT Borrcming 609 779 1,450 1,168 US$ KDn. Disbxrsement (1,102) (1,156) (1,972) (1,610) Amrtization (-493) (-377) (-522) (- 442) Public Debt, including 5,076 Capital n.e.i. (net) - 38 200 92 1 guarateed Increase in Reserves (-) 2/ -1,237 -1,168 -112 6 ln-guarnteed Private Debt .. Total O(tstarding and Disarsed 5,076 eoss Reserves (end year) 3, 4,113 5,420 5,633 4,894 Net Reserves (enld year) 3/ 4,106 5,416 5,630 4,892 3BULIC IBT SERVICE RATI0 F(Cl 1981 4/ % Amortization 6.0 Interest 10.3 Total 16.3 FEE OF EXHANG IBRD/IMk IENDINS, (December 31, 1981) (Million US$) Doember 31, 1981 IBRD 1k US$1.00 = Col$59.07 Col$1.00 = US$0.0169 Outstamding and Disixwrsed 1,165 21 Ukxiisbursed 1,073 - Deae,ber 31, 1982 Outstanding Including Uhdisbursed 2,238 21 US$1.00 = Col$70.29 CD1$1.00 = US$0.0142 1/ Prelimnary data. Ct C the Banking System. 3/ Of the Central Bank. 4/ Ratio of Public Debt Service to Exports of Goods and al Services. .Nt available.  COLOMBIA: ECONOMIC DEVELOPMENT AND POLICY UNDER CHANGINGr CONDITIONS SUMMARY AND CONCLUJSIONS i. Over the past two decades, Colombia has achieved significant gains in the standard of living of its people. There has been great success in sustaining rapid economic growth over a long period of time. The economic performance has been especially imapressive because a wide cross-section of the population, in rural areas and in the urban centers, have been able to share the fruits of development more fully than in many other developing countries. GNP per capita has reached roughly US$1,380, the economy has become more urbanized, industrialized and diversified than in the past, and in good measure more resilient to external shocks. ii. On the other hand, the country has experienced a significant slow- down in economic activities in more recent years. In many ways, the economy and its economic policy are at a crossroads: difficult decisions regarding exchange rat:e, commercial, fiscal, monetary, financial and investment poli- cies need to be made with urgency. A variety of favorable ext:ernal factors which fueled past growth are less likely to be at hand in the future. Coffee, a major source of the historical expansion, faces unpromising pros- pects,, and other minor exports are in dire need of a stimulus, particularly in the face of the recent downturn in the world economy. Meanwhile, domestic production costs have mounted, adoption of new technologies has lagged and supply prob:Lems are being encountered for many commodities. [n 1981 and 1982 growth has dipped, and macro-economic management has been presented with challenges of trade deficits and fiscal deficits in magnitudes unprecedented in recent Colombian history. Economic policies are now neededl which will contribute to a revitalization of development at a time when many old prob- lems have b(een solved, some others remain and new constraints are emerging. There are difficult issues ahead; nevertheless, it ;is within the Colombians' capability Ito tackle them successfully. Structure of the Report iii. This report provides a survey and analysis of Colom'bia's develop- mental experience and its principal features. The purpose of the review is to identify factors that have contributed tc past performance, to set recent developments and prospects in better perspective and to assist in framing the relevant issues and questions for policy formulation. The first three chapters trace major trends in economic growth, demographic factors, incomes and poverty, highlighting their salient characterislics and sources. The following two chapters are devoted to analyEes of macro-economic and financial policies that have, directly or indirectly, influenced past events, bringing out needed redirections in these areas. Niext, the public finances are treated in some detail, showing how their management can and does determine the flexibility and effectiveness with which economic policies  - ii - affect outcomes. A review of two key productive sectors--agriculture and industry--is presented in the subsequent two chapters, underscoring factors that have shaped past and recent developments and proposing a framework for policy. In the last chapter, which is concerned with recent events and with the current outlook, it is also attempted to reach some conclusions about future strategic issues for Colombia's development policy. Overview of Long-Term Trends iv. During 196C-81 the economy expanded at a pace of about 6% annually in real terms, with rates varying from an estimated 6.7% during 1967-74 to 5.4% in 1974-78 and to over 3% from 1978 to 1982. In spite of rapidly falling infant mortality and increasing life expectancy, the population growth rate has dropped from 3.1% at the beginning of the 1960s to about 2% at present on account of the spread of education, urbanization, greater female participation rates in the labor force and an active birth control campaign. GDP per capita is estimated to have grown by over 3% annually over the past two decades. v. In contrast to total population, the working age population was growing at about 3.4% annually up to the late 1970s, contributing to a worsening of labor market indices during the 1960s and early 1970s. Since tlhen, however, a remarkable reversal has taken place: from 1973 to 1978 employment is estimated to have expanded at a phenomenal 6% annually. Growing non-agricultural and urban activities and rising wages have given rise to rapid rural-urban migration. As a result, population pressure in rural areas is no longer increasing much, if at all. Rising per capita income, increased employment and expanded public services have brought significant improvements in the welfare to the poorest income groups in absolute and relative terms. vi. During the 1960s the annual inflation rate averaged less than 12%, in contrast to rates of about 21% and 25% since 1970 and 1975 respectively. The financing of increased Central Government overall cash deficits and the monetization of rising foreign exchange earnings after 1974 kept inflation at high levels despite government efforts at sterilizing a part of these earn- ings. A large part of this reflected rising food prices, resulting in part from supply problems. Food prices have been a major component of increasing consumer prices, given a weight of about two-fifths of food in the index. Ilistorically, the Central Government's budget had normally been in surplus. But as each succeeding administration expanded its programs and increased the coverage of much needed services and infrastructure without matching tax and pricing efforts, deficits of about 1% of GDP or more emerged during 1970-74 and also since 1980. vii. Prior to 1967, Colombia faced periodic balance of payment crises-- produced in some instances by falling coffee prices, and in others by the combination of large devaluations and import liberalization (para. 4.04)--to  - iii - which the Government responded with varying degrees of import controls. An outward-looking strategy was adopted in early 1967 (para. 4.05), combined with the expansion in the world economy and cdomestic cdemand, t'his turned the situation around during 1967-74. In the second half of the 1970s, the unexpected boom in coffee and illegal crops sustained growth and exports which together with capital inflows further increased international reserves and the debt-servicing capacity of the country. Trade deficits emerged in the last few years on account of declining coffee prices, a deteriorated real exchange rate and the international recession. Structural Clhange viii. Agriculture, manufacturing and trac.e have been the major contributors to aggregate supply, responding strongly to export incentives and expansions in domestic demand. Variations in the composition of demand, as well as e:xchange rate and trade policies, government investment and inflation, have affected the shares of these and other sectors in the GDP significantly. Aggregate demand has benefitt:ed from the participation of the private sector investment which in turn has periodically been stimulated by government policies. However, the stubbornly high share of food in total expenditure has constrained the demand for ncinfood commodities and the diver- sification of the economy. ix. Explanations of growth can be provided inter alia on the basis of changes in conventional factors of productiorL. Some rough econometric calculations suggest that increases in capital stock explain a'bout 50% of the growth durin,g 1963-80, out of which public and other construction accounted for about a third. Labor was responsible for about 25% of the economic growth, out of which over four-fifths arose f'rom the growth in the labor force and about one-fourth from more educaticin. The residual, 25% of the growth, incliudes the effects of other factors such as government policy, leading to improved resource allocation through favorable investment climate, export-oriented, fiscal, financial, and trade! and exchange policies. Implications of the Growth Experience x. Given the present rates of investme!nt, the capital stock is likely to grow at a'bout 4.5% annually. At the curre!nt incremental capital/output ratio, an additional 1% GDP growth is likely to require net investment to be stepped up by 2.5 percentage points. The allocation of investment can strongly affect outcomes: higher levels of construclion activities could stimulate growth. In particular, given the e!xisting unfulfilled demand for housing, efforts in this area could be effective without adding to fiscal and monetary pro'blems provided the indexed saving and housing finance system (UPAC) continues to mobilize additional resources for this purpose, as discussed su'bsequently (see paras. xxxiv through xxxix). To t'he extent that construction activities draw primarily on domestical'Ly-producedl resources with excess capacity or high supply elasticit:ies, and incomes thus generated  - iv - benefit (in the first round) relatively lower income people with low income elasticities for imported goods, major negative impacts on balance of payments can be averted; in view of the pressure on food expenditure, noted above, the achievement of better food supplies would be an important complementary goal. Even if the first round effect of more construction activities may not be large, the full impact after accounting for all multiplier effects, extending beyond the narrowly defined sector itself and also after spilling into imports, can be substantial. Over the longer term, investment in technological progress and cost reductions in productive sectors such as agriculture and industry are called for. xi. A gradual fall in the past growth rate of the labor force of about 3.4% is likely in the future, eventually converging with the population increase of about 2%. The education quality factor at the present rate adds a quality effect of about 1% annually to the labor force. Thus, as the rate of growth of the labor force declines, increases in the quality of labor are needed in order to maintain the historical contribution of human capital to growth. To achieve this, development expenditures are needed for educati'on, skills and job training, but the issue concerns not necessarily and not even primarily the size of the total expenditures, but their allocation and ef'fi- ciency of use. Additional momentum to growth will also have to rely on a greater contribution of the residual representing improvements in the econo- mic environment and policy reforms in trade, fiscal and financial matters. Demographic Trends and Employment xii. Long-standing patterns of human reproduction and mortality in Colombia have been significantly altered since the 1950s. The magnitude and speed of these changes are noteworthy especially since they took place in the absence of radical social change, or even strong population policies. These demographic events, moreover, have had considerable impact on labor supply, and their influence is likely to be felt for the remainder of the 1980s. The main force impelling the brisk transition has been the substantial improve- ment in economic and social conditions, especially in public health and eclu- cation, for a rising share of the population since the late 1940s. Govern- ment policies towards family planning, that moved from tolerance in the 15960s to moderate support of private initiatives in the early 1L970s, to more deci- sive interventions since the mid-1970s, added the impetus to fertility dec- line. Despite the sharp drop, Colombia's fertility level remains 84% above those in the industrial market economies, and how this remaining gap in demo- graphic transition is closed will significantly affect the future size of population and the rate of growth of labor force entrants. xiii. In the wake of a quickened demographic transition in the 1940s, which is surveyed in this report, growth of the potential labor force accele- rated greatly between the mid-1950s and early 70s. In the absence of a com- parable surge in labor demand, labor market conditions deteriorated. Then,  v as already noted, Colombia experienced during most of the 1970s what is seemingly one of the most extraordinary employment booms on record that brought labor market indices to their most favorable Levels since 1950. Moreover, the additional employment entailed in the boom was, ona average, at least as productive as pre-existing jobs among the lower-skilled workers. This development followed very rapid economic growth and coincided with continuing fast growth based on a significant expansion of labor-intensive production and exports in agriculture and manufacturing. xiv. In addition to sustained and substantial economic growth, other contributing factors to the 1970s employment expansion included: (a) growth in labor-intensive manufacturing for exports and the domestic market; (b) expansion in agricultural labor demand responding to the growth in sectoral output without- excessive mechanization: massive replantings of coffee with a new varLety and the illegal drug boom contributed heavily to the past employment generation; and (c) channelling of much of the income gains from the growth in manufacturing and agriculture towards the domestic tertiary sector. The vigorous employment response also suggest:s that the cost of labor, especially relative to that of capital, did not: escalate as in other Latin Amaericart countries, which in turn, owes in part to the Government's restraint with interference in the contractual relations between workers and employers. Most of these factors are still relevant ior future policies (the major exception being coffee and drug-related agricultural outpult). Wages, Incomes and Poverty xv. The issue of who has benefitted from growth has been 1the subject of much discussion. Based on a wide array of data, this report concludes, however, that the real incomes of the poorest workers improved significantly during most of the past decade. Some previous studies on the ol:her hand, have focussed upon the first half of the 1970s excluding the full redistribution impact of policies and developments which came to fruition by the mid to the second half of the decade; furthermore, the most commonly available statistics are incomplete and of variable quality. A more rigorous approach requires an analysis of the quality and coverage of various household surveys and comparisons of income distributions derived from surveys of similar quality. This report relies on such an approach, which is complemented by information on earnings obtained from periodic surveys of salaries and wages in manufacturing and in agriculture. xvi. The real wages of the rural and urban poor stagnated imtil the mid- 1970s, then improved significantly in the latter half of the decade, and have deteriorated somewhat in the past two years as the economy began to stagnate. Real earnings, however, improved more than the wage rates in view of greatly increased employment among household members, which wras more signific:ant in urban locations with greater opportunities for female employment. The income per member of poor urban families is likely to have  - vi - also benefitted from fewer babies in the 1970s as the fertility decline in urban areas was greater than in rural locations. xvii. Relative wages in low paid occupations tended to converge over the 1970s: those paid the least at the beginning of the decade gained; the better paid semi-skilled workers faced stagnant or falling wages; and higher paid workers suffered substantial setbacks. Factors contributing to the improvement in the lot of the poor include rapidly expanding educational opportunities since the early 1950s, reinforced by population, nutrition, and related social policies since the late 1960s. Strongly complementary was the employment boom of the 1970s which enabled increasing numbers of women and other secondary workers in the poorer households to find jobs. In the absence of strong performance on the part of either of these factors, alleviation of poverty in the 1970s would probably not have taken place. xviii. The economic position of the higher income groups also improved substantially over the 1970s. Rapid economic growth provided ample opportunities for rising entrepreneurial profits while substantial real interest rates were available to owners of capital. The decade also witnessed booming markets in coffee, emeralds, urban residential housing and illegal trade leading to substantial profits for more than a few well-positioned individuals; with collected taxes on some of these enhanced earnings remaining modest, it is not surprising that the higher income groups prospered as much as the evidence indicates. In comparison, those in the intermediate range of the income distribution fared poorly, which to some extent was the by-product of the surging supply of educated manpower during the 1970s. Policy Implications xix. This review shows that rapid economic growth without too much dis- crimination against agriculture or in favor of large capital-intensive indus- try can lead to employment generation and higher incomes for urban and rural unskilled workers. A reduction in excess agricultural labor is vital in lowering absolute poverty. A switch from a high protectionist strategy coupled with subsidies to some capital-intensive investment projects, to freer foreign trade and a less repressed financial system coincided with faster growth and decreases in unemployment. The shift to freer trade and lower interest rate subsidies was gradual, however, and at no time were exchange or financial restrictions dismantled. xx. Fiscal policy also contributed to decreasing income dispersion. Through a process of tax reforms, taxes have, by and large, remained progres- sive, and government expenditures more so. Public expenditures in education, health, housing, land reform, and public services have benefitted the poor more than proportionally, and the redistributive impact of the budget has probably increased through time. Furthermore, with some important excep- tions, public expenditure has not been a source of inflationary monetary  - vii - expansion and the Government has therefore noi: been a cause of regressive redistribution through inflation. Serious efforts have also been made to concentrate resources in areas with poor beneiiciaries. xxi. Human resource policies of the seventies (especially in the latter part of the decade) by and large appear to halte been effective. Similarly, although it would be easy to point to imperfections in the labor markets--the cesantia (see para. 3.33) is perhaps such a slhortcoming--when compared to other Latin American countries, the overall impression is favorable. The degree of flexibility of labor markets, particularly at the enterprise level, in responding to a downturn in the economy, nevertheless, needs to be further examined in view of the more recent experience. It would appear, in general, that in human resource and labor market policies, marginal changes, corrections and fine-tuning rather than radical departures would constitute a recommended approach. Growth and Stabilization xxii. High growth rates have been accompanied by both low and high rates of inflation. The economy also absorbed rapid increases in the labor force at constant and then rising real wages, without seriously disrupting the development process. Coffee prices fluctuatec widely, alternately creating unexpected balance of payments crises or rapic. accumuLation of foreign exchange reserves but in general the economy adjusted to these shocks without major disruptions. In the late 1970s, the country also turned into a net importer of petroleum from its earlier exporting posiltion and, as a consequence, has had to adjust to rapidly rising petroleum prices for increasing quantities of petroleum imports. Again, the adjustment was accomplished relatively smoothly. xxiii. While Colombia faced varying fortunes in the external sector, good economic management helped the economy to achieve stability reasonably well without losing the growth momentum during most of the last two decades. A historical review of the development plans of the last four administrations shows that despite some differences in their individual features, a remark- able consistency has characterized their objectives and policieS. Particu- larly successf.ul were policy directions of the late 1960s and early 1970s, which are worth replicating. Replicating HiLgh Growth xxiv. Favorable external circumstances which emerged during the early to mid-60s, clearly contributed to the 1967-74 performance as Colombia started to take advantage of them. World markets were expanding rapidly, Colombia established effective cooperation with several aid dorLors, and lhe terms of trade improvecl. In addition, non-traditional exports were encouraged by improvements iin the real effective exchange rate and special incentives. Imports were allowed to increase faster than GDP in Colombia improving  - viii - resource allocation. At the same time, the financial system moved away from directed credit, repressed interest rates, and unrealistic values of the Peso through greater incentives to save and improved credit allocation. xxv. Trade Policy. Non-coffee exports grew from US$117 million in 1967 to US$671 million in 1974--a 28% yearly average rate--to a considerable degree in response to improvements in the real effective exchange rate (REER). An adequate exchange rate led to better capacity utilization, higher export-oriented investments, more employment, particularly in manufacturing and an improvement in resources allocation. In contrast, at the end of 1981 the REER had deteriorated by about 21% compared to its level of 1974, hurting the performance of minor exports. The lack of adequate export incentives in 1981 meant, for example, that Colombia did not fulfill even half of its textile quota in the EEC, and Colombian exports of manufactures have per- formed relatively worse than those of most LDCs in recent years. xxvi. In Colombia, non-traditional, non-resource based exports are more labor intensive than import competing ones and, therefore, import substitu- tion in place of export promotion is likely to reduce the long-term demand for labor. Colombia's experience shows that both export promotion and import substitution can be used to stimulate the economy as long as output remains depressed below "normal" levels. However, once capacity becomes more fully utilized, export promotion and import substitution begin to conflict, as both compete for scarce resources and will bid up costs and prices. In such a situation, it is possible to maintain a high growth rate and increases in real wages by concentrating resources in industries with a comparative advantage in exporting or import substitution. xxvii. It is not easy to predict which industries would achieve compa- rative advantage in Colombia, if they were to benefit from a generalized policy allowing competition to select the most efficient development path instead of tariffs and import prohibitions favoring sectors chosen by planners. The industries which are most competitive internationally, be they exporting or import competing, will receive the greatest stimulation. In particular a more favorable real exchange rate should gradually be substi- tuted for specific tariffs and subsidies. Plausible sub-sectors, such as agro-industries, textiles, and capital goods, should be stimulated with technical assistance and investment loans for restructuring rather than wit:h specific protection. In the case of imports, a general tariff should be substituted for tariff rates varying across industries, and changes in minimum tariffs, rather than surcharges, should be used when increased protection is deemed necessary. xxviii. Resource allocation would be made more efficient if exchange rate policy were guided by a desired balance of payments target at higher ratios of exports and imports to GDP, thereby allowing a greater specialization and exploitation of comparative advantage in domestic activities. The failure of  - ix - the 1966 import liberalization illustrates the importance of choosing an exchange rate consistent with both the changes in protection and the balance of payments target. Colombia may suffer some short-term losses as resources are reallocated, while the inflationary effecls of a more rapid devaluation can be cushioned by the reduction in protection. In the longer-term, how- ever, the economy will be in a better positioIn to take advantage of a world economic recovery. xxix. Financial and Fiscal Issues. Greater reliance on market mechanisms in the financial sector would help to raise the private savings rate and/or improve investment allocation. In particular, attempts to fix domestic inte- rest rates through administrative decisions should be avoided. Recent mea- sures to improve the indexing on the UPAC instruments are likely to produce a resurgence of housing construction, which has suffered from slow growth during much of the past eight years but was one of the chief motors of the 1967-74 boom. A possible hindrance to further financial reform is the re- appearance of large Central Government deficit:s in 1981-82. Sufficient revenue generating and budgetary control measures are needed to reduce the size of the deficit. Although a fiscal reforrn was introduced in mid-1983 to address this issue, its full effects are like:Ly to be felt in 1984 rather than in 1983. Meanwhile the deficit can be fLnanced by tapping international capital markets and through domestic resource mobilization. xxx. In the short-term, the domestic sale of bonds bearing competitive interest rates could provide some fiscal resources. Currently, the domestic capital market probably could absorb some Government bonds bearing competi- tive interest rates, given the lack of demand for private investment. While the Government has recently taken some steps iJn this direction, longer-term exclusive dependence on this source is probab:Ly impossible, since without additional foreign capital inflows, local bond financing is likely to "crowd out" credit to the private sector. Thus, as the economy begins to pick up, it will be necessary to move toward a fiscal surplus. xxxi. The other means of financing the deiicit domestically is through continued reliance on Central Bank emission and the inflation tax. With an annual average inflation rate of 26%--roughly the average rate of the last three years--and the ratio of money base to GIP of .083 in June 1981, the Central Bank can issue new money to purchase new debt from the Government in amounts equal to about 2.2% of GDP (see para. 5.27). Thus, inflationary financing of about 2% of GDP is consistent wilh roughly 26% annual rate of inflation, a level clearly in excess of what Ls desirable. Moreover, a fiscal deficit of this magnitude deprives private investors of a similar amount of credit--in this case equal to about 10% of private domestic saving and 20% of savings passing through financial :Lnstitutions. Until the deficit is reduced significantly, it will be impossib:Le to reduce further the reserve requirements which now divert these resources to the public sector. Thus, the need for inflationary finance will perpetuate a large spread between average lending and deposit rates and hinder Einancial reform. Moreover,  x while in principle the system of directed credit based on forced investments could be reduced during a period of fiscal deficits, the high real interest rates which the deficits produce will create political pressures for credit at below-market interest rates. xxxii. Policy Coordination and Direction. In attempting to replicate the 1967-74 experience, the Government must aim for stability and consistency in policies. Frequent policy changes will usually prevent any single policy from achieving its full effects, and it often is necessary to maintain it for some time. This same argument also suggests the importance of adopting policy actions which may have short-term costs but longer-term benefits. The policy framework also needs to be internally consistent. The failure of the brief-lived, 1966 liberalization is attributable to an inconsistency between the large reductions in protection, the insufficient compensating variations in the exchange rate, including the revaluation of the minor export rate, and the low level of international reserves. In the current situation, exchange rate and inflation targets should be broadly set in line with projected use of inflationary financing of the fiscal deficit. While it is possible in the short-term to repress inflation somewhat by slowing devaluations and controlling prices, in the longer-term such a program will collapse because of internal and balance of payments pressures, unless the fiscal deficit is also reduced to a level consistent with the target rates of devaluation and inflation. Moreover, it is also not desirable to pressure domestic interest rates downwards through administrative measures since such actions would produce increasing pressures on foreign exchange reserves. This makes it all the more important to coordinate domestic interest rate and exchange rate policies with international interest rate developments to avoid capital outflows. xxxiii. Macroeconomic policy targets to reactivate growth with stability clearly should also be based on a full appreciation of current domestic and world circumstances that are different from those that existed in previous high growth periods. Exchange rate depreciation should take into account the need for offsetting policies to discourage coffee production, and also for more vigorous export promotion for minor exports to capture foreign markets under sluggish world economic growth. A lowering of existing import rest- rictions is particularly important in combating inflation, given higher inflation rates and fiscal deficits today than before. A lowering of the fiscal deficit, on the other hand, must recognize the need for care in averting any depressing effect on aggregate demand, given existing reces- sionary conditions: efforts should focus on reducing the purely inflationary component in the Special Exchange Account (SEA, see paras. 6.21 through 6.23), in providing greater flexibility and discretion to the Government in its expenditures, in improving efficiency of resource allocation and use, and reducing "crowding-out" of the private sector; a variety of essential revenue and tax measures can also be consistent with efficiency and growth. Finally, there is a greater need now than before to gear up the productive sectors with technological improvements and careful investment--to diversify agricul-  - xi - ture, and improve non-coffee export options and import: substitution possibi- lities in agrLculture, industry and energy. These issues are further deve- loped in Chapters VII, VIII and IX. Financial Pol:icies for Development xxxiv. The financial system's response to changing government policies over the last 15 years clearly suggest the beneficial effects of easing rest- rictions and eliminating distortions. Further efforts along these lines --maintaining a free Certificate of Deposit (CD) rate,, lifting interest rate ceilings on savings accounts and UPAC (see para. 4.10) deposits and reducing the burden of reserve requirements and forced investments, decreasing taxes on saving and correcting treatment of depreciation, reducing restrictions on medium and long-term foreign borrowing, and renewed ef-forts in slowing in- flation--are recommended. In addition, other key areas such as the regula- tion of financial intermediaries, concentration of financial groups (see para. 5.12) and universal banking will require the increased atitention of the Government, which has begun to address some of these iissues. It: has elimi- nated the marginal reserve requirements on commercial banks and unified the reserve requirements on deposits in public and private banks. M4oreover, it has appointed a commission to make recommendations for improving the ope- rations and stability of the financial system. xxxv. Mosi: institutional changes in the financial markets have been res- ponses to changing government policies rather than financial inlovations. For example, the growth of the Central Mortgage Bank in the earLy 1970s re- flected the liLfting of ceilings on its certificates, while its decline resulted from the growth of the more attractive UPAC system (see para. 5.05). In turn, UPAC benefitted from a monopoly on indexed instruments, which were highly attractive in the mid-1970s and became the only available mechanism for privately financing housing constructiorn in times of high and variable nominal interest rates. On the negative side, credit to agriculture slowed because the Caja Agraria (the agricultural credit bank) was inef- fective in corapeting for deposits in the new environment. After financial sector and government resistance, the competitiveness of other interest rates was restored as ceiling deposit rates were raised to near positive real rates. This permitted the financial sector to increase its mobilization of resources. xxxvi. In response to the financial restrictions (on credit expansion and reserve requirements and also prior deposits cn import:s, forced investments and directed credit) of the 1977 stabilization package (see paras. 5.08- 5.12), the unregulated market gradually grew in relative importance. In addition, limitations on the legal importation of capital diverted capital inflows to this market through a black market where dollars actually traded at a discount. The policies of the 1980s reversed this process by freeing the CD rate and easing restrictions on foreign loans t:hereby increasing loan- able funds in the regulated markets. The commercial finance companies  - xii - became the most rapidly growing institutions because of the liberalization of CDs and because the changes in restrictions made it profitable to legalize the unregulated market in which they had previously operated. xxxvii. Although the financial innovations were beneficial in general, in some cases institutions have profited from favorable regulatory treatment while in other cases distortions were created. In order to reduce advantages arising solely from differences in regulation, as well as to improve the security of the financial system, the Government might consider making capital requirements uniform by increasing the average capital requirement of the highly leveraged institutions and strengthening the Superintendency of Banks by giving it authority over all financial intermediaries. xxxviii. At first glance, the system may seem to be subjected to a high degree of risk because the individual institutions seem to be fairly specialized owing to their origins in taking advantage of niches in the market which appeared because of changing government policies. However, the disadvantages of specialization in a particular market (in terms of risk) are mitigated because most institutions lend to many sectors and at varying terms. The principal exceptions are the UPAC system and the mortgage and agriculture banks. Moreover, many institutions are associated with an economic group which controls many types of financial intermediaries with interlinked assets. Under these circumstances, a move to universal banking probably would simply convert the group's institutions into divisions of a multibank. Although this might increase efficiency somewhat, the decrease in loan interest rates is likely to be small in percentage terms. The principal determinants of interest rates remain the expected rates of inflation and devaluation and the reserve and forced investment requirements. xxxix. In order to lower interest rates on credit in the free market, the Government could consider the joint policy of gradually lowering reserve and forced investment requirements on non-CD deposits, combined with a gradual increase in the real interest ceilings on these forced investments (paras. 5.21-5.23). The effect of this policy would be to increase the availability of non-CD based funds, gradually driving down the average interest rate. At the same time, it is also necessary to gradually increase the real interest ceilings on deposits in financial intermediaries in order to avoid having the increased earnings of these institutions end up entirely as bank profits. The real ceilings could be raised either by raising the nominal ceilings, or decreasing inflation, making the existing nominal ceilings not binding. While some forced investment requirements should be maintained in order to retain control over the expansion of deposits, they should be lowered to the CD level on savings accounts while dropping marginal requirements of the total portfolio. These measures would lower interest rates to borrowers iln the free market and raise interest rates to depositors. If these recommend- ations had been adopted in 1981, the Development Funds' subsidy could have been reduced by more than half (see Table V.3).  - xiii - Public Expenditure and Resource Mobilization xl. Over the last three years, fiscal proolems have resurfaced after a period of relatively improved performance of the public finances. The growth in tax revenue that followed the implementation of the 1974 fiscal reform brought with it a substantial growth in Central Government spending relative to GDP. As the impact of the reform has been reduced through a combination of increased avoidance and evasion as well as subsequent modifications in the tax code, the overall deficit has risen from 0.2% of GDP in 1979 to 2% in 1981 and over 3% in 1982 while government spending has averaged about 11% of GDP over the same four-year period. This inability to reduce spending in response to declining tax yields has been caused to a considerable degree by the mechanisms governing the allocation of ptLblic sector resources and the loss of budgetary flexibility brought on by those mechanisms. Thus, to a large extent, the current problems are not only those of strengthening the GoverrLment's revenue generating potential, but also of budgetary control. During 1983 the Government introduced a fiscaLl reforn designed to address the above issues. The reform focused on income and complementary taxes, a series of tax measures to strengthen the finances of the departments and municipal- ities and changes in the nature of SEA. Public Expenditures xli. The Central Government is responsible for a share of all public sector financing that has been increasing steadily for many years. At the same time, it exercises a decreasing direct authority over consolidated pub- lic sector expenditures. This has resulted in part from the proliferation of national decentralized agencies and public enterprises over the last fifteen years, and partly from the growth in transfers to the departments and munici- palities. Transfers to these other levels of goverrnment and to the agencies and enterprises of the National Government have risen extremely rapidly since 1974, and by far the greater share of the growth has been in current rather than capital transfers. The rise in transfers has been associated with a fall in the revenue efforts of the departments and municipalities. In the last few years, moreover, the Central Government's cash deficit has risen abruptly to 2% and over 3% of GDP in 1981 ancd 1982 respectively. The overall pattern that emerges from these data has been is a steadily diminishing control over the level and composition of public exptenditure. Any budgetary reforms that are undertaken should be designed to restore that control. xlii. If the level of transfers from the NationaL Government is to be reduced, it is necessary that public enterprises and decentralized agencies rely on self-financing to a greater extent and greater tax efforts by the departments and municipalities. Specific examLples oE opportunities for greater self-financing by public enterprises and decentralized agencies include, but are by no means limited to, the following: continued real in- creases in petroleum prices and electricity tariffs, rationalization and overall increase in the structure of seaport user fees, together with in-  - xiv - creased expenditure controls on the finances of the ports, and substantial increases in tuition at the public universities. While there are many dif- ferent taxes over which the departments and municipalities might be given authority, including the recently increased tax rates on local excise taxes, a new tax on vehicles, higher fees on the operations of commercial, indus- trial and financial firms, and the updating of cadastral values, a very simple and easily administered inducement to increased tax efforts would be to transform the existing transfer programs into matching grants. This would partly reduce the tendency to view such funds as relatively costless, thereby increasing the care with which they are spent. xliii. Reduction in tax earmarking is desirable but this policy must be implemented carefully. Any reduction in the earmarked portion of the gaso- line tax that at present, for example, automatically increases the amount of remaining funds that constitute the base on which the situado fiscal (trans- fers to the departmental level of government, see para. 6.14) is calculated, thereby substituting the first reduction in earmarking by an increase in the amount of transfers. While such a reallocation may be desirable, it should nevertheless be the result of a conscious decision rather than a byproduct of the complicated system of non-discretionary budget allocations. Additional- ly, the distribution formula of the situado fiscal should be changed to increase the correspondence between a department's level of need and the amount of funds it receives. xliv. Serious consideration should be given to reform of the National Government's budget process. Two valuable changes would be a shift to a cash budgeting system and a change in the emphasis of the Comptroller's office away from excessive control over the disbursement of funds toward systematic performance auditing of ongoing government programs. This is a necessary step toward the long-term retention of whatever control over public expend- itures may be attainable over the next few years. Indirect Taxes xlv. An essential element in understanding Colombia's fiscal problems is the functioning of the SEA (para. 6.23). The replacement of the purely inflationary component of SEA by other taxes would not only provide the Government with more flexibility to combat inflation but could also provide it with the needed revenues to finance its programs as elaborated in Chapter VI, Section C. The 1983 fiscal reform introduced changes on the nature of the SEA. The greatest significance of these changes is that a substantial amount of funds will no longer be part of the general revenues but instead will be earmarked for a newly created Public Investment Fund reducing the amount of the situado fiscal and effectively freeing a significant amount of Central Government revenues from being assigned into current expenditures. Similarly, the revitalization of municipal and departmental revenues through recently approved tax measures should reduce the level of Central Governmient transfers to these government units and restore to them a greater degree of fiscal autonomy and direct accountability for expenditure decisions.  -xv- However, these reforms by themselves are not likely to reduce the overall Central Government deficit to historical levels. If that problem is to be dealt with successfully, it will be necessary for the Government to give con- sideration to new revenue sources in addition to a more vigorous effort in reducing tax evasion. The greatest potential. for rapidly increasing indirect tax revenues lies in broadening the base of t:he saler, tax (see paras. 6.27- 6.28). This has been authorized by Congress as part of the fiscal reform and requires Government implementation. In addit:ion, raising the tax on gasoline and other fuels and increasing the tax on coffee exports could increase tax collections. Income and Complementary Taxes xlvi. Additional revenues could be collected in income and complementary taxes with well focussed efforts in reducing tax avoidance and evasion and improving tax administration. In this connection, the fiscal reform package of 1983 contains many desirable alterations in the structure of Colombian public finances. The revision of the income tax, highlighted by the cut in tax rates (especially at the low end of the of the structure), the extension of presumptive income tax to family companies (limitadas) and the amnesty on declaration of assets (patrimonio), are important steps toward remedying the adverse effects exerted by "bracket creep" and rampant evasion on the distribution of the tax burden. They are also necessary pre-conditions to a more vigorous effort in tax administration and collection. Agricultural Strategy xlvii. With an annual growth rate of 4% during the past two decades, agriculture has contributed about one-fifth of overall growth. Although the sectoral share in GDP has been falling during this period, it is still about 26%. Well over one-half of the growth in agricultural output during the 1970s was accounted for by additional agricuLtural exports. The centerpiece of this phenomenon has clearly been coffee, although the combined value of other export items such as cotton, sugar, banana and flowers has also risen significantly, doubling in real terms during the past decade. Most of the additional domestic production of non-coffee products, however, has contributed to higher domestic consumption, which combined with more imports, has raised consumption by nearly 5% annually during the 1970s, implying increased per capita consumption of roughly 2.7%. xlviii. Sectoral performance, however, has deteriorated markedly in more recent years, with an annual growth rate of 2%-2.5% estimated for 1979-82. A part of this recent slow-down can be attribut:ed to adverse weather conditions and to the stagnation in the general level of economic activity . While the food price increase has not softened, the terms of trade for agriculture as a whole are no longer improving as in the first: three quarters of the 1970s, while production costs have been increasing somewhat faster than the output prices. In addition, however, the external ,iemand facing Colombia for its coffee and other exports remains depressed wLth the appreciation in the real  - xvi - effective exchange rate being a significant factor in the depressed state of non-coffee exports. On the supply side, the utilization of additional acreage for cultivation is projected to involve greater investment than in the past, implying the increasing need to pursue the option of achieving yield improvements. xlvix. During the 1970s, additional coffee output--much of which stemmed from yield increases--explained about 32% of the rise in the value of crop production; the sectoral growth over the past three years would have been halved if coffee were excluded. Of the increase in crop output other than coffee during the past decade, about 44% was derived from yield improvements and 56% from acreage expansion. With the country's coffee stocks being nearly equal to one year's exports, sectoral strategy will now have to be based increasingly on coffee diversification in order to restore non-coffee export and output growth. In the case of non-coffee commodities, although past yield increases have by no means been small, the potential exists for achieving substantial improvements. 1. Some scope exists for raising agricultural incentives through an improvement in the real effective exchange rate and through reforms in output pricing even though the prices of most non-perishable food items subject to price interventions are already at or above world price levels. The potential exists to modify the operations of IDEMA, the government purchasing agency, in restoring more reasonable seasonal price spreads. A more effective avenue for affecting incentives is in reducing input costs, most of which are above international levels on account of tariffs and other taxes and high transport and distributional costs. ii. The coffee boom of the mid-1970s slowed down and reversed the previous trend in raising the relative price of exportables other than coffee, and the negative effect on the so-called minor exports (non-coffee exports) has been significant. In response, the current Government has established higher export subsidies for agriculture, in conjunction with some tightening of import restrictions. A preferable alternative, however, would be to accelerate the rate of the depreciation of the Peso and at the same time lower import tariffs and restrictions on agricultural inputs. lii. Rural wages have, and may be expected to continue to, benefit from a production-oriented approach. At the same time, integrated rural development programs, which have shown to be capable of advancing rural welfare, should also be vigorously pursued. Furthermore, in order to perrait the most gains to be secured from policy reforms, it will be necessary to capitalize on the existing vitality of such institutions like FEDERACAFE and the private sector in general, and to strengthen the functioning of other institutions that face serious difficulties. iiii. In addition to productivity gains, additional land could and should be brought under irrigation and/or drainage for more intensive cultivationi. Priority should be given to lower-cost alternatives, such as rehabilitatioin  - xvii - of existing irrigation districts and drainage and flood control. rather than to new schemes. Watershed management and forestry development should also become integral parts of a long-term strategy for growth and for conserving the natural resources. The generation and delivery of technological innovations should receive priority in the array of long-term measures. Research institutions are in need of rehabilitation aLnd strengt:hening, and considerable scope for stepping up research and extension has been identified. There is substantial unexploited potential for int:ensifying the use of fertilizer and new seed varieties in non-coffee food crops. Marketing constraints also need to be relaxed if higher production is to be sustained: present problems include inadequate information networks (a steadier and more accurate supply of nationwide price data for producers could be! helpful), institutional constraints and transport bottlenecks. Sufficient credit availability for production and marketing is also essential, and existing credit policies should be carefully examined, as discussed below. Issues in Manufacturing Industry liv. While industrial development has been quite successful over the long-term, performance has worsened significantly in recent years. Over the past two decades manufacturing industry grew at a rate of about: 6.4% annual:Ly exceeding the average for the economy through most of this period. Its share in GDP increased slightly from about 16% in 1960 to over 18% today. Growth rates have varied from 8.6% in 1967-74 to 5.2% in 1974-78, followed by a sharp downturn to about 1% in 1978-82. lv. Manufacturing exports rose from US$25 million in 1967 to US$480 million in 1974, climbing rapidly from 3% of industrial production in 1970-71 to about 10% in the mid-1970s. While less important than the sitrong expansion of internal demand, the rapid growth of manufactured exports was the driving fEorce in the high industrial growth between 1967 ansd 1974. Expor expansion aft:er 1974 has been slow, however, and the recent setbacks in domestic production have been more than proportionately constituted by a slowdown in exports. Between 1979 and 1981, the valute of manufactured exports hardLy changed in nominal terms, implying a 1.5% decline in real terms. Thus,, in 1981, the share of exports in production was only 7% and the share of manxfactured exports in total exports only 16.4%. lvi. Industrial output and employment depend on many more variables than foreign trade strategies particularly since exports and imports combined are less than 25%, of GDP and exports of manufactures, while reaching over 30% of total exports, averaged only 10% of total industrial production during their best years. Nevertheless, the evidence presented in Chapter V.[II shows that manufacturing development and employment growth went through d:Lfferent development stages, which correspond closely with trade strategies. The analysis in Chapter VIII suggests that an appropriate medium- and long-term policy for sustaining industrial growth involves a reduction and equalization of protection, combined with a faster rate of devaluation, which would reduce  - xviii - distortions and provide generalized incentives for exports of labor intensive manufacturers and for competitive import substitutes. In contrast, restraining imports will raise their local prices and is likely to raise costs and prices in general, with negative effects on growth (see para. 4.29). Increasing credit more rapidly than import prices will produce balance of payments pressures and a rise in inflation. lvii. Appropriate trade policies should be combined with increased efforts to promote technological innovation, both in the production and marketing processes. Joint ventures with foreign partners should allow domestic industries the access to technology and marketing know-how they need to penetrate foreign markets. Subcontracting of parts to domestic firms is a promising approach to combine the cost advantage of local producers with foreign technical and marketing know-how. Here, a revision of the Andean Group foreign investment law is in order to facilitate creation of such export-oriented joint ventures. Improvement of export related infrastructure remains also a priority item. Ports and customs are still causes of frequent delays and high costs. Recent Developments and Policy lviii. Real GDP growth decelerated to about 3.7% during 1978-81, and to an estimated 1.4% in 1982. The production of coffee, cotton and oil seeds dropped sharply in 1982, as a result of low international prices, reduced fertilizer use and adverse weather. Both acreage and yields declined in many agricultural products and production was estimated to have fallen by about 1% at year end. Industrial activity remained stagnant in 1982 mainly on account of depressed aggregate demand, and unutilized capacity continued to increase, particularly in manufacturing. lix. After experiencing a surplus for six years, a deficit emerged in the resource balance in 1981 of about US$1.5 billion, and a slightly higher deficit, about US$1.8 billion, resulted in 1982. The latter year's deficit resulted mainly from a drop in real exports by about 4%; major reasons were the slowdown in world demand, especially for agriculture commodities, and the reduction in Colombia's coffee export quota in the ICA. Net foreign exchange reserves declined by about US$800 million, in spite of which, a balance equivalent of about 8.5 months of imports of goods and non-factor service resulted at year end. lx. Inflation has begun to slow since November 1982, leading to a 25% average in 1982, and 21% at present (as against 28% in 1981). Explanations for the continuing inflationary pressure can be found, among others, in the Central Government's overall cash deficit which reached over 3% of GDP in 1982, and the use of the SEA already mentioned. The financial sector has also been in difficulty with high real interest rates showing few signs of falling rapidly in the short-term in the face of the financial restrictions to be discussed in Chapter V, despite a drop in US prime rates and Eurodollar  - xix -- rates. Unemployment in 1982 was slightly above the historical rate of 9% of the labor force; to some extent this reflected the continuing growth in participation rates rather than a fall in employment. Developments in 1983 indicate that the downturn of economic activity has bottomed out and that the economy is gradually beginning to recuperate with reail GDP growth of 2% expected for the year. Development Issues and Government Strategy lxi. The Government's stated medium-term strategy is to restore the long-term growth, while avoiding balance of payments difficulties, as a means to expanding employment for the growing labor force, improving the income distribution and welfare. The Government is also committed to reviving growth with price stability in the short-terni, by st:[mulating domestic demand, regaining Colombia's international competitiveness and curtailing inflationary expectations. In this connection, and also to protect interna- tional reserves, the Government introduced import and exchange restrictions in early 1983. These are envisaged to be lifted as the real effective exchange rate gradually moves to its equilibrium level and Colombia's non-coffee exports respond to these incentives. These short- and medium-term goals call for a careful consideration of policy options and timely policy coordination. lxii. Growth stimulus is expected to be achieved through an expansion of housing construction utilizing the UPAC system, toget:her with increased pro- duction and exports in agriculture and agroindustries, capital goods indus- try, and small- and medium-scale industries. Crawling peg adjustments to- gether with the continued expansion of the Government's investment program should increase aggregate demand. Such an approach could also, however, increase inflationary pressures and further crowd oul private sector invest- ment unless complementary policies are put in place in a timely manner. Such policy measures are: (a) lower import protection along with the faster crawling peg as the economy begins to reactivate; (b) reduce the overall Central Government cash deficit below 1% of GDP throu gh revenue and expendi- ture measures over the next two years; in particular, eliminate the in- flationary component of the SEA over a two-year period and replace it by increased real revenues. Given that monetary' policy has been increasingly constrained by the size of the cash deficit, it will be difficult to reduce the growth of the money supply and inflation as long as the deficit remains high as a percent of GDP. liowever, as the deficit is reduced, the growth of the money supply should decline gradually, and the Government should avoid a substantial increase in the lending program of the Central Bank; (c) ease restrictions in the financial system further, particularly, the burden of forced investments and reserve requirements, and interest rate ceilings on savings accounts and UPAC deposits; and (d) modify the current treatment of the cesantia to curtail inflationary expectations.  -xx- lxiii. Given the present favorable reserve position, the low debt service ratio and the low ratio of external debt to GDP at present, government strategy should continue to emphasize public sector investment in energy, transport and the social sectors, together with policies and investments for a rapid expansion of agricultural, industrial and mineral production. In support of such efforts, Colombia will need to borrow heavily from external sources during 1983-86 to finance a large part of the foreign costs of its investment program. However, to finance the local costs and some of the foreign exchange costs, the Government will have to adopt timely and adequate pricing policies, and ensure adequate public sector savings for the financial viability of its investment program. Moreover, if overall investment is to be increased and the productive capacity of the economy is to be improved without straining the absorptive and debt servicing capacity of the public sector, increased private sector involvement in the energy, mineral and manufacturing sectors will be necessary, particularly to develop petroleum, natural gas and coal resources and to modernize industry. Macroeconomic Outlook lxiv. The current level of accumulated reserves and the Bank's present forecasts of gradually rising commodity prices beginning in 1983 are factors offsetting rising import costs. With a public external debt service ratio of 16% in 1981 and favorable medium- and longer-term export prospects under policy improvements as discussed in this report Colombia should be able to mobilize considerable external financing. Under these assumptions, growth could rise from an average annual rate of 2.7% during 1980-82 to about 3.7%--or almost 1.7% per capita--during 1983-86. Achievement of the latter growth rate would, of course, require progress in dealing with the develop- ment issues described in this report. lxv. Balance of Payments. In the medium-term, balance of payments pros- pects will be largely influenced by policy decisions on the following inter- related matters. The demand for Colombian exports are expected to be af- fected by the Government's response to the problems of low growth in the world economy, an appreciated exchange rate, falling domestic productivity in manufacturing and high input costs in agriculture. Colombia's answer to depressed world coffee prices should include an expansion of its non-coffee agricultural, agro-industrial and manufacturing export capacity, and the exploitation of its advantage in developing its energy exports utilizing domestic energy production. Real domestic price increases for petroleum derivatives could dampen the growth of the large amount of petroleum imports by increasing conservation. Moreover, recent discoveries indicate the existence of sufficient oil reserves which could make Colombia self- sufficient in oil towards the end of the decade. A more rapid development of these resources could produce the same effect by the mid-1980s. Although the current high level of foreign exchange reserves provides a cushion for rising imports until the mid-1980s, serious balance of payments constraints could emerge by then if the development of the coal, petroleum, and hydropower resources is delayed.  - xxi - lxvi. Successful balance of payments management over 1983-86 will require, in atddition to considerations in para. lxii above: (a.) policies to increase direct foreign investment, particularly in export activities with the purpose cof introducing modern production technologies; and (b) fiscal and other incentives to attain higher levels of labor productivity and replace technologically obsolete capital equipment. lxvii. External Capital Requirements. Gross external capital requirements are projected to total US$10.6 billion over thle 1983-86 period for an average of about US$2.6 billion per year. Net foreign investment is ex:pected to increase rapidly during the next several years, primarily in the mining and petroleum sectors, averaging about US$362 million per year for a total of US$1.4 billion during 1983-86. Direct capitaL inflows should provide about US$0.1 billion. The total of these inflows slhould provide about 14% of the gross external financing required during the projection period. Of the remaining 86% (US$9.5 billion) about US$4.5 bLllion (over two-fifths), has been either committed or is expected to be secured from multilateral and bilateral sources, while the difference, US$5 billion, will need to be secured abroad from financial markets and suppliers' credit sources during 1983-86. lxviii. Debt Management and Creditworthiness. Colombia's public and publicly-guaranteed external debt, repayable in foreign currency, was estimated at about US$7.6 billion at the end of 1981, of which about US$5.1 billion was disbursed and outstanding. Public debt outstanding and disbursed is projected to rise to about US$11.3 billion by 1986 and about US$20.1 billion by 1990. lxix. The public debt service ratio is projected to increase to about 24% in 1986 from 16% in 1981, as a result of the rapid increase in public external debt foreseen for the future and of imodest export growth. Projec- tions past 1986 show the public debt service ratio peaking in 1989 at about 28.3% as the country pays off the heavy borrowqing incurred over the decade. The debt service ratio of total public debt would then start declining, also as exports of coal and nickel are envisaged to expand, and would be about 28% in 1990 as compared to an estimated 10% in 1980. lxx. Although the projected debt and debi: service ratio would be relatively high compared to past levels, they would decrease substantially in the early 1990s when increasing exports from new energy and mining projects, together with import substitution in the petroleum and fertilizer industries and an expansion in agricultural and manufacturing exports, should improve substantially the balance of payments. By then, a major part of the invest- ment infrastructure required should be complelted and new commitments for public sector investment projects should decrease substantially. The success of the high investment strategy planned by the Government would therefore depend heavily on a rapid expansion of non-coffee- exports, Ion the timely  xx i- development of domestic energy sources and on progress in executing the export-oriented mining projects contained in the public sector investment program. Given the continuation of sound economic and f-fnancial managemenLt, Colombia is expected to maintain its creditworthiness through and beyond the 1983-90 period.  - 1 - CHAPTER I FEATURES AND SOURCES (IF GROWTH 1.01 This chapter traces major features of economic growth in Colombia over the past two decades, with the purpose of identifying factors that con- tributed to past performance and of setting recent developments and scenarios of future outlook in better perspective. The focus oi- the preseant discussion is on changes in conventional sectoral aggregates, production relations and consumption patterns. Clearly a full and adequate explanation Df historic trends cannot be provided on such a basis alore. This chapter will therefore be complemented by analyses, presented subsequently in this report, of macro- economic and sectoral policies affecting growth, price stability and income distribution. 1.02 Reviewed below are broad indicators of long--term performance and the sectoral contributions to growth. Also surveyed are structural changes in production and consumption relations that have been associated with the growth process. An accounting framework is used to bring out the contribu- tions to growth of conventional production factors, of- technological change, and of policy reforms which might have affected performance. A. Long-term Trends Overall Performance 1.03 Over the 1960-81 period the economy expanded at an impressive pace of nearly 6% annually in real terms. With a population growth varying from 3.1% at the beginning of this period to about 2% at present, GNPl per capita is estimated to have grown by over 3%, reaching US$1,380 in 1981. 1.04 Total population is estimated to have risen from 17.9 million in 1964 to about 27 million in 1982.1/ In spite of rapidly falling infant mortality and increasing life expectancy, the population growth rate has dipped sharply on account of the spread of education, urbanization, greater female participation rates in the labor force and an active birth control campaign. The working age population, on the other hand, was growing at about 3.4% annually up to the late 1970s, contributing to a worsening of labor market indices during the 1960s and early 1970s. Since then, however, a remarkable reversal has taken place: from 1973 to 1978 employment is estimated to have expanded at a phenomenal 6% annually. Growing non-agricultural and urban activities and rising wages have given rise to rapid rural-urban migration. As a result a point has now been reached where population pressure in rural areas is no longer increasing much, if at all. Moreover, the combination of rising per capita income, increasec. employment and expanded public services has brought about a significant improvement in the welfare of the poorest income groups in absolute and relative terms. These and related developments and some of their causes are reviewed in Chapters II and III. 1/ The last population census took place in 1973. The source of the total population estimate is the National Planninlg Department (DNP).  1.05 During the 1960s the annual rate of inflation averaged less than 12%, in contrast to rates of about 21% and 25% since 1970 and 1975 respec- tively. The financing of increased Central Government cash deficlts and the monetization of rising foreign exchange earnings after 1974 kept inflation at high levels despite government efforts at sterilizing a part of these earnings. A large part of this inflation reflected rising food prices, resulting in part from supply problems. Food prices have been a major component of increasing consumer prices, given a weight of about two-fifths of food in the construction of the index. Historically, the Central Government's budget had normally been in surplus.2/ But as each succeeding administration expanded its programs and increased the coverage of much needed services and infrastructure without matching tax and pricing efforts, deficits of about 1% of GDP or more emerged during 1970-74 and also since 1980. Chapters IV, V and VI analyze stabilization, financial and public expenditure policies that have affected inter alia government finances, inflation and growth. 1.06 Prior to 1967, Colombia faced periodic balance of payments crises--produced in some instances by falling coffee prices and in others by the combination of large devaluations and import liberalization (para. 4.04)--to which the Government responded with varying degrees of import controls. An outward-looking strategy was adopted in early 1967 (para. 4.05), this combined with the expansion in the world economy and domestic demand, turned the situation around during the 1967-74 period. In the second half of the 1970s, the unexpected boom in coffee and illegal crops sustained growth and exports which, together with capital inflows, further increased the international reserve position and debt-servicing capacity of the country. A trade deficit re-emerged in 1981 for the first time in six years on account of declining coffee prices, a deteriorated real exchange rate and the international recession. Chapter IX is devoted to development and outlook in the external sector, creditworthiness, and the economy's growtn prospects. Growth Trends 1.07 A disaggregation of the long-term real growth rate of about 6% into distinct time periods brings out significant differences over time as shown below in Table I.1. Until 1967, the economy was growing on average at a rate distinctly less than 5%, largely based on production for domestic consump- tion. Low coffee prices and declining terms of trade exerted a dampening effect on economic expansion. Diversification and expansion of non-coffee exports, together with increasing domestic savings and investment resulting from the gradual liberalization of the financial system, propelled a signifi- cant take-off during 1967-74 when the growth rate averaged nearly 7%. A growth rate slightly above 5% was maintained during 1974-78 when coffee and illegal exports experienced a boom. The pace of growth has tapered off since then, first as some of the stabilization measures directly or indirectly restrained expansion below the economy's potential and then as the world economy went deeper into recession (see Chapter IV). 2/ In Colombia, the size of the Central Government's overall deficit is important because of its inflationary implications. The Central Government also generates the bulk of the Public Sector's current account surplus (see Annex Table 5.1).  - 3 - 1.08 Agriculture, manufacturing and trade3/ have been the major contributions to aggregate supply, responding strongly to favorable export incentives ancd expansions in domestic demand, as set out in Table I.2 and reviewed in detail in Chapters VII and VIII. Agricult:ural expansion picked up sharply in the late 1960s and was maintained through most of the 1970s as the country capitalized, through rapidly increasing exports, on good coffee prices. About: a third of the 1970-81 growth in crop production was accounted for by coffee. Manufacturing growth was most pronounced during 1967-74, when the sector responded strongly to favorable export incentives and the expansion in domestic demand. The sector's performanc.e has decelerated markedly since then as a consequence of a number of factors: the slowdown in domestic economic activity, lagging technological progress behind tariff walls in the domestic industry (see paras. 8.14 and 8.30), a declining real exchange rate for exports in the face of the coffee bonanza, and the downturn in the world economy. The trade sector has sustained a fairly steady and high growth rate until recently, responding either to high manufacturing growth or to agricultural expansion. Since 1978, as both agriculture and manufacturing suffered, the growth in trade has also clipped. 1.09 Taking into account both growth rates and sectoral shares, contri- butions to GDP growth of individual sectors have been calculated in Table I.1. Details for all sectors are in given in Annex Table 2.2. Agriculture, manufacturing and trade combined have accounted for about three--fifths of total GDP increment during 1960-81. Transport, communication and power have Table I.1: CONTRIBUTION OF SELECTED SECTORS TO GROWTH, 1.960-81 (in percentage) Real Growth Rate Sectoral Cont;ribution t:o Growth a/ Agr i- Manufac- Agri- Manufac- cult:ure turing Tradeb/ GDP culture turing Tracleb/ Totalc/ 1960-64 3.1 5.9 6.0 4.8 21.1 2,0.2 20.0 100.0 1964-67 2.9 5.1 4.7 4.6 19.3 1.8.8 16.9 100.0 1967-74 4.3 8.6 8.3 6.7 18.0 23.1 21.4 100.0 1974-78 4.5 5.2 7.4 5.4 21.9 1.8.0 2'5.5 100.0 1978-81 3.2 1.5 2.7 3.7 22.2 7.3 1 3.8 100.0 1960-81 4.2 6.4 6.8 5.7 20.5 20.0 21.2 100.0 a/ Sectoral growth rates weighted by the corresponding sectoral. shares in GDP. h/ Comprised of commerce, banking, finance and insurance. c/ The other sectors' contribution to growth is not shown, therefore the sum of agriculture, manufacturing and trade does not add to 100)%. Source: Annex. Table 2.2. 3/ The trade sector is comprised of commerce, banking, finance and insurance.  - 4 - maintained a high and steady expansion but their proportions in incremental GDP have been held down by the smaller relative sizes of these sectors. The growth rate for agriculture has been less than the average for the economy, a characteristic feature of the growth process of most countries as demand shifts relatively more towards secondary products and services; nevertheless, the high share of agriculture in GDP has helped to maintain its prominent contribution to economic growth. In exports, in particular, it has played a dominant role in generating foreign exchange. Prior to 1974 manufacturing grew well above average, making the most significant contribution to GDP, closely followed by trade. Since 1974-78 agriculture and trade--fueled by the coffee boom--have had the leading edge. 1.10 During 1978-81 the three traditionally major sectors grew at rates below average, although agriculture and trade did not suffer as much as manu- facturing. The contributions of the remaining sectors belonging to the ser- vices category helped to improve the overall situation. For instance, the contribution of public administration and defence to incremental GDP was elevated to about 11% from less than 5% in 1974-78. The construction sector which had fluctuated widely contributing about 5% to the 1967-74 growth and 2% in 1974-78, raised its share in GDP increment to over 6% in response tc increased amounts of the foreign exchange earnings invested.4/ B. Structural Change Supply Side: Sectoral Shares 1.11 The Colombian economy has undergone a significant structural trans- formation, not unlike the experience of other countries going through a com- parable growth process. Over the period 1960-81, manufacturing, power, transport and communication, trade and services grew faster than the average of 6% per year for the economy, while agriculture, mining and public adminis- tration grew slower, which changed the relative sectoral shares. From 1960- 64 to 1978-81, the agricultural share fell from 33% to 26% while manufactur- ing and trade rose from 16.5% to 18%, and from 16% to 19%, respectively as shown in Table I.2. The share of construction does not show a secular trend. 1.12 The trend of a falling share of primary activities and rising shares of secondary and tertiary activities is normally associated with shifting demands during economic growth. The Colombian experience in fact, reveals less structural changes than might be expected on average: for instance, the share of agriculture remains about twice the average for courn- tries of Colombia's income range, reflecting the continuing large proporticin of high-value crops and the relative uniformity in the performance of various sectors. 1.13 In addition to variations in the composition of demand, the rela- tive shares of sectors in GDP have also been affected by other factors, such as exchange rate and trade policies, government investment and inflation. To 4/ Construction sector in the national acounts refers only to value added and differs from "construction activities referred to later in the context of sources of growth as the latter uses a final goods concept.  -5- evaluate the Lmpact of inflation5/ an equation relating sectoral shares to a time trend and the inflation rate was estimated for the period 1963-80. The results indicate that the share of agriculture has been falling at an average rate of about 1.3% per annum, while manufacturintg shows about a 0.5% annual average increase. Trade, transport and communication have been increasing their shares more significantly. Having separated changes in sectoral shares that fall along a trend line, the model brings out the distributional impact of the inflation rate on the sectoral growth rates. The! model suggests a differential effect of inflation among sectors. Agriculture is estimated to have suffered most on account of a general inflation (production costs rose rapidly and sq[ueezed profit margins), while mantLfacturirig gained from it (i.e. in relative terms); the effects on the remaining sectors were smaller and less statistically significant. Table I.2: SHARES BY INDUSTRIAL ORIGIN IN GDP, 1960-81 (in percent at 1970 Peso factor cost) 1960-64 1964-67 1967-74 1974-78 1978-81 Agriculture 32.7 30.7 28.2 26.3 25.7 Industry 24.2 25.0 26.8 26.0 24.8 Mining 2.3 2.4 2.0 1.3 1.1 Manufacturing 16.5 17.0 18.0 18.7 18.1 Construction 4.3 4.3 5.2 4.3 3.8 Electricity, Gas and Water 1.1 1.3 1.6 1.7 1.8 Services 43.1 44.3 45.0 47.7 49.5 Transport/Communication 6.9 7.1 7.5 8.8 9.5 Trade 16.0 16.6 17.3 18.6 19.0 Public Administration 7.0 7.1 7.0 6.7 6.7 Other Services 13.2 13.5 13.2 13.6 14.3 GDP 100.0 100.0 100.0 100.0 100.0 Source: Banco de la Republica and mission estimates. Demand Side: Changes in Expenditure 1.14 The major components of aggregate demand in Colombia have moved, by and large, similar to their behavior in other middle-income countries over 5/ The following equation was estimated for each of the sectors: Fi =eYi + - iT + YjI), i = 1, ...9;with cross equation zonstrainlts