DISCUSSION PAPER Report No.: SUDD 34 HOUSING FINANCE IN COLOMBIA by Alvaro Pachon (Consultant) August 1983 Water Supply and Urban Development Department Operations Policy Staff The World Bank The views presented here are those of the author, and they should not be interpreted as reflecting those of the World Bank. Dr. Alvaro Pachon is a consultant based in Bogota, Colombia. This paper was prepared in 1983 as part of an on-going review of the experience in housing finance in Latin America. It was prepared jointly for the Urban Division of the Latin America and Caribbean Regional. Office and the Urban Development Department of the World Bank. ABSTRACT This paper presents a comprehensive overview of the housing finance system of Colombia in 1983. It provides a. brief 'description of the recent macroeconomic environment in Colombia, of the general structure of the financial system and of the place that hou.sing finance institutions occupy within that system. It charts recent trends in finairzial intermediation anxd provides a description of the role and significame of the major housing finanre institutions operating in the system. It cornluies with a review of some of the issues that are presently arising in the development of the Colombian housing finance system. l',fi'LE OF CONTENTS Fage :~ I. INTRODUCTIOIN i II. OVERVIEW OF T'HE COUINTRY 5 A. Macro-Economic Overview 5 B. The Financial System I 5 1. Structure of the Colombian Financial System 15 2. Recent Developments of the System 21 Z. Interest Rates 28' 4. Resource Mobilization ;S; 5. Current Conditions 34t C. The Housing Sector .5 1. The Stock o+ Housing :5 2. The In+ormal Sector 41 3. Non-Financial Constraints 44 III. HOUSING FINANCE 47 A. Trendis in Level of Intermediation 47 B. The Land Credit Institute (ICT) 5i C. The Central MortgaFe Bank (BCH) 67 D. Housing and Saving Corporations (CAVs) 75 1. Indexation 78 2. Terms and Conditions 79 3. Assets and Liabilities 83 4. Relationships with the Central Ban k 86 5. Credit Allocation as E. Other Institutions 89 IV. ISSUES IN HOUSING FINANCE IN COLOMB4IA 94 A . Introduction 94 B. The role of HoLusing Finance 95 C. A+fordability 9 DJ. Rental Market of HoLusing 101 E. Public Sector Aaencies i 02 F. The Public and Private Mortnage Interest Rates 104 G . SuLb si d i es 106 L liDS OF LfA;LE S Tabl e No. F'a3e Nc'. 1. Construction Share in GDF Colombia 1970-1981 ............b 2. GDP G~o-owth by Sector. ................................e 3. Gross Capital Formation Colombia 1970-1981 ........... 10 4. ConstrLiction Economic Indicators .................. 11 5. AnnLual Urban Floor Space Building .................. . 12 6. Demographic Indicators o$ Colombia .................... 14 7. The Colombian Financial Sector ...................... 16 8. Financial Sector: No. of Firms and No. of+ Offices... 18 9. Financial Liabilities Colombian Financial System.... 20 10). Level of Intermediation - Financial System 1970-82 s 2 11. Selected Reser%a Requirements........................ 25 12. Distribution of Financial Assets Colombia 1970-82 ... 27 13. Financial Assets: Colombia 1970-1982 ............. . 29 14. Selected Deposit Rates Colombia 1972 - 1981. .........30 15. Housing Stock: Colombia 1950 - 1986 0............... 36 16. Estimated Urban Income Distribution Colombia 1982... 40 17. Housing Output by Financing Agency Colombia 1981. 42 18. Individual Mortgage Credit Colombia 1950 - 1981 ......4? 19. Construction Financing By Sources 1965 - 198 ....... 20. Distribution of New Resources Used in Construction.. 51 21. I.C.T. Revenues 1970 - 1981 ...................... 5 o: Z2. I.C.T. Revenues by Source 1980 -1981 ................. 57 23. I.C.T.'s Outstanding Loans 1978 - 1981 ............ 59 24. I.C.T. Expenditures 1977 -19.1 ....................... 61 25. I.C.T. Selected Housing Statistics 1942 - 1981...... 62 26. I.C.T. Sale Conditions by Program 1981.............. 64 27. I.C.T. Financing Conditions by Program 1961 .......... 65 26.. I.C.T. Profit and Loss Statement 1976 - 1979 ........ 66 29. Distribution of Construction Credit 1960 - 1981 .....6? 6 30. B.C.H. Liabilities 1l970 - 1981 ....................... 70 31. B.C.H. Approbals and Disbursements, 1982 ...............74 372. Housing Units Financed by the CAVsq 1973 - 7 86 .b :;. UPAC Financing Colombia 1973 - 1981 ................... 77 34. CAVs Effective Returns 1976 -1981 ................... 84 35. CAVs Assets and Liabilities i973 -1981.. .............. 5E 36. CAVs Profit and Loss Statement 1974 - 17.............-7 Z 7. F.N.A.. Sources of Funds 1971 - 190.. ................. 9 38. F.N.A. Financial Terms .............................. 39. fA+fordability of ICT Proqrams bv Tyoe of Prooram.... s 40. Relative A+fordabilitv of the UPAC System ........... .1 41. Implicit Subridies in the UFAC System ................. 108 StUdy wi IL I provi dE crne model o+ hou5a na f 1 nance devel opinert t- contrast with others that jave been used in other develocinc couLntries. The fLtnct.ions that a housing Finance system perform will be illustrated by the study of the Colombian case. The way ir: which financial intermediaries mobilize their resources will also be analy-Zed in the Colombian context. The important role of non-bank instittutions in the Colombian housing finance system will be explored. The ways to protect a housing finance system against inflation will be illustrated by the Colombian experience wiith indexed instruments. The mix of housing finance provided by specialized institutions and by general purpose financial intermediaries will, also be studied. Final-ly, the Colombian experience coutld be used to explore some o+ the following i SSUCes: Should housing finance institutions be run by the public sector? Should market lending operations and. government low-income hoLtSing programs be run through two windows of the same institution or two separate institutions ? It is also hoped that the study will help to clarify some o f the issues adressed in Colombia. The National Planning Department (DNP) in a recent document (198i), has mentioned the following 1 SsLtes: 1) Reducing the housing deficit while improving the quality of the stock o-f housing units, 2)- reducii6g the instability of housing finance f unds by reestructuring the capital market to provide adequate funds for housing, 3) improvi.ng access to low-income hoLoseholds. 4) minimizing the extent of low density and The objective of this paper is to present a description of housing finance in Colombia and of the main issues' related, with the financing of formal housina units. DuLe to lack of quantitative information on the informal sector it has not been studied. It is hoped that the Colombian e>:perience with her varied institutions. and the relative sLuccess in coping with very rapid growth of Urban population, could help in uLnderstandi.M, ways to solve similar probl ems. The importance of housing construction in Colombia can be appreciated by the vaiue of its oLutput (Col 12,1C?5 Million in 1979), by its contribution to employment generation (183,365 man-years), by its contribution to gross capital formation (13.7%'. of the total) and by the share of savings going to the housing sector (46.9 '. of the total). Because of the multiplier effects of hoLusing in other sectors, housing construction has been considered in some development plans -as one of the. leading sectors of the economy. This study o+ housing finance in Colombia will contribute to illLustrate some of the main issutes identified by RenaLtd 1 Pi2) . It is hoped that the study will help to understand the di++erences between hoLusing finance and the financing of other sectors. the 1cJ >)-rif development, J cot-ntrolling the large i ncreases in 1 a.r, values,t 6) reducina the he.-,vy btrden on ptblic utilities dLue t- the rapid increase in Lurban population. A+ter this i ntrod ucti i the paper presents an overview of the Colombian economy and its firnance system. Looking first at macro economic conditions. then at the finance system, and finally at the housing sector. The paper continLies with a stLudy of hoUSiino finance. Lookcing at the trends in financial intermediatiorn in housing, at low income finance by the Land Credit Institute (ICT). at the r:,le of the Central Mortgage Bank , the Housing and Savings Corporations, and the rest of the institutions financing shelter. The paper concludes with a presentation of the main issuas in housing finance in Colombia. It studies the role of housinq finance, affordability by low income households. and ends with a study of implicit subsidies to housing. Given the limited scope of this paper, it does not pretend *to summarize all what is known about the development of Colombia, urban development, the +inancial sector or hoLusing. The World Bank. has made care+ul studies about all these topic!. in this context.1.f -------------- 1/ See for example, The Colombian Investment EBanking System and Related Financial Sector Issues. World Bank, 1i933; Urban Policy in Colombia Selected Issues and Some Directions for Change, World Bank, 1983; Colombian Economic Development and Folicy Report, World Bank, 1983; Strassman. Paul W. The Transformation of Urban Housing, 1982; and the work carried out unoer the City Study reserch project. 4 With its viork it has contributed to ae cl ea:rer Llnderstanding c, recerit Coiombiari development. This paper has benefited trom this wor k Colombian researchers have also contribbuted to advan'ce our knowledge aboLut recent economic ev,ents. Most of the qLanti tati ve information in the next two chapters come from careful work made by Colombians. Financial sector information comes mainly from the work done by Arango et al.v (1983) and by Carrizosa et al., (1982). Information on hoLusing finance comes from Carrizcsa ano associates., from Villate (19S1) and from BCH sp,nsored research undertak en by Sistemas de Informacion (1982). 11 UVERVlEW OF 1 IE CCOUNJIRY A. Ma.cro-Ecmnomic Overview The constrLlction sector has been considered in Colombia by the present Government as one of the main hopes for the reactivatior of the economy. As in other countries constri.ction has experie.nced more volatility in its growth than other sectors (table 1). the rate of growth of the construction sector during the peri od 1971-1982 has been 4./6%/ while the rate of growth of GDP was 5.2%. If one looks at four years periods, one can observe the rate of growth of both GDF and construction have declined ..GDP growth rate declined from 6.7'. in the period 1971-74, to 5.5%'. in the period 1975-78 to 5.3.% in the period 1979-82. Construct.ion's rate of growth declined from 5. 7% to 4.67. and to 3.57. in the three four-years periods mentioned above. Graph 1 presents the variation in money supply and prices; most c+ the variation of the money supply were caused by variations in foreign reserves, the largest variation was observed in i976, in a year of external bonanza. Prices have moved in the same direction than Money Suppl y. The shAre of constrLucion in total GDF in 1981 was .-.. 7 considerably less than its share in 1970 A--0.). i" slight recoverY TABLE 1 CONSTRUCTION SHARE IN GNP COLOMBIA 1970-1981 ( 1970 Millions pesos) Ano GOP Construction Share % 1970 130 361 6,566 5.04 1971 137,889 6,897 5.00 1972 148,630 7,032 4.73 1973 159,195 7,884 4.92 1974 16B,786 8, 190 - 4.85 1975 175,226 7 830 4.46 1976 183,296 6 723 3.67 197.7 192 187 7 101 3.69 1978 209,369 7 379 3.52 1979 220,091 7 277 3.30 1980 229,294 7,981 3.48 1981 235,028 8 752 3.72 Source: Revista Banco de la RepCublica 'vE$.FFQ~ (J P..trtE P. lpL" AP.1 D P FR I';F 19e71 t 89m t 1973 19i74 1 97r: t1976 197. 1975 1 9X t 983ti l9l 19 ' .1-icm in constrLuCtion nas been observed in the eighties (table 2). I- z-Th accordinc r-o a stLidy made by DNFP, the value added by the constrLuCtion sector was Cols 49qC)81 Millions or 4.17. of GDF ir cLtrrent pesos. In that year 3.51.495 man-years were labored ir construictions correspondi ng to 4.9 of the e-bnomy' s employment (table .3) . Construtction is one of the main components of gross capital formation. The share of constrLuction in GCF was 36.3% in 1981. hoLtsing represented almost half of construction or 17.2-% of the total GCF in Colombia (table 4). According to some estimates of DNP tor each peso spent in building construction 45 cents are spent in other sectors inputs and generate 117 7. of indirect. employment. Housino production is more than 75,' of total building in Colombia (table 5). Housing production is largely concentrated in the four largest cities: Bogota , Cali, Medellin, and BarranqLLilla. According to building permits information, 647. of the area was built in the four largest cities. In recent years. however. a relati'vely higher growth in intermediate cities has been observec ard consequLently the share of the XOLUt largest cities in constructicn has declined. Housing e>:penditures in Colombia. as in other coLuntries are e larae 'fraction of total household expenditures. Accordini to the TABLE 2 GDP GROWTH BY SECTOR Anios GDP Agricul- Manufac- Trade Construc- Mining Finance Transpor- total tural turing tion tatlon 1971 5.8 2.5 8.5 6.3 5.1 0.7 15.9 5.6 1972 7.8 5.9 9.2 6.7 1;9 -7.4 11.0 7.9 1973 7.1 4.0 8.9 8.6 12.1 5.5 8.1 -9.4 1974 6.0 7.2 5.6 5.0 3.9 -7.0 15.4 10.8 1975 3.8 6.0 0.7 1.3 -4.3 -5.5 13.9 7.2 1976 4.6 1.8 7.1 6.3 -14.2 -1.9 10.0 9.0 1977 4.9 2.3 4.1 5.5 5.7 -3.9 9.7 8.8 1978 8.9 10.3 8.5 10.5 3.8 4.4 9.6. 10.9 1979 5.1 4.4 4.7 5.8 -1.4 -1.2 1.2 9.0 1980 4.2 2.9 2.0 2.8 5.2 7.8 8.1 7.7 1981 2.5 2.4 -1.1 -0.1 6.2 6.6 6.1 6.2 1982 J1.4 -0.8 -1.1 -1.4 4.0 7.5 -- 6.6 Source: Banco de La Republica 10 TABL E 3 GROSS CAPITAL FORMATION COLOMBIA 1970-198j ( 1970 Millions pesos) YearC Gross Hous irg Construc- capital tion formation 1970 26,441 4,475 16.9 13, 824 52.3 *1971 23,067 4,449 19.3 14,577 63.2 1972 27,786 3,826 13,8 14,724 53.0 1973 29,156 5,157 17.7 16,817 58.0 1974 31s,700 5,151 16.2 17,890 56.4 1975 31,844 3,729 11.7 16,363 51.4 1976 32,802 4,043 12.3 14, 400 43.9 1977 34,487 5,001 14.5 15 426 44.7 1978 38,737 5,771 14.9 16, 783 43.3 1979 39,719 4,781 12.0 16,020 40.3 1980 45,860 4,518 10.1 17,434 38,0 1981 48,005 8,250 17.2 17,436 36.3 Source: Revista del Barnco de la Rep6blica TABLE 4 CONSTRUCTION ECONOMIC INDICATORS 1979 Share Values in the Economy I GDP Construction 49,080.6 100.0 4.1 a. Housing 15,404.1 31.4 1.3 b. Other construction 33,676.5 as*6 2.8 II Factor Shares of construction a. Salaries 35 ,901.9 73.1 9.0 b. Profits 13,178.7 26.9 - III Employment. (Man - years) 351,495 100.0 4.9 a. Housing 18 e,360 53.1 2.6 b. Other construction 165,135 4649 2.3 IV Gross Capital Formation (millon $) 103,067 100.0 41.7 a. Housing 33,914 32.9 13.7 b. Other Buildirg 9, 181 8.9 3.8 c. Other construction 59,972 58.2 24.2 V Financial saving going to 72,735 46.2 bultding Source: DNP (1981) 12 CUADRO 5 ANNUAL URBAN F=LOOR SPACE: BUILDING (000 SQ m) 1970 - 1980 Year Housing O Other Total bui tdirg 1970 4,0440.5 82.1 968.2 17.9 5,408.7 1971 4'309 .5 79.0 1.,143.2 21.0 5,452.7 1972 3,796.g6 7644 1,172.8 23.6 4,959.4 1973 5,151.3 79.7 1,310.6 20.3 6,461.9 1974 5,080.2 72.1 11972.5 27.9 7,052.7 1975 3,627.4 75.0 1,212.2 25.0 4,839.6 1976 3,864.7 74.6 1,1308.4 25.3 5,174. 1 1977 4,972.9 77.3 1, 463.8 22.7 6,436.7 1978 5,795.6 80.0 1,447.9 20.0 7,243.5 1979 4,717.7 77.4 1,376.0 22.6 6,093.7 1980 4,468.8 78.5 1 223.9 21.5 8,692.7 Source: 0. N. P. 1981 Banco de la Rep6blica. Cuentas NacionaLes y Revista del Banco de la Republica (varias); Dane, Soletin Mensual de Esta- d(sticas; Albert Berry, "W La Industria de la ConstrucciOn en la Deca- da de los anos Sesenta"l, Economta Colombiana, No.92 (abril, 1972), pags . 16 - 30. National Gtatimticil 0+fice (DAcNE), blUe collar wqrrkers sporrd +i±th ot their income in housing while white-collar work.ers sper: 23% of their income in hoLising. The housing stock in 195i was approximately 1.9 million units, three decades later, in 1980 there were approximately 4 miillior Lunits in Colombia. The housing stock grew at an annual rate o- 2.67.. In these three decades Colombia passed through a process o).~ rapid ULrbanization, it passed +rom a .9% .urbanization ratio ir 1951 to an estimated of 69% urbanization ratio in 1983 (Table 6). Eventhough,. Colombia has experienced high rates of popLtlatior growth her urban growth rates have been even higher. DLuring the sixties and the middle seventies the Lurban growth rate was douLble the countryp's population growth rate. EBy the late seventies the relative importarice of migration' has decreased conseqLlentlv the urban growth rate was only 507. higher than the country growth rate. Migration's contribution to growth was approeximately fift iy - percent in the sixties while it is only one third in the eighties. The rate of growth of popLulation has declined very rapidly ir Colombia. From a peak o+ more than Y/. at the beginning o+ the sixties it has reached a level of approximately 27. per year (Table 6). The birth rate has declined from 4.37. in 1964 to an estimatec 2.87% in 1983. The death rate has also declined, it was i.57. inr 1964 and it is estimated at 0.67. in 1983. Family size has declinec in the last decade5 dLue to declines, in fertili ty rates and tc 14 TABLE 6 DEMOGRAPHIC INDICATORS OF COLOMBIA Y EA R 1964 1973 1978 1983 Tctal Population 17,918 .8 22,904.8 25,655.6 28,77e.5 Growth rate (%) 2.80 2.26 2.32 2.19 G B R (%o) 43.3 32.7 31.2 28.5 G D R (%o) 15.2 10.1 8.0 6.6 Urban Population 9,265.2 14,013.0 16,772.5 19, 727,5 Growth rate (%) 1/ 5.6 4.6 3.6 3.3 Urbanization. rate (%) 51.7 61.2 65.3 68.6 1 / Urbanization rate Urban population x 100 Total population Source: Base - Line run CCRP's SERES MODEL improveme!nts in economic conditions. Lonsequently, the nLumber c families have increased 'more rapidly than population. AlthoLugh there is no stLudy of the possible consequernces tha-. changes in demographic conditions can have on the demand for housing in Colombia, it is possCible to think of' the more likel'. ones. Firstly, it is **ery likely that the lower rates of grow-th o- urban -nd total population colld help in improving the housinc .conditions in Colombia. SecondlyS, smaller families will require less space and consequLently the same reSoLurces couldl be used tc provide more units. Finally, lower migration rates-will imply that the main demographic force driving the demand for housing will be natLural increase. Therefore the composition of new housi ng oLutpLt will change. B. The Financial System 1/ 1. Structure of the Colmmbi'an Financial System Table 7 presents a summary of the Colombian financial system, it consists of one operating branch that includes deposit institutidnsq non-deposit institutions and related institutions regL}lated by the government and a regLl atory system consisting of 1/ This section is based mainly on: Caballero (liSO); Carrizosa et al. 1 ¢92); Arango et al. (19:); World Bank (1983 b). The Worlc Bank: Fep.ort has been tsed as the base Tf-or nLtmerals 2 3 ;and x4. TABLE 7 THE COLOMBIAN FINANCIAL SECTOR Central Bank [Comercial Banks (Banco de la Rep6blica) Irvestment Banks SuMonetary Board Credit Saving and Housing Corporations Superintendency of Banks Savings Bank Savirns Cooperative - 1 l Finance Cornpanies Financial Insurance Companies Capitalization Compantes Fiar-a-icial Ivsmn ud Intermediaries Oherestment Funds Livestock Funds Financral Leasing Companies Financitati Factoring Companies I nstitution [ Stock exchange Related capital . General Wareh0use Companies Trust Comparies Stock Brokers Macket Insurancer Brokers Source: ArangQ, et.at. El Sector- Financiero 1 7 the Centr -E. Ban4.. the Mlonetarv BEoard and the Superintendenr,cy o Oankhs. Irs 9'82 there were -7 Commercial 13anks. s0 Investment Bank= (CFs), Io Saving and Housing Corporations (CAVs) , 4s) Finance Companies (:CFCs). 11 Investment Companies, 70 Insurance Companies and 21 Leasing Companies (table 8). The commercial banks have the largest barsach network (almost 2700 offices in the country), they are followed by the CAVs with more than .4C)Q offices. The CAVSs and the Finance Companies began their operations in the seventies and have experienced very rapid growth in the last decade. In addition a stock market exists with trading in shares of about 200 major firms, government bonds and CATs. Mutual funds exist and deal in the listed stocks and other equity instruments. Firms--mainly those in manufacturing and mining--may obtain direct loans +fom abroad for .specific ends and: some intermediation oF foreign funds is also done by the financial sector. An Lunregulated credit market also exists, althouLgh its importance has declined recently owing to the freeing of the CD rates, the impositions o-f regulatiQns of transfers of funds from the regulated marketq, and the incorporation of many of its firms into the regulated market as commercial +inance companies. The large number o7' branch offices of the various instittutions provides access to diverse financial intermediary sersvices for most oa the populationq, 50. of which live in cities of over 20,000 persons.. Four tvpe of institutions are allowed to take deposits from the TABLE 8 FINANCIAL SECTOR# OF FIRMS AND # OF OFFICES -Ia-ncial 1970 1971 1972 1974 1975 1076 1978 1979 1980 1981 1982 -iterme- [larias 1)(2) (1) (2.) (1) (2) (1) (2) (1) (2) (1) (2) t9) (2) (1) (2) (1) (2) (1) (2) (1) (2) Commercial- Baoks - 25 1880 25 1853 25 1970 26 2290 25 2456 28 2455 27 2449 27 2613 27 2606 24 2695 27 2696 rvestnent arnks 14 14 :17 17 17 17 0 20 23 24 29 30 128 30 128 30 128 30 100 AVs - - 7 10 48 10 10 2a 10 260 10 388 10 394 10 388 10 400 Iinance -C ompaniles - 10 12 18 34 39 98 40 75 40 *100 40 76 -ivestme-it 3ompariles 10 10 10 10 10 10 10 10 10 10 11 11 11 11 nsurance Companiles 67 65 65 C 64 70 Leasirg 1 1 1 1 3 3 3 3 17 17 42 21 -. - - - -- -.. Source; Arangoet.al.EI Sector Fi-ianclero Colomblaro (1) * of firns (2) * of offices 19 p y 1: c L cmllt'ri a 3. banks. CFs. CAVs and CFCs. CommerciE:t.L b art' commands more than two third of the resources of these +our t,pcE- o+ instituLtions '(Table 9). Commercial:Banks. The main source of fuLnds of commercial banks is the demand d6posit, followed by certificates of deposit( CDs) and savings deposi-t5 (table 9)-. Commercial Banks primarily grant short term loans to indcustry and other btusinesses. They also provide long term funds at concessionary terms from resources provided by. the various Financial Funds of the Central Bank. 1tfort.gff_BLnk. . In Colombia there is only one bank that provides medium and long term funds for housing . This is the Central Mortgage Bank (BCH), a pLublicly owned bank whose major shareholder is the Central Bank. Mortgage bonds anrd some forced i lnvestments prov,'de the main sotirce of housing funds. Since BCH is one of the main actors in housing finance a more detailed presentation will be made in the next. chapter. Investment Banks. The investment .banks (CFs) main objective are"to promote the creation,trieorganization,and transformation of firms (in the manufactLtring, mining, and aaricultural sectors), tco participate in their capital or facilitate the participation of others, and to provide them credit". Initially CFs resoUrces came from long term bonds but when inflation accelerated they were dallowed to issLe short term liabilities. TABLE 9 FINANCIAL LIABILITIES COLOMBIAN FINANCIAL SYSTEM ( Col. $ Millions) 1977 1978 1979 1980 1981 1982 1. Commercial Banks 3436 160,352 242,670 346,159 412,036 Demand deposits 62,P696 81,404 100,739 130,950 159 ,142 195, 615 COD 12,790 16,230 13,576 53,327 114,291 132, 143 Savings deposit 24,178 36,653 46,037 58,385 72,725 84,278 2. 1 -ivestme-it aanks : 771 15$645 192453 862 COD 8,771 15,645 13,315 19,453 26,380 28,862 0 3. Saving Housing Corporation 24,400 34,956 53 204 83 242 122,209 171 ,847 Savi-gs deposits 15,832 20,924 37,226 58 ,614 84,044 122,759 COD 6,742 8,727 12 ,96 21,929 35,004 46.,145 Ordinary deposits 1,834 5,305 3 ,009 2,699 3,161 2,943 4. Finance Companies 7t470 9 743 18 774 p553 4472 Fu)ds from the 5,797 7,470 9,743 18,774 36,653 44,J722 public Total 138,640 192 438 236,614 364 ,139 531, 281 657 ,467 Source: Banco de La Republica - Departamento de I nvestigaciones Economicas t 1 Savings and- Ho4ai n .cgr- tions. One of the principal developments ii the seventies was the creation of the CAVs. Since its creation irs 1972 they have experienced a very raoid rate of growth. The CAV'- main innovation was the indexation of both assets and liabilities. CLtrrently they are the main actors in housing f-inance. T'he next chapter dealing with housing finance will present a.detailed descr-iption of this type of intermediary. Finance Commanies. The Finance Companies (Companias de Financiamiento Comercial o- CFCs ) represent the second innovation of the seventies. CFCs have experienced a very rapid growth since 1973. Originallv( before 197t CFCs operated in the unregeLlated markets in th-t date they became regulated. In the middle seventies there were very low barriers to entry and conSequtently the number of irntermediaries increased very rapidly. In 1979 the Government increased their capital requirements and as a consequence the number of CFCs remained constant for a long tirme. 2. Recent De . clo opments of the System Several studies have been made of the recent evolution of the financial sect=.z: since 1970. The stLudies shtow that during the period the finar-cial sector has crown mare rapidly than GDFL. The ratio of finnanc: al savingis to GDP has increased from 6. 1. in 197/0 to 29.-.7. in 19.:1- .Tab Ie 1I0) . 22 TABLE t0 LEVEL OF INTERMEDIATION - COLOMBIAN FINANCIAL SYSTEM 1970 1982 Year Financial savings/ M1+ Financial GOP saving /G DP 1970 .061 .212 1971 .062 .209 1972 .064 .204 1973 .070 .205 1974 .077 .199 1975 .077 .210 1976 .099 .215 1977* .110 .227 1978 0120 .236 1979 .112 .225 1980 .116 .227 1981 .176 .306 1982* .170 .297 * Estimated Note: Fin.savings Savings deposits, UPAC, COD, mortgaga bonds, coffee gr:>wers bonds, and TAC Source: Francisco J. Crtega, Notas sobre la reciente evolucion econo- mica e institucional del sector financiero. "Ihie diversity of todaAy's t:inancial intermediaries largely reflects the system 's response to changes in government policies.Fitteer years ago it consisted of a dominant grouLp of commercial ban,:s. :the Aararianr Bank. the BCH and the CF'"s. The system was constrained by numeroLus restrictions and distortions , carried over from past sporadic attempts at directing credit,, financinc government expenditUre, and controlling. foreign exchange.the principal restrictions were low ceiling rates on deposits and loans, forced investments requirements, and directed credit. During the late 1960s and early 1970s, the system began to be gradutally liberalized, partly in line with the Government's philosophy, and partly as the unintentional consequence of other p6licies. In the late 1960s, and early 1970s interest raites on publ-icly held government debt and BCH mortgage bonds were raised, partially to compete with the CAT's; ceilings on term deposit rates were also raised. The starting oi the UFPAC system in late 1972 was a further move in this direction. By the end of 197. the UPACs had captured nearly 20% of the yearly change in non-monetary +inancial assets and 3.% of the year-end stock:. ConstrLuction activities which had lagged in the 1960s, began to play a role in the coLuntry.'s growth process. In ALugust 1974 the eovernment began a systematic financial reform. based on the general philcsophy of giv/ina greater incentives to Sef L C4 t reEtcricl its C.P± ict Y .C: a C, c,r-, cr c- a to t mot.st pro+ i ti e Ct. -i t ieS and o0 r-citC1 na tnte IOerrmel r (I. wt t h the Ll tImate objective o^ improvinq oaarea-co pr- u3tc t ivitY and increasinq the savina rate. T'he rmovements toward easina rest;rictions and el i mi n a i nr distor- ons were. 1-c ever, p-mrtly o++set byw c-tter measures. Fi: rt, thf e the UPAC sy-?stem, was limited by increa-sed ,erve recuLrefment5 Ec a :'Y/% ceiling on the rate of jndexati-on. becornd the 1974 1-,-! Reform nad side ef f ect o- decre2zsing the a+ter-ta-l- r-teturn ar. savirics. In particuilar. taxes were. pl-=ed orn Saving accokuSnt i-eterest and retLurns on CATs. *ihe 1977 stabilization program reverse- the cgains achieved :.n easing restrictions a.nd ela.ninat in- distortions in the - inanciaa2 s,.ystem dureling the first hal+ o-. the I9?/"s') oart. iCU.L.ariyr I . restrain,ing the arowt- o+f llquidity and inflation thro0ugr- tre. repression of financial sector and bv a pol ic oi SC: stringency. This period is marked bv' tre ex:2stenc-e o dras---c controls such as liOO% mnarcinal reserve r-equiremernts e Tab1e I;z The program was motivated by the sharp increase in the -a.te- inflation as foreign exchange earnings rose because of the c o 4- bona,rza and the ilLecal expoort boom. In acdittion, a FL2 -' s5i owdowr; in deval uatI on, relative to l;-cal in± i. ati on. g-. * ; stimul at-ed increased private focreiign borrowi na initiallv,v Since ii;78 the Government has made ar- dual c j u stments in inan ci -i SeCtOr poL icV. In light ±+ =hanrcino e.t -r.I! *. TABLE 11 SELECTED RESERVE REQUIREMENTS (tend of year) 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 2/ 2/ 2/ Demand deposits 31.5 38 18 18 18 18 18 18 18.5 18.5 35 36 43.5 46.5 45 45 45 45. Savings depositb 20 20 20 20 20 20 20 20 20 20 Government deposits -- -- 45 80 80 80 80 80 80 80 60 COD Comm.Banks 19 19 10 10 10 20 20 25 10 10 COD Inv.Banks -- - 20 25 10 10 UPAC savings Dpts. - -- 15 15 15 20 20 13 13 13 UPAC (6m COD) -- -- 10 10 10 15 15 10 6 6 UMPAC (12m COD) - -- --3 3 Ordinary Dept.CAV -- -- - - 15 15 15 15 15 15 Finance companies -- -- 10 10 20 20 20 25 10 10 1/ 100% marginal reserve fot Increases over balance at XI-20-75 2/ 100% marginal reserve requirement for increases over V-31-77 Source: Carrizosa et.al. Analisis Economico (1982) Tabla XI, p 70 domesti c economi c deve.I opments, general 1 y esti nQ restrl Ctl os . 19 7, the lQa. r marcinal *reserve requlremrent on foreian currenc deposits was replaced by an increased average rate. The e+fect o; the intlation orn the retutrns to saving was recognized--t-: brackets and nominail capital gains wiere adjusted by an increasinc fractions of the inflation rate until the Tax Alleviation Law ot 1979 adopted 1007. indexation. In JanLuary 198Cw) the CF's were allow ed to i ssue CDs wjith a free interest rate and a 10%. reserve requirement -- to be invested in the Funds. Banks wiere alsc allowed to enter this market. Reserve .requirements on checkting accounts were lowered in January 1980 and again in September. Finally, in February and May 1982. additional reductions.were made in reserve requirements by permitting the frozen funds to be LusedT as the intermediary's portions of loans discounted with the Funds. The net effect of the above measures has been to increase real returns to depositors and ultimately resource mobilization. In addittion. the measures have improved the competitiveness of the regulated market relative to the unregulated one. If one concentrates on the period 1970 to 1982 , one can identi+fi several interesting trends (table 12). 1; Rapid growith of new financial intermediaries such as the CAVS and the CFCs. 2. Dissappearance of long term instrLuments such as the mortgaae bonrds. of BCH. 3. Decreasing share of demand deposits due to the acceleration O+ in+lation, high reserve requirements End competiticn from other instruiments with high liquidiity. 4. Ratp i d 27 TABLE 12: DISTRIBUTION OF FINANCIAL ASSETS, COLOMBIA 1970-1982 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 19Ev 1981 19:; 1- Financial Svstei 94.9% 94.8% 94.8% 94.2% 91.3% 92.0% 89.0% 85.21 87.6Z 86.7% 87.3% 90.9% 89.; 1. Checkina Actounts 54.8% 52.8% .52.4% 50.2% 45.1% 41.7% 31.9% 43.3% 35.2% 34.5X 29.7% 25.7% 25.r- Coasercial Banks 51.5% 50.1% 49.61 47.4% 42.5% 39.0; 29.4% 31.1% 33.2% 32.3% 27,27 23.9! 23. ' Agricultural Bank 3.3% 2.8% 2.9% 2.8% 2.6% 2.6% 2.4% 2.2% 2.0% 2.2% 2.0% 1.7% 1. 2. Savings Accounts 13.6! 14.3% 16.1% 15.61 14.5% 12.8% 18.5% 17.6% 14.5% 15.8% 13.2% 11.8% 1 1.Z Coeasercial Banks 7.0% 7.5% 8.9% 8.7% 8.4% 8.1% 13.3% 12.8% 9.87 11.1% 9.5Y 8.4% 5,: Agricultural Bank 6.5' 6.8% 7.2% 6.8Z 6.1t. 4.7X 5.2% 4.8% 4.7% 4.6% 3.8% 3.4Z 5-; 3. C Ds 0.9% 0.6% 0.4% 0.5% 4.4% 6.5% 10.4! 11.5% 13.8% 9.2% 16.51 22.8% 21.'% Commercial Banks 0.4% 0.4% 0.2% 0.1% 4.0% 5.8% 7.3% 6.8% 7.0Z 4.7% 12.1% 18.5% 17.'. Investment Banks 0.6% 0.2% 0.2% 0.3% 0.3% 0. 7 3.2% 4.7% b.8% 4.6% 4.4% 4.3% 3 4. UPAC Systes 0.0% 0.0% 0.Z% 9.1% 12.5Z 15.1% 16.0% 13.0% 15.1% 18.1% 18.8% 19.8% 22. Savings Accounts 0.0% 0.0% . 0.0% 4.5% 7.9% 9.2% 9.5% 8.4% 9;0% 12.8%- 13.3% 13. 61. 16.: CDs 0.0% 0.0% 0.0% 4.61 4.71 5.9% 5.9% 3.6%. 3.8% 4.4% 5. 5.7 ?X b. 7 Ordnary Saings 0.0% .0% 0.0% 0.0% 0.0% d.0X 0.7%- 1.0% 2.3. 1.01 0.6% 0d5% O.-I 5. Mortgage Bonds (BCH) 20.6Z 24.1% 22.8% 16.2% 12.4% 9.2% 7.5% 5.3% 4.5% 3.7% 2.6c 12.0 1 a. Contractual Savings (BCH) 2.3% 2.4% 2.2% 1.8% 1.6% 1.3% 0.8% 0.6% 0.5% 0.4% 0. 51 0.6% 0.Z. 7. Finance Companies (CFC) 0.0% 0.0% 0.01 0.0% 0.0% 3.7% 2.7% 3.2% 3.5% 3.7% 4.4% .1% 6. B. Investsent Bank Bonds 2.7% 0.61 0.9% 0.9% 0.7% 1.7% 1.2% 0.8% 0.7% 1.3 1.6% 2. '1 2.?: II- National Cof4ee Fund 0.0% 0.0% 0.0% 0.0% 0.0% 0.4% 2.4% 3.4% 3.6% 2.3% 1. 3 0.8% 0-V. 1. Coffee GrowErs Bonds 0.0% 0.0% 0.0% 0.0% 0.0% 0.4% 1.5% 1.6% 2.2% 1.4% .9 0.6% 0.c f' 2. Coffee Savings Titles (TAC) 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.9% 1.8% 1.4% 0.8. 0.4% 0. 21 0.;' III- National Government 4.9% 4.5% 4.7% 5.7% 7.2% 5.8% 4.5% 2.6% 1.7% 2.2% 1.4% 1.Q0% 0.4T% 1. PAS 0.0% 0.0% 0.0% 0.0% 0.0% 1.2% 3.7% 2.3% 0.0% 0.0% 0.% 0Z 1 .0% O. 2. Economic Development Bonds 2.8% 2.0% 2.3% 3.a% 4.4% 3.9% 0.57. 0.0% 1.4% 1.8% 1.1! 0.77. O.4" 3. Tax Credit Certificates CAT 2.2% 2.5% 2.4% 1.9% 2.8% 0.7% 0.4% 0.31 0.3' 0.3% 0.2% 0.3% 0.0 IV- Central Bank 0.2% 0.7% 0.5% 0.1% 1.6% 1.8% 4.0% 8.8% 7.61 8.8% 10.0% 7.3% Z.:t 1. Participation Titles 0.0% 0.0% 0.0% 0.0% 0.2% 0.6% 0.7% 0.0% 0.0% 2.4. 4.E' 3.0%. 2.5'. 2. Exchanaeable Titles 0.0% 0.0% 0.0% 0.0% 0.7% 1.0u 3.1% 1.5% 1. 5Z 2.0% 2.2% 1.6% 3.!1 3. Exchance Certificates 0.2% 0.7% 0.5% 0.1Z 0.7% 0.2% 0.2% 7.2% 5.6% 4.47 3.0X 2.7% 4.11 ~~- -- - -------- .----- -- ---- --- ------ -------- - - -- ---- ------- ---- ----- --- -- ------- Tatal Assets 100.0% 100.0% 100.0% 100.0% 100.OX 100.0% 100.0% 100.0% 100.0' *100.0% 1 no. . 1o0 % 10.2,. - ------------------- ------------------------ ----------------------------------------- ---------------------------- S.p,arcE: A'rango at al.: El Sector finsciero Colomfbiano inc re-as.- CJt ssets with pcistive rates of retLtr-n such a! UPiL. depposits and CDs. '. Leginnlng o+ Open Market Operations by Jthe Central sank followed by rapid growth of Q. H1. instrLuments SUch as the exchange certificate in 1977 and the participation titles in 1 980.) Table 13; shows the evolution of the main financial assets in real terms. lotal assets in real terms more than doLubled from 1970 tc 19862 growing at a rate of 6.77. per year. The financial system rate of growth was relatively similar to the rate of growth of total assets (6. 2%). Demand deposits remained basically at the same level during the period. Central Bank liabilities increased dramatically in the subperiod 1975-1980. The rate of growth o+- Central Bank liabilities was 22% per year in real terms. Z. Interest Rates During the period 1970-1982 there were major changes in.interest rates with a tendency to a decrease in their variance (tabie i4). Return in savings deposits of commercial banks and in CDs were adiusted upwards while CAVs borrowing rates were moved downwards. The average interest rate increased by 12 percentage points from 14.4 in 1972 to 26.2 in 1980. Since the inflation rate moved frof 14% in 1970 to 24,.' real interest rates remained practicaliv eq±lai to zero during this period. 29. TABLE 13 FINANCIAL ASSETS COLOMBIA 1970-1982 (1970 PESOS 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1- Financial System 23916 23735 24473 25942 23914 27575 29325 34917 36350 35103 42646 50199 49362 1. Checking Accounts 13007 13222 13533 13836 11824 12491 10490 13645 14583 13976 14489 14173 14064 Commercial Banks 12981 12532 12796 13056 11134 11700 9698 12723 13774 13091 13526 13211 12998 Agriculturil Bank 826 690 736 780 691 792 792 921 810 895 962 962 1066 2. Savings Accounts 3425 3588 4169 4290 3801 3850 6094 7224 6023 6387 6465 6515 6059 Commercial Banks 1774 1875 2299 2404 2209 2434 4389 5262 4059 4509 4622 4625 3268 Agricultural Bank 1651 1714 1870 1886 1591 1417 1706 1962 1964 1878 1843 1890 2790 3. C. 0. D. 239 140 103 125 1149 1947 3430 4692 5705 3731 8052 12600 11576 Commercial Banks 99 90 45 34 1059 1730 2393 2783 2905 1884 5900 10239 9501 Investsent Banks 140 50 58 91 90 216 1038 1909 2800 1847 2152 2361 2075 4. UPAC System 0 0 0 2501 3V86 4525 5279 5312 6256 7344 9210 10948 12355 Savings Actounts 0 0 0 1226 2061 2745 3116 3445 3745 5165 6485 7529 8826 C.O. 0. 0 0 0 1274 1226 1780 1936 1467 1562 1761 2426 3136 3318 Ordii.ary Savings 0 0 0 0 0 0 227 399 949 417 299 283 212 -5. Nortgage Bonds (BCH) 5186 6032 5879 4453 3260 2761 2483 2169 1847 1489 1258 1110 535 6. Contractual Savings (BCH) 576 597 563 498 417 403 268 227 205 168 262 332 318 7. Finance Companies (ZFC) 0 0 0 0 0 1098 894 1320 1450 1483 2134 3381 3307 8. Investsent Bank Bonds 683 156 228 239 175 501 387 329 292 525 777 1140 1148 II- National Coffee Fund 0 0 0 0 0 129 804 1385 1493 915 641 450 295 1. Coffee Growers Bonds 0 0 0 0 0 129 497 662 893 586 439 355 289 2. Coffee Savinos Titles ITAC) 0 0 0 0 0 0 307 723 600 329 202 94 7 lIl- Natianal Government 1246 1115 1213 1568 1879 1730 1483 1071 707 882 673 533 240 1. PAS. 0 0 0 0 0 357 1209 958 0 0' 0 0 0 2. Economic Development Bonds 696 499 593 1037 1142 1170 153 0 577 747 560 379 240 3. Tax Credit Certificates CAT . 550 616 621 531 737 204 122 113 130 135 113 154 0 IV- CentrXi Bank 46 187 132 37 413 529 1322 3587 2922 3576 4999 4046 5295 1. Participation Titles 0 0 0 0 S5 165 218 0 0 980 2365 1663 1285 2. Exchangeable Titles 0 0 0 0 172 307 1033 635 611 799 1052 875 1723 3. Exchange Certificates 46 187 132 37 i84 57 71 2952 2311 1797 1482 1507 2.287 Total Assets 25208 25037 25918 27547 26206 29964 32934 40959 41473 40476 48859 55227 55192 -Source: ---no -t --- - -.---- -- ------.- - El S i- Co--in Source: Arango et al., El Sector Financiero Colombiano TABLE 14 SELECTED DE-POSIT RATES COLOMBIA 1972 - 1981 YEAR StiALtOry Implicit Savings COD Savings Free Excahnags Participation 6 month COD Average rates rate deposis.- Banks market certificates titles Gut Bank rate Bills (1) (2) (3) (4) (6) (7) (8) (9) (10) (11) 1972 19.3 19.6 5.7 13.1 13.6 14.4 1973 26.4 26.2 28.8 8.8 18.7 13.6 17.6 1974 27.0 26.2 .27.4 10.1 27.1 21.6 18.6 20.6 1975 26.0 24.5 , 25.7 18.3 26.0 21.0 28.8 26.2 20.3 1976 23.0 24.0 22.3 23.5 18.7 25.5 21.6 28.6 26.2 19.4 1977 22.8 23.3 22.7 23.9 19.3 26.0 24.4 28.6 26.2 24.3 1978 22.5 21,5 21.6 22.4 109. 31.0 29.6 26.2 23.1 1979 24.1 23.8 24.7 27.5 19.3 33.9 83.2 34.6 25.7 23.6 1980 25.6 24.2 27.0 29.5 20.2 34.0 30.0 38.3 34.0 26.2 1881 27.4 26.0 27.0 29.5 22.7 37.3 30.2 30.6 37.3 Source. Carrizosa et.al. Anallsis Econ6mico - Table XIX p 157 1 ' I he pri nci p al c va1 v erble-s a+ +ecti rti both thtG nominal and re.-: deposit rates are: (a) the e:pected rate o+ inflation (b) th- expected rate of devalLiation; (c) ceilings on deposit rates; anc (d) rates of taxation on asset income. O: these, the expected rate of inflation is probably the most important. For most assets. particular'y those held by savers who mainly invest in Colombia mark et, the nominal rate of return has roLughly followed the inflation rate. The relationship is not exact becaLuse of errors ir, expectations, ceilings on deposit rates, competition from unregulated assets and variations in the . underlying rate o, return . In the last two years the expected rate of devaluation seems to have had a maior effect bn the CD rate. Since 1980, when the rate w.as +reed, it closely followed the US CD rate adjusted for devaluation, even thoLugh this implied an annual average real rate of .5.5%. Thus, the 1980-81 rise in CD rates can be said to reflect mainly the run-Lp in US rates.. The relationship between the two "'D rates is probably a reflection of the behavior of Colombian investors, who regard the instruments as good substitLttes. * 4. Resource Mobilization Over the past +ifteen years, the ratio of total saving to GDPP irn cLurrent prices has remained suLrprisingly constant on average, rising +rom 207. to .22%, despite the doubling Or the average date .... - . . , -.-l.. . ,-, .. , ........., ,. . .. ....,.. !...~ s,.e.. o+ inrt1atiori betwe-ri the 196l-74 annd 1975-L-40 perio:ds; While overali. nominal savinq rate increased by about 10'. and the personal saving rate by even more (averaging slightly over 6'/' o+ GDP dutring 1978-C)). the real investment rate declined slightly. BUsi ness saving has been the main source of investment finance during both of the periods, averagina about 50% of investment. Within business saving, depreciation allowances have been three to four times as large as retained earnings. 1Most of the year-to-year variations in business saving resulted from cyclical fluctuations in retained earnings, with depreciation allowances fluctLuating less. The only important shift in the sources of saving ocurred beteween. personal and net foreign saving. Between i9b7 and 1974, capital inflows financed an average o+f 13.5% of investment and personal saving 8.1% (about 3% and 1. 6. of GDP respectively). Most of the capital inflows went to the public sector from bilateral and .multilateral sources. There were also small amounts of commercial loans and direct foreign investment. On balance, the private sector was a net purchaser of foreign assets (effectively a net seller of foreign exchange), partially offsetting these inflows. In contrast, between 1975 and 198Q, net capital outflovws averageg about 4% of total investment (or abouit 0.9% of GODP). while hoLusehold saving financed 27;.r of total invJestment (aboLlt 6%/c GDP). Thus, dom6stic resource mobilization increased as a fractian. of GDF because of the rise in personal savinq and.a rcuomlve 33 constant rate of 6overnment savinr, . Lbut the ratio o+ investmerit t- GDPF remained constant owning to a fall in the rate o1: +oreic- savingg The public sector increasingly tapped domestic privat- saving to finance its investment program. Two possible reasons for the increase of the ratio of personal *saving to GDP can be Qiven: Firsts households increased their saving rate in order to spread oLut over time the tranrsitory gains obtained from the coffee and illegal ex>port boom. Second, the Government maintained a roughly constant positive real interest rate wliich induced households to maintain a roughly constant ratic o+ interest bearing financial assets to GDF. Although the higher rate of inflation during i975-80 reduced the ratio of non-interest bearinQ assets (the money suipply defined as currency plus demand deposits) to GDP, it did not do so by ver;y much. The net result was that gross saving in the forrm of f financial assets actLually rose as inflation rose to about 25f. on average during the mid-and 1 ate- 3I,y 70s. The particular e>tperience in Colombia--where attempts were made to mai ntain positive real interest rates, but credit growth was restrained-- meant that pri'vate saving rate and international reserves increased as inflation rose. The . resulting low net international debt position has allowed the current Government greater flexibility in deali*ng with the fall in coffee prices. Finally, the rise in domestic savinQ was accompanied by inflation a^nd an increased dependence on the Central iDank.'s FuLnds, whic- worsened the allocation o+ credit. -These developments suggest the +followinng conclusions or fruther reSOUrce mobi lization: '.c positive real interest rates on deposits should be maintained t: preserve Colombias good record on r-esource mobilization durinc inflation; (b) as inflation is redLLcedq any decline in private saving should be offset by increased pLublic saving; and (c.: forei6n borrowing should proceed cauLtiously, essentially tL finance high priority projects and taking account the public sector's debt servicing capacity. 5. Current Conditions The Colombian financial sector has just passed throudh a crisis. The Superintendency of Banks, which is in charge of controllinc the financial intermediaries intervened in July 1982 one of the financial groups. Thereafter it has intervened another bankr several CFfs and CFCs. Some of these intervened institutions have been nationalized while other haJve gone out of bLusiness. Since there is no deposit insurance in Colombia some investor have movec their funds to minimum risk instituLtions sLuch as government ownec Banks and CAVs. C. HOUsing Se-ctor i. The Stock of Housing Estimates of the housing stocik exi.st in Colombia only for the censLus years. Although, +figLres for intermediate years can be estimated, in principle, by the perpetual inventory method, that is by adding to the stock in a censuLs year the newJ uLnits butilt ir, that year and subtracting the number of units retired from the stock. In practice, it is not possible to do this since there is no information in the number of units retired from the stock and even more problematic, the number of new Llnits are highly uLnder reported, because a large fraction of new utnits are built by the informal sector. It has been estimated that almost fifty percent of units are built. As an alternative some researchers (Carrizosa et al.., 1982) have used the value o+ rents from the national accounts to derive figures of the value of housing stock . Table 15. present Carrizosa's estimates. According to these estimates the percent growth of housing stock, measured by the ratio of the valu.te of investment *in housing to the value of the hoLusing stock, has increased more rapidly than population. For example between 1951 and 1964 the growth rate of hou-sing was 9.4% or more than three times the growth of population (2.87%). If these estimates are correct it is possible to conclude that the Va Ilue of the hoLcisrig TABLE 15 HOUSING STOCK: COLOMBIA 1950 1980 Housing Percaptta Housing Gro FC _ousing Stock Intercensal growth rate stock rent Investment Annual (%) Intercensal Total (%) Urban. (%) Year (mnill. 1970 $) average (1) (2) (3) (4) (5) (6) (7) 1950 20,170.oo 208.8 2, 265.3 11.2 . 1951 19,948.33 200.1 1,597.9 8.0 1952 20,670.83 200.8 1,811.2. 8 .8 1953 21,532.50 202.7 1,955.6 9.1 1954 22i477.50 204.9 2,497.9 11.1 1955 23 644 16 208.8 2,191.8 9.3 1956 .24,921.66 213.2 2,384.2 9.6 1957 26,310.83 218.0 2,519.1 9.6 1958 27,784.16 222.9 2,695.7 9.7 1959 29,758.33 231.2 3,114.6 10.5 9.4 2.8 5.6 1960. 31,980.83 240.9 2,796.9 8.7 . .) 1961 34.229.16 249.3 3,127.3 9.1 .I 1962 36,425.oo 257.0 3,195.7 9.9 1963 38,925.oo 267.6 3,311.7 8.5 1964 42.317.50 280.7 3,460.9 a 8.2 1965 45,343.33 291.8 3,465.2 7.6 1966 48,317.50 302.0 3,490.9 7.2 1967 51,428.33 312.3 3.445.3 *6.7 1968 54,484.16 321.7 3,715.2 6.8 7.0 2.26 4.6 1969 57,791.66 332.3 3,736.2 6.5 1970 61,125.oo 342,3 4,474.7 7.3 1971 65,159.16 355.5 4,449.0 6.8 1972 69,255.oo 370.0 3,825.9 5.5 1973 72A 677.50 378.1 5,157.3 7.1 1974 77,812.50 296.1 5,151.0 6.6 1975 83,902.50 417.4 3,729.2 4.4 1976 88J180.83 430.2 4,043.4 4.6 1977 93,399.16 445.7 5,001.3 5.3 5.2 2.2 3.2 1978 100,611.66 470.3 5j770.4 5.7 1979 109.345.oo 500.7 4,780.6 4.3 1980 116,592.50 523.1 4b618.1 3.9 ' Sources; Carrizosa etal (1982) and Table 32 5tctck percapita has increased very rapidly in Colombia. In spite of the relatively rapid rates of urban popLulation growth in Colombia in 1970s, the pLiblic corporations charged with providing infrastructure seem to have done a qcod job. Among the aprox imately 18 million person living in urban areas, 81 percent have access to piped water and 65 percent had access to sewage conhections in 1981. Electricity has practically become uLniversal in most urban areas. Several characteristics of housing construction are common to developed and developing countries. Among them, housing generally presents a cyclical behavior, In Colombia as can be seen in graph 2 housing construction presents a cyclical behavior. Peaks in housing construction c.an be observed in 1954, i959, 19624, 1970. 1973, and 1978. As in other countries aggregate housing demand has been shown to depend primarily on the growth df the nLumber .of families, the growth cf household income, and the decline in prices. According to econometric evidence income is the main explanatory variable of housing demand. Studies of the. housing market in Colombian citi£es have shown remarkable similarities between houLsing demand demand eqLtations in Colombia and other countries (Ingram, 198;1 Strassman, 1 2) 2 38 f25 i I F Housi 1.ng A Investment Percapt&-- - (1970 Cot$ -. I. -..,e' -I -! -''i 1nrcome cAnd price elasticities derived from hotsehold survey datE are less than one and similar in maanitLtde to those toLUnd in th=- U.S. using microdata. Income eslasticities of housinrg demanr derived from time series are greater than one. Studies of the effect of the amoutnt of credit in housing ouLtput provide conflicting evidence. One of the increase in specializ-ed&> hoLusing firnance have had a posi'tive influence (Sistemas de Informacion, 1982), while other did not find any influence (Carrizosa et al., 1982). If- to the econometric evidence reviewed above we add likely macroeconomic outcomes slower economic growth, decelerati nc popul.tion growth, it is very likely that housing demand will grow very -slowly in. the near future. 'Consequently, the additiornai houtsing output considered in the new development pl an could have difficulties on finding households interesed in them. It is verv likely that the additional formal sector units will substitute' for units taht are currently provided by the informal sector. The evidence on income distribution in Colombia points to high unequal distribLution of households income. Estimated i ncome distribution for 1983. (Table 16) shows that the poorest half of the popLilation receive only one sixth of the total income. The Gini coefficient for household income distribution is estimated as -0.52 showiinrG considerably income inequLality. 40 TABLE 16 ESTIMATED URBAN INCOME DISTRIBUTION COLOMBIA 1982 Decite Income % of % of Population % of Income Range Income (Cummulative) (Cummulative) 1 0 - 6.856 1.1.2 0.10 1.12 2 6.857 - 10.003 2.41 0.20 3.53 3 10.004- 12.966 3.15 . 0.30 6 6. 4 12.967- 16.491 4.23 0.40 10.91 5 16.492-20.989 5.33 0.50 16.24 6 20.990- 26.763 6.70 0. 60 22.94 7 28.764- 35.336 8.59 0.70 31.53 8 *35,337-.50.514 11.49 0.80 43.02 9 50.515-117.967 17.47 0.90 60.49 1 0 117.968-and more 39.51 1.0 100.00 Gini Coefficient 0 . 52 41 This ineqLtality of income in Colombia has several conseqLuences -For housing firnance. First. households in the poorest deciles will find very di++icLtlt to afford dwelling uLnits of "reasonable" quality. To satisfy this demand it will be required to decrease hoLusing standards. As the experience in Bogota has shown, adopting minimal standards contribute to satisfy the demand within the legalized sector. Secondly, increasing resource mobilization by poor households will reqLuire some institutional innovations that address their specific problems. 2. The Informal Sector 1/ Unlicensed construction (in historically unauthorized sutbdivisions) is responsible for the creation of about 60, 0,C units a year, or half the overall increase in the stock[ of urban housing, (Table 17). The prevalence of this unofficial construction process suggests that shelter policy should be based on a knowldge of some of its pecLuliarities. First, the households in the informal sectors tend to develop their units incrementa lly, over a period of years, varying the rate to suit household composition, tastes, income, and assets. --------------- 1/ The material on the informal sector is based in work done by fndrew M. Hamer summarized in the Colombian Urban Sector FReport (World Bank. 19B3 c). 42 TABLE 17 HOUSING OUTPUT BY FINANCING AGENCY CO LOMBIA 1981 Financing Units built ° Agency I.C.T. 20,000 16.7 B.C.H. 10,300 8.6 UPAC 21, 000 17.5 F. N.A. 5,500 4.6 Others 1/ 3,000 2.5 Total 59,800 49.9 Informnal Sector 2/ 60,000 50.1 Total 119,800 100.0 1/ Caja de Vivienda Popular (Popular Housing Bank) Caja de Vivienda Militar (Military- Housing Bank) 2/ Estimated 43 This means, amorag other things, that a static dwelling coLunt W1i. miss the quantitative and qualitative improvement in the unit tha- will take place over time. As an example of this process, one car cite a Study of Cartagens (Strassman, 1983) that found that rooms per dwelling unit rose 24 percent in just five years in the mi c seventies. Second, it follows that for a large number of hoLseholds, shelter conditions at the begining of their family life cycle will be significantly worse than at later point in time. Shared hoUsing, whether at lower or higher incomes, is a rational response for reducing expenditures on houtsing when household heads are young, incomes are limited, family size is small, and residential and job mobility are high. At the same time, shared housing places less -pressure on the extension Of the city and its inf rastructure by allowing for increased densities in built-up areas. Third, the self-help shared housing mechanism is also an important source of income for the owners of unlicensed, unauthorized housing and at the same time contributes to th-e supply o+ rental units. Detailed estimates of shared houtsing suggests that, for large citien like Bogota, the principal Souirce of rooms-for-rent is now the self+-built, peripheral dwelling and not the inuLtilinatca or more central, commercial rooming hoDtse, as is often believed. .. . . . . . .. . . . . .. .... .. 44 No.-+tinancial Constraints 1/ Some generalizations on non-financial constraints on hoUSino sector do appear possible. Raw land seems to be transferred from rLural to Lirban Luse in a fairly efficient manner. However, the servicing of that land with infrastructure is slowed by numerous hurdles, including unrealistic development standards, cLumbersome administrative procedures and lack of coordinated metropolitan investment planning. Yet, over time, the proportion of serviced households for different types of basic infrastructLure i s increasing rapidly. Finally, throutgh mechanisms such as unlicensed construction of dwellings in the informal sector and the widespread utse of shared and rental housings the urban popul ation' has been experiencing an improvement in shelter, be it measuLred in terms of dwelling space, structLural quality, or services. The market of raw land in Colombian cities appears to operate reasonably well. Conventional wisdom in Colombia points to "exaggerated" price increases as evidence that owners of large blocks. of undeveloped land of the periphery of cities are holding land off the market for long-term speculative gains, thus severely constricting supplies. Carefully docLumented evidence fo'r such a --------- ------ 1/ The presentation is based an reserch done for City Study staff and sLummarized in the Colombian Urban Sector Report- (World EankS 1983 c). 45 hypothesis dc5s riot seem to e,xSt, however. On the contrary, studies which reviewed data for Bogota sLuggest that prices +or land up to 4 tkm from the center, or average, e-xperienced stabilit/ or declines in real values. Areas located 4. to 1C kms from the center increased in value at an average real rate of 2 to :%. per yearR while in more peripheral zones land prices r&se at an average of no more than 5'Y per year. If there is one area of shelter provision that may merit c,loser attention it is the process of improvement of existing Lunits. At presernt, Lhe only national inventory of housing qualit-y is the 1973 demographic census. Looking at the dwelling occupied by houLseholds living in the country seats, one finds only two-thirds of the units were constrLlcted with only "permanent" m,aterials , conversely only five percent were "shacks" built of nothin but temporary materials. . CoLunty seats, or cabecera, data for major urban areas of Colombia sLuggests that, as of 1973. 2(0 to 5C0 percent of all units could benefit from some form of physical upgrading, i.e. substitution of temporary -for permanent bLlildina materials. Research on Bogota and Cartagena suggests that poorly developed access to loan funds in the formal financial mark%ets among lower income owner households combined wiith delays in the delivery of infrastriucture arekey reasons why shelter upgradina does not proceed more expeditiously. These prove to be a critical bottleneck, more so than the rate ot new shelter creation. As the rate of popuLlation growth declines in uirban areas in the 198C)s is 46 the condition oQ the enormrous existinq stock of hotIsing be come a irnpor-tant fOCLiS of concern for poiicymakers. I I I. HOUS 1 Ihl F 1Nf1'-4CE A~. Trends in Level of Intermediation When the Colombian Finance System was reformed in 192:3 the financing of housing was assigned to the commercial banks through their mortgage sections. Because of the 19-sO5s crisis and the fa'ilure of some banks the mortgage sections of the banks were closed and the BCH (Olortgage C;entral Bank) was established. In 1942 the Land Credit Institute (ICT) was tounded to serve low income housing. In 1972 CAVs were established and after a rapid development have become the main sources of housina financing. Other financial intermediaries sLucha as instrance companies and £ap talizadoras have granted some mortgage loans. Funds to finance housing finance come also from .the commercial banks. from developers own resources and from the unregulated financial market. Although, a large fraction of the funds that finance housing c6mes from the public there exist cLumpolosory schemes that provide fLrnds for housina. The main are reserve requirements, Social InsLurence Trust Funds and severance payments reserves of pLublic employees. Funds also come from governments that transfer funds to puLblic hot-tsing agencies to cover operating de+icits. 1 he Lol ombl an covernment as part o; its devel opment strateqv hac, tried to increasCe the level of intermediation ot the houS:nC finance sector. If one looks to the ratio of new fLtnds for hoLusinc finance to the value of investment in new housing one can see- (graph 3) that in 1967 there was a dramatic shift. In 1973- with the start of the UPFAC system the ratio attains a peak. It is important to notice that the increase in the coverage of the hoLtsing finance sector corresponds to a shift from other sources. As can be seen in tables 18-20 several indicators of hoLusing finance such as individual mortgage credit, construtction financing in+ormation from ..building permits and new loans to construtctior, show the shift in sources of financing construction. The tables also show the evolution of the three major soLtrces of Fi.nancing: the Land Credit Institute, the Mortgage Central Bank .ana the CAVs. Several subperiods can be defined. One frcm 1950 to 1 958 in which commercial banks played the major role in financing ronstruction, a second from 1958 to 1972 where government institutions such as BCH and ICT were .the main actors, and finally a third from 1973 to the present in which private institLutions such as the CAVs become the main providers of construction funds. The main institutions financing housing in Colombia are the privately owned CAVs. The public sector instituttions speciali-zinq in housing finance are the Mortgace Central Sank, the Land Credit InstitLute4 the National Savinas Fuind and Lo Sal v3avinQs EBanks or 49 TABLE -18 INDIVIDUAL MORTGAGE CREDIT COLOMBIA 1950 - 1981 (millon pesos) Year B. C. H. CAVs Insurance Capitalization Companies Companies (1) (2) (3) (4) 1950 116 23 n.a 1951 135 31 n.a 1952 168 40 n.a 1953 198 19 n.a 1954 221 23 n. a 1955 233 27 n.a 1956 293 34 n.a 1957 344 43 nria 1958 399 57 n.a 1959 504 62 n.a 1960 626 68 51 1961 712 84 72 1962 810 100 105 1.963 ,958 - 114 ,116 .1964 1, 080 136 138 *1965 1,227 154 158 1966 1,397 163 171 1967 1,578 176 199 1968 2,100 195 229 -1969 2,973 n.a n.a 1970 4,523 n.a n.a 1971 6%821 n,a n.a 1972 e', 917 14 n.a n.a 1973 11, 370 1,124 n.a n.a 1974 10,991 4,396 n.a n,a 1975 10,977 6, 924 n.a n.a 1976 . 10, 04 1 1, 868 465 n.a. 1977 10,788 18, 146 560 n.a 1978 11,409 28, 113 n.a ri. a 1979 12,440 40,050 n.a n.a 1980 16,124 59,103 n.a b.a 1981 19,824 79,349 n.a n.a Source: Carrizosa etal (1982) 50 TABLE 19 CONSTRUCTION FINANCING BY SOURCES ACCORDING TO BUILDING PERMITS 1966 - 1980 Year B. C. H. Other Private Non-profit owri Total I. C. T. credit irvestors Institutiors resoues (millons pesos) (%) Instit. (%) (%) Currents Constants (1) (2) (3) (4) (5) (:s) (7) 1965 15.1 2.7 n.d. 82.2 1,446 2.325 1966 16.2 3.2 n.d. .80.6 1, 344 1.911 1967 15.5 2.2 n0d. 82.3 1, 550 2.034 1966 . 26.0 2.4 n.d. 71.6 2,121 2.516 196e 28.7 1.9 n.d. 69.4 2,294 2.529 1970 33.4 1.8 n. d. 64.8 2,,660 2.660 1971 31.2 5,3 3.8 0.7 59.1 31262 2.839 1972 19.2 5.7 5.4 0.9 68,7 3,175 2.435 1973 14.8 15.2 3.8 0.7 65.6 4,362 2.853 1974 10.6 26.2 3.2 1.0 59.1 5,748 2,820 1975 9.8 23.2 3.4 1.3 62.3 5,479 2.180 1976 7.6 19.7. 5.7 1.5 65.5 7,634 2.542 1977 14.1 23,8 3.5 1.2 57.4 12,940 3e361 1978 14.4 25.1 3.4 1.0 56.0 17,365 3,441 1979 9.4 25.8 5.2 1.4 58.2 17,731 2,504 1980 12.0 27.9 4.4 05 55.1 27,870 2.907 Source: Carrizosa et. at. 1982 Table 1 Page 11 51 TABLE 20 DISTRIBUTION OF NEW RESOURCES USED IN CONSTRUCTION COLOMBIA 1950 -1980 Year, B. C. H. C.A.V. Commercial Irvestment I. C. T. Total Total (0) (%0) Banks Banks (%) (Million (million (%) (%) c4rrens constant pesos) pesos) (1) (2) (3) (4) (5) (6) (7) 1950 42.8 48.2 8 ;9 56 368 1951 40.3 47.8 11.9 67 416 1952 45.4 47.5 7.1 99 611 1953 38.3 49.6 12.2 115 688 1954 28.9 47,1 24.Cr 121 688 1955 27.6 - 51.0 21.4 145 780 1956 34.8 57,2 8.0 187 926 1957 32.4 52.2 15.3 222 969 1958 35.0 43.5 21'.4 248 939 1959 48.3 39.1 12.$ 294 1,1O10 1960 45.5 37.0 17.5 389 1,182 1961 40.1 33.8 26.1 414 1,193 .192 32.5 31.2 36.3 603 1,562 1963 32.0 29.8 38.2 762 1,536 1964 35.0 41.3 23.7 669 1,190 1965 37.6 , 44.0 . 18.4 700 1,125 1966 34.4 43.6 22.0 884 1 s257 1967 36.7 s38 .9 24.4 1st98 1 ,572 196f 53.2 26.5 20.3 1,959 2,318 1963 64.0 19.4 16.6 2,403 2,349 1970 68.3 15.0 16.7 3,,365 3,J65 1971 70.3 13.5 16.2 41,642 4,040 1972 52.1 2.1 20.8 25.0 4,119 3,159 1973 22.4 46.5 15.2 15.9 7,103 4,546 1974 4.3 68.2 12.0 15.5 9,703 4,761 1975 9.9 50.7 16.5 22.9 7,431 a,957 1976 5.5 42.5 20.4 12.5 19.1 10,877 3,622 1977 7.7 45.8 23.4 8.4 14.7 19,721 5,1*22 1978 6.4 53.4 19.4 7.6 13.2 29,821 5,910 1979 3.7 55.2 19.0 9.1 13.0 25,865 3,652 1980 11.6 52.5 16.5 10.2 9.2 'Source: Carrizosa et.al. 1982 Table II Page 13 52 . , I ,j , -- -iQRA.PH±. 2 Mortgage ! - i Financa - I Ratio -! ., 19. .. !6 :? . t . - . -1\-- . . ..1I- ._ _ _ -_ _ _ . _ __,_ _ '. _ _ _ _.__ _ _ _ _ _ 53 Caj, de Vivi.end,i. Ile CAVs are other +irnancial intermediaries are cont:rolled by tht.e Monetary Board and by the Superintendency o-f Edanks. 1he national hoLusing agency irn principle are control1ed byg a 111inist-tr butt in practice they are aLutonomUs. The- general manager is selected directly by the President and sometimes is more power+ul than the Minister in charge of the Instittution. In order to solve the problem caLused by the lack: of coordination among hoLusing agencies it has been proposed the creation of a new Ministry of HOLIsing and Urban Development. The current government has established a Housing fAdvisor to coordinate housing policies. in the past there have extisted some short-lived committees and a Housing and Saving Board. In th& following sections-a detailed descriptions of each one o0 the three major institutions will be presented while the ne:.t chapter will discLuss the main issues in hoLusing finance B. The Land Credit Institute (ICT) 1/ This is the national houising institute.It i s affiliated witrh the Ministry of Development.It was founded in 1942. CuLrrently it offers four different type of plans.In each one of the plans there ---------------. 1/ This sections is based mainly in several ICT annual reports, *DNP (1941) , Camacol (1983:). and the Colombian Urban Sector FRocrt- (World Bankq 11583 c). are sever-il classesi n+ oroarams. These four planc are: I- -roaressive development o+ below normal zones (D.Z.S.) -Technical assistance. -lenulre legalization. - -Public services dotation -Housing improvements. -Community dotation. II- Progressive development housing construtction(C.V.D.) -Serviced plots. -Core units. -Minimal soluLtions. -Basic solutions. -Tn+S?rmmediate solutions. -Ma ximal solutions. I IIi-§-ommunity Services Construction (I.S.C.) IV Community Development. ICT is then the responsible tor the construction of new hausina for low income hoslSeholds. ICT financed apprcO:mately one third o4 the dwelling units builtW by the formal sector. Thease 2 OC)4ooo units used one ninth o+ the financial resOurc's cf the sector. ICT is al=o responsible for other plans suLch ze the inrt.egrati-n o- c:'-rvices and commtinity par-ticipation that look. +or the improvement b t the utrban environment. In part dLue to the. muLltiplicity of objectives ICT has been characteri-Zed by some inefficiencies. It also has su++ered of variations on +unding. As a consequence of financial liberalization dLuring the seventies ICT soLurces of compLilsory low yielding investment have decreased substantially. Table 21 shows that ICT's reSOLurces have been very unstable both in th,Eeir Value and in their composition. ICT's resources have increased at an annLual rate of 21 % in nominal terms,- bLtt only at 1% in real terms. The main sources of funding .for ICT (Table 22) are: 1) Current revenues generated by the Institu-te activities suich 'as i nterest payments4 amortization payments,down payments and what are called commercial operations which consist mainly in sale of land to private developers, 2) capital revenues including: a) budget appropriations (ordinary, special, and amortization of Development Bonds), b) domestic credit consisting of compulso&ry investments ot: Insurance Companies,, savings departments of commercial bank:s; short term borrowing; scme points of the reserve requirements O-T the commercial banks; borrowinQ from the National Saving Fund (FNA). the Central Mortgage Ba:nk (BCH)., the Banco Fopular, and the HoLtsing and Savings Corporations(CAVs) j and previo.isly c) Foreign credit consisting o-f IDB and i;i7D {uninds. 'TABLE No. 21 INSTITUTO BE CREDITO TERRITORIAL - REVENUES (*Utliton Col Si Concept 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1931 l. Current Revenues 401.6 474.2 N.D. 728.1 970.9 1w140.3 1,396.3 1,793.6 2 758.3 3,023.0 3,961.2 3,676.1 (11 11. Capital Revenues 509.7 756.3 1I.0. 1 439.7 1A423.2 1,652.0 1,903.9 2,590.0 3,027.0 2,634.2 3,612.0 6,008.8 A. Government Transfers 128.2. 124.1 N.D. 285.7 298.9 430.6 410.8 681.5 691.3 1,168.0 1,764.8 2,099.8 0. Domestic Credit 352.3 513.2 N.D. 890.6 575.3 I1,53.0 1,336.8 1,731.0 2,274.2 1,410.8 1,847.2 3,908.9 C. foreign Credit 77.4 62,4 N.D. 147.3 330.7 - - - - - - III.From previous year 31.8 56.5 tl.D. 116.1 218;3 68.4 156.3 177.5 61.5 55.4 Q7.3 290.4 TOTAL 991.3 1,230.5 1,976.1 2,167.8 2,394.1 2,792 ..3 3,300.5 4,383.6 5,785.3 5,658.0 7,676.6 9,975.3 Percent Change 24.1 60.6 9.7 10.4 16.6 18.2 32.8 32.0 (-2.2) !5.7 29.9 SOURCE: D'IP "%LaEdJficacO6n de Vivienda en Colombia". Revo:'>j de Planeaci6n y Desarrollo. Volumen XIII. No. 3. Julio- C/lnu 1. JCi. Annual Report. TABLE No.22 ICT. REV ENUES BY SOURCE 1980-1981 Concept 1980 1981 I. 'CURRENT' REVENUES Interest 14.3 12.3 Amortization payments 13.4 13.9 Down payments 6.4 4.1 Commercial Operations 13.6 2.8 Others . .4.0 3.8 II. CAPITAL REVENUES Government transfers 22.3 22.6 Credit . 24.1 43.1 Others 0.7 0.5 III. FROM PREVIOUS YEAR 1.3 3.2 TOTAL 100. 0 100.0 SOURCE: ICT. Annual Report, 1981. As can be seen in the pre vious tables, cuirrent rever,enes hatve be:- relatively stable. How3Jever it is very likhely that CLirrent re:'enUez- cou.lid be improved since more than halt o+ ICT loans are delinUentr ¢table 23). Furthermore delinouency rates have been increasinq ir the recent past. On the other hand, capital infusions have filuctuated widel.y. Fart of the 'variations were dLue to changes in the financial sector that have been mentioned above. Other were responses to chances in the ioreign sector cautsed by the boom in exports and the corresponding increase in foreign reserves. The government decided that the housing sector shoLild be financed with domestic resoLtrces. Pai-t of the changes are also due to change in the role played by pLtublic houLsing agencies. It was decided in 1972 that the priVate sector should have a major role in _the financing o+ housing and Oonsequently the public sector shouLld have less importance. Fi.nally some of the fluctLtations are due to changes in f iscal policy. DLlring the trade sector boom part of the adjULstment was borne by the government sector which adjusted capital outtlays in order to generate a fiscal surpIlus. T he present government has chosen housina as one ol the 1 eadinQ sectors and has targeted most of the putblic sector construLci on to iow income families. in order tzo provide resCUrces fQr these pIans it l,as made seVeral ci,ancies in the sOurce ot [CT r; -ancing. na Omnla TABLE No. 21 ICT's OUTSTANDING LOANS 1978-1981 Concept December 78 December 79 December 80 December 81 OUtstanding loanS (Amount) 7,851.9* 8,976.2* 10,724.2* 13,537.6* Amortization payments 1,403.5* 1i685.1* 1,920.5* 2,449.6* Amottization in December 117.7* 151.8* 164.9* 188.3* Delinquent loans (Amount) 217.3* 239.9 312.3* 414.1* National Delinquency Index 2.76% 2.67% 2.91% 3.06* Outstanding loanis (number) 164 405 161J 057 162,700 169,983 (D Delinquent loans (nUmber) 77,947 77,189. 81,673 88,398 Delinquent payments (number) 383, 528 361,660 417,294 453,370 Delinquent loans (%) 47.4% 47.93% 50.20% 52.0% SOURCE: ICT. Annual Report 1979-1981. * Millions of Col$ themn: 1) new boni o-f popCt ar hoLuSinci with 4 years to mati.tr t an: nominal inter-est ratecs o4 2Y.'),' 2) butdqet appropriations of- 88 . mi11irse of col peso5s , .) a IC) percentage points incr-ease ir reserve requtirements of commerci.al banks savinqs deposits from 2C- Lip to 3C) points. From these 30 points, 13,.5 points will go to finance ICT, 1iC) will be invested in the new popular hoLtsina bonds (N .EV.F.)q Sand three and a hal$ points in B-Class Savin:s anc HoLIsing Bond, and 4) a requirement for the CAVS to invest two oF the 8 points of savings deposits reserve requirements in NBFvp. it is estimated that. the new bond suscription will generate ColG 12,804.7 millions. ICT exipenditures* have fluctuated folowing changes in revenues (table 24). A major shift in investment took place in 1951. Investment outlays share in total expenditure increased from 5; in 1980 to 71 % in 1981. Chahges in investment 6uttlayv are associated with changes in ICT outputs (Table 23). ICT has been charged with the responsiblity of providing housinc For the urban poor. In order to target its output to the Loi income population it has tried to establish some ceilings in the income of the beneficiaries. Since 1977 the InstitLute has defined these ceilings in terms Of minimum wage rates. . Before 1977 ICT plans were measured in terms of square meters of construction and in terms ofF current pesos. As credit terms vari ed with the Value of the hoUseCq being more onerous as the ;value of the *nit in-creased, and since inflation was relativelv hi,ih it haocienen TABL E No. 24 I.C.T. EXPENDITURES 1917 - 1981 1977 S 1978 1979 X 1980 X 1981 X . Current Expen- ditures 1,484.9 34 1 31 2,0390 38 3 583.3 48 3 066. 7 29 Operating expendi- tures 498.8 617.0 801.1 1,041.t1 1,155.5 Debt Service 986.A 1134.0 1,230.7 2,542.3 1,911.2 11. Capital Expen- ditures 2 892.9 66 3 5945 69 ] 374.3 62 3,840.0 52 7,608.4 71 Direct Investment 2,856.0 3,916.3 3,333.3 3,738.2 7,257.1 Indirect Invest-_ ment 36.9 29.1 41.0 1o0.n 351.3 TOTAL 4*377.8 100 5,697.2 100 5,414.1 1on 7.,423.3 100 10,675.1 100 SOURCE: ICT. Annual Report 1977-1901% TABLE No.22a SELECTED HOUSING STATISTICS ICT - 1942 - 1981 Year Completed Improved Site and Total Total In- Total Invest Construction Dwelling Dwelling Services Housing vestment ment constant Cost Units Units Units Units current Prices. Index Prices (1970 prices) 1942-1970 162,872 - - 162,872 3,161.4 3,161.4 100,0 1971 13,997 - - 13,997 '751.1 705.9 106.4 1972 20 190 1, 200 379 21, 769 1,031.5 900.9 114.5 1973 23,887 1,129 190 25,206 1,131.6 838.2 135.0 1974 18,926 420 1,345 20' 691 1,503.0 799.5 188.0 1975 20,268 696 3,196 24,160 1, 698.6 792.6 214. 3 1976 22,282 1,609 1,194 25, 085 2,076.0 816.0 254.4 1977 22,214 8.072 697 30,983 2,892.9 929.3 311.3 1978 22,684 4,837 1,184 28,705 3,945.4 954.1 413.5 1979 15.906 1,598 - 17,504 3,374.3 628.4 537.0 1980 23,799 2,616 - 26,415 3,839.9 550.1 698.0 1981 20,998 4,163 960 26,121 7,606.9 893.5 851.3 Total 338.023 26,.340 9,145 423,508 33,012.6 11,969.6 SOURCE: ICT, Annual Reporta 1981. that as tirme pasSed credit conditions were becoming more stringer,: and the taroet popt,Alation was becoming more difficult to reach. Tc. make hoLtsing more afordable ICfT allowed the Luse of severance payments .cesantias) as additional annual contributions. Table 26 shows credit terms relevant for 1981. Income ceilings anc maximUm housing valuLes are expresed in terms of minimLtm wage rates. (Some of these ceilings were changed by the new goverment in 1982). The fraction financed by ICT varies from 45 to 1C(. depending on the program. There is no downpayment for sites and services plots, for core Lunits, and for the so called minimal solutions. For other programs providing better housing, and targeted for middle income houtseholds the down payment can be as high as 25%.. It should be noted that ICT has been financina housing without downpayment even before the present government was inaugurated. However, cLurrent plans try to expand these programs since one of the promises of the presidential campaign has been housing without downpayment. ICT credit terms in respect to loan period, interest rates and annual growth of payments varies with the type of city and the type of program (Table 27). Loan periods varies from 11 to 14 years, interest rates from 14 to 23 oercent and the rate of growth of the payments from 10 to 15 percent oer year. It shoutld be noticed that for the so called max:imal solutions II the credit terms are equLal to those used by the CAVs. TABLE No.?6 ICT. SALE CONDITIONS BY PROGRAM 1981 (Acuerdos 004/77-005/79-007/79-015/80) Program Maximun Housing Value 1./ Maxlmun Housing Income 1/ System 2/ Serviced Plot LU 400 MDW HMinimun Daily 1.5 AtilW (Mlinimun Monthly AC, CC Wages) (176,000) Wage) ($8.550) Lot with sanitary unit LTS 650 N4OW 1.5 MMW AC, CC ($123,500) ($8.550) MQinimal Sol u- tion Sm 1.500 MDW 2 141MW CIS, AC, CC, ($285,000) ($11.400) CF, PT Basic Solution SB 1 500 -2,142 NDW 2 MMW CC, CF, PT (t285, o0-$406, 980) ($11.400) intermediate 2 142 3,310 MlDWI 3 M14MW CC, CF. PT sol ution SI ($406,980-$628,900) ($17100) Maximal solu- tion I MI 3,310 - 5.780 MDW 5 m14M CC, CF, PT ($6281900-$1098,200) ($28.500) Maximal solu- tion so ml No restrictions Ho restrictions CC, CF, PT 1/ 1981 Col $; Exchange rate/81: Col$ 1,000 1 US$ 16.93 2/ AC=Supervised Autoconstruction; CC= Constracted construction; CIS= Individual supervised credit; CF= Cofinancing contract; PT- Employee cofinanced contract. SOURCE: Annual Report. TABLE No.27 FINANCING CONDITIONS BY PROGRAM. ICT. 1981 Program ICT's Share Down Loan pe Interest Annual Paynt riod - Rate growth rate A. Serviced 0'I6ts 100% 0 % 12-14* 12-16* 10-15* B. ServicedFi ots with sanita- ry unit 100X 0 % 12-14* 12-16* 10-15* C. Minimal Solution 50-lOO%** O%-1O% 12-14* 12-16* 10-15* 0. Basic Solution 50-100%** 15% 11-13* 14-18* 10-15* E. Intermediate solution .45--oo0%** 20% 15 20 15 F. Maximal solution I 50-100%** 25% 15 23 15 G. Maximal solutions 50-100%** Same as Same as Same as -Same as CAVs CAVs CAVs CAVs * Depends on City Type ** Depends on Operating System SOURCE: ICT. Annual Report. 1981 66 TABLE 28 PROFIT AND LOSS STATEMENT I.C.T. 1976 - 79 (MILLIONS OF PESOS) 1976 1977 1978 1979 (+) Interest received 649.4 804.1 1.089.4 1.255.65 (-) Interest paid 515.0 756.0 904.7 1.294.7 (+) Net commissions 8.2 14.8 20.4 19.5 + Other revenues 615.1 551.5 683.1 933.1 = Net financial revenues 757.6 614.4 888.2 915.5 *- Balaryand administrative expenses 732.6 604.5 898.3 1.168.5 Profit (loss) 25.0 9.9 (10.1) (255.0) == === * =: Interest rate on liabilities 12.04 13.69 15.30 15.10 Interest rate on assets 13.00 13.05 13.92 13e.79 Spread 0.96 (0.64) (1.38) (1.31) Budget Appropriations n-a 675.3 684. 1.159.0 Source table XV-7 D.N,P. Firnanzas intergubernamentales in Colombia P 323 67 1'here- are several observations about the e+iciency of ICT, as can be seen in Table 2B, since 1977 the rate of interest paid by ICT is higher than the interest rate in the loans gr-antei. It is importart to see that ICT borrowing rate is well below the market rate of interest. Another problem with ICT is the high fraction of expenditures devoted to administration. There have been complaints about some land purchases made by the Institute above commercial valuLe. Finally there have been complaints about cost accointing used by ICT since lai,d is carried at its historic value and consequently in a period of high inflation it leads to implicit sLtbsi dies. C. The Central Mortgage Banik (BCH) 1/ -- ------- ------- ---- --- Since its founding and up to 1972 the Central Mortgage Bank(ECH) was the only bank financing housing In49o. the bank started its operations with equity participation of private banks '.20 7) public credit instit'Ations (20%.) and the Central Ban k (6C7.)% Mortgage bonds backed by the loans should be the main soUrce o4 funds, cedLulas de capitalizacion (tong term saving contracts that could win their final face value by a lottery system) and 1i This section is based in se-teral BCH annutal reports and the Colombian Urban Sector Report (World Sank. fi983 c). 68 indLstrlal bonds to finance constrLCtion industtry sLppliers would provide additional resources. In 1949 becautse there were few investors willing to take mortgage bonds the government forced the commercial banks to invest 30% of their savings deposits in these mortgage bonds. In 196/0 to complement SCH resoutrces a. fixed percentage of the Social Security TrLust Fund were transfered to BCH BCH has lost its major role as a howising financing institution because of the creation of the UPAC system. Table 29 shows the relative share of housing financing of the three main financing institutions. Two major reasons can be given for BCH's stagnation. In the first place, the profitability of the main instrLtment (cedulas hipotecarias) declined in real terms becaLuse of accelerating inflation that occured simulltaneouLsly with the creation of the CAVY. Secondly., the government made in 1974 a major tax reForm that removed the tax exemption of thei cedcl as. In 1972 BCH formed a CAV that 1later was incorporated as a section of the bank . As can be seen in table 28 BCH has five main type of liabilities. In the first place, mortgage bonds or cedulas hipotecarias. In the early 7(C)s mortgage bonds reprsesented more than 80 percent of B:CH'iUs liabilities, in 1981 it was less than a quarter of the liabilities. FLurthermore. as can be seen in the table, there were some Years in which the level oT' this liability TAB L E 29 DISTRIBUTION OF CONSTRUCTION CREDIT 1960 - 1981 Year I.C.T. B.C.H. C.A.V.s TOTAL 1960 27.8 72.2 100.0 1961 39.3 61.9 - 100.0 192 52.9 47.1 100.0 1963 54.3 45.7 100.*0 1964 40.6 59.4 100.0 1965 32.0 68.0 100.0 1966 39.0 61,0 100.0 1967 39.9 60.1 100.0 1968 27.6 72.4 100.0 1968 20.6 79.4 100.0 1970 .19.7 80.3 . 100.0 1971 18.7 81.3 100.0 1972 31.6 65.8 2.6 1/ 100.0 1973 18.8 26.4 54.8 .100.0 1974 17.6 4.9. 77.5 .100.0 1975 27.4 11.9 60.7 100.0 1976 28.4 8.1 63.4 100.0 1977 21.5 11.3 67.2 100.0 1978 18.1 8.7 73.1 100.0 1979 18.1 5.2 76.7 100.0 1980 12.5 15.9 71.6 100.0 1981 16.4 14.3 69.3 100.0 Source: Estimated based on DNP (1981) Table 16 1/ From September/ 72 TABLE No.Z30 BANCO CENTRAL HIPOTECARO -LIABILITIES 1970- 1981 (millions Col$) Afio HipotecarSas CapttaliZaci6n UPAC Dec. 687/67 Fiduciaria Total Valor X Valor V Valor V Valor t Valor V Valor S 190 5 8. 19 563 91361.8 5.7 210.7 3.3 6,335.3 100.0 1971 6,973.5 84.9 576.3 98. 357.9 4.3 256.3 3.1 8,277.5 100.0 1971 82183.6 81.3 786.1 1.8 18.0 0.2 767.7 7.6 302.8 3.0 10,096.8 100.0 1973 7,533.3 69.0 926.5 8.5 977.0 8.9 1,Y123.2 10.3 359.6 3.3 P,919-5 100.0 1974 B,136.4 62.6 1,056.6 8.1 1,731.0 13.3 1,669.8 12.8 422.2 3.2 13,016.0 100.0 1975 8,345.0 55.7 1,219.2 B1A 2,313.0 15.9 2,570.2 11.2 463.4 3.1 14,970.8 100.0 196 8539 4. ~3.4 59 3,951.1 22..4 3s,434.3 19.5 711.0 4.0 17,650.7 100.0 1977 9,227.8 46.7 1,045.-5 5.3 3,,913.1 19.8, 4,884.1 24.7 687.3 35 1,5. 0. 1978 10,268.6 41.7 1'142.7 4.6 6,099.8 24.8 6,391.3 26.0 707.4 2.9 24,579.8 -100.0 1979 10,717.1 34.7 1, 317.3 4.-3 8,492.9 27.5 8943 2. 1,24 46 30864.0 100.0 1980 11,356.4 28.5 2. 330.9 5.8 12,865.6 32.3. 12,039.7 30.2 1,252.3 3.1 39,849 0. 1981 11,831.8 23.0 3,569.7 6.9 19,162.1 37.2 15,943.7 3o.9 1,013.2 2.0 51,520.5 100.0 SOURCE: DNP. Rgvista de Planeacit6n ,y Desarrollo. Vol. 13, No. 3, July-Dec/1981. BCi. Informe de Gerencia. 71 remained prcL1icaj.1y constant impilving that mortgage bonds did nct contribute to the sources of new loans. A second source of funds is the UPAC section. In 1981 it constitLied the main instrutment. UPAC's share was 7.7% .It should be noticed that thi.s section is the largest CAV in Colombia with more than 18% of the resoLurces of the system. With the crisis in the financial sector that. took place in 1982 there has been an increase in the deposits in the UPAC section of BCH. A third soutrce of resources corresponds to the Soci al InsLurance Trust Funds or what,.are called Decree 687/67 funds. Its share has increased from a little more than five percent in 1970 to more than 30 percent in 1981. A fourth source of funds .a contracutal saving instrument called cedulas de capitalizacion. This savings scheme can be described as follows: savers are classified in groups, of say 100, and each one commits itself to save a given monthly amount after a period of time each one receives the face value of the contract as an incentive to save there are monthly lottery drawings with the winner getting the face valele inmmediately. As a further. incentive to save at the end of the contractual period the Bank can graht a loan at concessionarv terms. Since the implicit interest rate of this instruiment has not kept with mariket rates its share has declined from almost nine percent to seven percent. 7Z2 Funds come also 5 rom the trutst section tJidUciaria) whose shar. has been almost three percent of the total. In 1961 BCH developea a new instrLument called cedulas de inversion to finance ol= housing. lt works as follows when buyer and sellers of old hoLusing aet together and agrees to a price they go to the bank ano BCH issues a bond for 70 percent of the value to the seller anc grants a loan to the buyer by the same amount. During the first two years the seller gets interest at below market ratec and after that he receives both a return and gets repayment of the principal. The seller can sell the bond in the stock e-change at a discount (the current 198; discount is 30(%). The bLtyer can buLy bonds in the stock exchange to repay the loan. Since the interest in the bond is below the markcet rate of interest and there is a discount in the bonds , the bond has not succeeded. Other source of funds for bCH has come form its inventory of land. BCH still owns some parcels of land that were acquired beFore 1960 when BCH carried out some constrLuction work in its own account. It also owns some land that ICT has sold in order to generate funds for its construction program. In 1981, B(CH financed aboLut 17% or 10,300 of the formal sector units, with about 187. of the sector 's financial resources (col $8 . million), including those oF its own saving and houtsina corporation. BCH's mortgages have been principally a-+-ordable to the top two to three deciles of the mrban income distribution and benefit from ffiscal interest rate dedLuctions as those of the Corporaciones de fhorro y Yivienda (CAgs) mIiportant exceptians t o 73 LDCH' s lending have been the si te and ser /ices and core hoLtsirn: programs in Cali and elseewhere financed throuLgh its infrastructure fund. lThe Fondo Financiero represented, howeverq a small thougt growing proportion of all BCH credit. BCH offers two kind of programs, one that is called the traditional program which offers fixed interest rates and the indexed rate or UFPAC plan In the traditional plan, it lends to developers at 28% for six months beyond the construction period. It also lends to individuals buying hoLuses below5,000 UPACs (Colt 3.4 million in 1983) the interest rate is 26%. In order to adjust to the rate of inflation BCH has changed the lending rate from 11% in 1971 to 18% in 1974 to the current level of 26.. In the UP.4C plan the credit terms are the same as those used by the CAWs.. In recent years BCH has adopted several innovations.It has started a plan that allows for the pre-selling of housing, and other for the financing of used housing. Since the pre-seiling demands les- resoLurces from the bank and redLuces the uncertainty of the develioper consequently redLucing his cost, it has been rapidly adopted . As can be seen in Table 31, two thirds of the approvals in 1982 cor-responded to the pre-selling plan. 74 TAaLE .No.31 BEC. HAPPROBALS AND DISBURSEMENTS, 1982 (Millions Cot $) -ogotA Country Total Short Term $207957 Traditional $ 276 $1, 334 UPAC $ 620 $2,43 Pre-selling $1,899 $4.,530 Long Term $1 91 $5 312 Traditional $1,509 $3, e9 .UPAC $ 392 $1,23 TOTAL $ 4,696 $14,119 DISBURSEMENTS S.C.H. 1982 Traditional $1,758 $5,113 UPAC $1,011 $3,040 Pre-selling $ 389 $1,725 TOTAL $3,158 $9,878 Source: CAMACOL Informe del Departamento de Investigaciones Econ6micas, Girardot, April 1983 75 L. HoL1s.in.g: ano Saving Corooratiorns CUAVs) I/ 'The third institLution in charge of housing finance in Colombia ic the CAV or Corporaciones de Ahorro y Vivienda (Savings and HoUsinq Corporations).' At the present time the CAVs are financing more than a third (21,000 ) of the formal sector units uising more thar 70 percent of the sector reSoLurces. The Colombian System o- hotsing and saving started in 1972, its nominal assets have growr at a rate of more than 150J% annually. In more than ten years o+ existence it has accomplished many things. For example, in the first 41tll year of activities it financed more than 30q000 LInitc (table 30).. in the first nine years it financed 180.,00 units. The system financed 51% of the value of Lurban bLuildings and 62% of the area of urban .building in the first year and is -Financing 27% of tha value and 45 of the area (table 31). The Colombian System of Saving and Housing has attracted a lot of interest. Part of the interest that the system has attracted is due to the use of indexed instruments that has allowed tFor a rapid growth in an inflationary environment. Part of the interest has been associated with development policy since the zystem of savinc and hoLusing has been adopted as a part of the 1971 development --------------- I/ The Section on the Colombian Saving and Housing System is based on Sisten-ras de Informacion (1982); Carrizosa et al. (1982): Villate (1981). The first years b4 the system ha>ve been StLtdied by Sandi1Lands (1980). 76 TABL E 32 HOUSING UNITS FINANCED BY THE. CAVs,COLOMBIA ,1973- 1980, Year .urirg Cumulative the year total 1973 32,176 32 ,176 1974 22,,083 54 ,259 1975 2,001 560260 1976 139581 69 841 1977 21, 134 90,975 1978 21,762 112,732 1979 21,200 133 ;937 1980 23, 305 157 ,242 1981 21,000 178,200 Source: ICAVI Experiencia Colombiana, Table 26 p 230 TABL E 33 UPAC FI NANCI NG COLOMBIA 1973 - 1981 Year Loans ' Square Building Urban UPACs Share approverd. , meters value building share in (millons 1970 finance (millonsl970 sq m invalue sq m Col. $) (000) coI $) (%) (1) (2) (3) (4) (5) (6) 1973 3,359.9 4,022 6,550.4 6,461.9 51.3 62.2 1974 2,350.0 2,760 7,155..4 7,052.7 32.8 39.1 1975 250.0 793 4,878.3 4,839.6 5.1 16.4 1976 1 520.3 1, 698 5, 262.1 5 173.1 28.9 32.5 1977 2,065.4 2,641 6,518.1 6,436.7 31.7 41.0 1978 1,910.4 2,720 7,067.3 7,243.5 27.0 37.6 1979 1,586.0 2,226 6,084.8 6,093.7 26.0 36.5 1980 1,633.1 2,581 6,003.8 5,692.7 27.2 45.3 1981 2,372.8 2, 561 Source: Carrizosa et. al. Anralisis Economico Table 0XIII, p 168 plan Ohe Fso-tr Strateciies) that choose constrLction as one of the leading sectQrs. ihis section of the report presents a brief description of tne system. Linder the following headings: i ndexation, terms and conditions of the loans4, stLtcture of the assets and liabilities of the system, relationship of the system with the Central Bank. and allocation 'of credit. In the following chapter some o+ the main issLues related to the system will be covered. 1 I nde- ati Ln One of the main charateristic of the Colombian system is the use- of indexration for both deposits and loans. According to its proponents a major obstacle in the development of th.e construcion sector was a lack of funds for its financing. It was thoLught that savings would flow to finance construction if the real rate of in terest were positivet. since in an inflationary environment fixed rates tend to lag behind the inflation rate it was thought that a solution will be to have indexed rates of interest. To index the rate of interest the Consumer Price Index was chosen. Savings deposits will oay a real. rate of interest plus a monetary correction'reflecting the change in the CPI. The correction will be made d>+ailv while the intersEt wjill be c:alculated ULuarterlv1 Saving deposits will get a return o4 5 ¾4 anco CODs 5.5 ¾. T-he loans made by the system will ca rry a rate ot interest o+ 7.5Y. pl.us monetarv correction when made to individuals and 87. pIUS monetary correctidrn when made to developers. The refer. - ,-e period for th,e index:ation was the previous three months. In accordance with the ideas of the system, monetary correction was granted exemption for income tax purposes. During the last ten years the system has been subjected to several changes. The reference period for the definition of th6 indexation has changed from 3 to 24 months. In i974 a ceilinan was established $or the rnOnetary correction. Since the ceiling vwas beloow the inflation rate in practice.the UFAC system was transformed into a fi xed rate system. The tax: reform of 1974 also changed the ta. statLus of the monetary correction limiting its exemption to the first 8 points. The real return has also changed in the last ten years it has been reduced to 4% and raised again to the current rate of 5% for deposits. 2. Terms and conditions One of the principal reasons for the success of the savings and housing System w5as the innovative character of the terms of the loans granted by the CAVs. EBecause of the indexation of the principal it was possible to offer graduated payment plans that considerably improved access to mortgage credit. F3ecause of the flexibility of the plans offered by the CAVs, it was pos=-.ble for 80 thr. potential bUE;er- to choose a plan that best sLlited hic circLumstances. In some ot these plans the first payment was less than hal+ the correspondinq one in a level payment plan aria consequently it was very attractive to someone with a cLurrently l ow income butt with possibilities of fLiture increase. At the beginning of the system the loan to valLue ratio was 70.% for all loans. Later., the ratio was increased for low income hoLusing and decreased for high income housing. Currently, the loan to value ratio varies from 1007. for popultar housing to 707. for Lunits whose va.ue is greater than 5,ot3 UPAC (Col$ .4 million). Since- the maximLum loan is 7OO)O UPAC . loan to value ratios are a declining function of loan size for units whose valutes are above 10000 UPAC (Col* 6.7 million). The standard payback period in the UPAC system is 15 years. When the system started all loans were f,or 15 years., buit recently th e CrAWVs have tried to shorten the terms of the loans. Some of the Corporations have redLuced the maximum payback period to 12 years. and are promoting even shorter maturities. At the beginning of the system, all loans to individLual's carriied a 7.5%. real rate of interest. In 1974 the government established a ceiling in the monetary dorrectjon and as a conseqtence the real rate of interest became variable4 decreasing in per-iods of accelerating inflation and increais-ina in oeriods of 'deceleratior, in intlatiorin As 6r Linintetnded consequence of thi.s refform thE,- demand tor hcLuing became procyclical, raising in periods o± hiori acgregate demand arnd declining in recessionary periods. Currently, interest rates varies with the size 0+ the loan ! from 6%. for Ltnits below 1,3OO UP'AC (col$ .9 Millions) to 11i for Lunits above 10.:)C, UPfC (Col$ 6.7 million). Two reasons have been given for the change ir, interest rate policy. First, with different rates it becomes possible to cross-subsidize low income buyers with the higher rates charged to high income buyers. Second, the differential in interest rates can compensate the implicit sLubsidy due to the income tax: dedckction of interest payments. The return offered to savers by the Savi ng and HoLtsi nq Corporations varies with the type.of liability. For UPAC deposits the CAVs off ers a real return of 5. with monetary correction, and for CDs the retLurn is freely neaotiated. For ordinary deposits a nominal return of 199% is paid to savers. At the time that the UPAC system began, interest rates in other financial assets were very low and negative in real terms in most of the cases. Since the return of the system was positive in real terms a large gap existed between the average market rate and the rate of return on indexed instruments. Graph 4 sho;ws the ratio between indexed and market deposit rates. As can be seen from the graph., in 1973 the UFPC deposit rate was twice the average. by I'476 the two rates have become equal, whil e the UFPC rite has beer-, - - - - - N r N N N La C 0 v 4 0 CD 0 - N U 4 E * - 4 0 N U 4 U 0 - W 0 0 GRAH a 1'' . t'1 - weak. They ToUnd 9' +ol lowincn a model dev elope d b%y k rceLIs and Hteltter t 1t97'$* tt tr.- main determinants o0 constrLLCtlon are permanent income ar;t interest rates. .They concltutde that the maLn e+4fect of in-creasir.- mortgage tinance is to decrease ±inancing from other alterniati,t- soLurces. Even thoutgh it is not the purpose of this document to discuss th- econometric evidence that support Carrizosa's claim. it car reference other resuilts, such as the BCH sponsored research. (Sistemas de Inforrdacion, 1982) that concluded just the opposite i. e. that the increase in financing resulting fromf, the cr-eatior: of the UPhC sy4tem increased the qLuantity of housing. It can be said that in principle what it has been in discussion is the *Va1lU;e of the elasticity of the sLupply of new hoLtsing -and thaat to obtain reliable estimates symultaneous equation methods should be used. Even if Carrizosa's evidence is accepted as a prooT that an increase in housing finance does not contribute to an increase ir, housina it is possible to argue that it is worthwhile to increase the financing and that it contributes to an increase in wealTdare. It only needs for its validity the existence of secimented credit mar kets. If an increase in hous.ing finance leads tio lower costs. for the infrormal sector it seems that this would lead to ar. imnro,ament in welfare. Financial innovations such as mori-ty.aae i nsLrance s.nd sesondarv mortcace markets coLld. ihave positive, impacts in the housink m.-ark-tret. Another important issue is how affordable are the di+T'erent uinits oF+ered by different inStitLutions. UnFortUnately there is nc evidence at the country level of the degree in which the sLupply.- agrees with the possibilities of the popuLlation. Some informatior ex i st for specific cities. For example. Strassmanr's work iin Cartagena and the City Study work: in Bogota. Despite the concentration of shared housing among the poorest (the average 1.4 household per dwelling in Bogota may increase to 1.77 in poor neighborhoods),, the real demand For new, owner occupied lower-income housing may be redLiced by the relative attractiveness of the sub-renting markets, by the limits on payment capacity, and by the willingness to pay the renters. An analysys of the income distribution shows that half of Bo0ota's population couild a-fford tcx purchase a housing unit of more than 1,000 UPACs. At present., about 64% of the 1982 portafolio of ICT is composed of units of less than 1,000) UFACs, at its present subsidized rates. ExclLidinc betterment lcans, which almost dwindled to. nothing, ICT produced an avergae of- no more than 16,000 sLtch Lunits ppr year between 197-:. and 1981. By contrast, only a nominal portion of the activitv o+ the other DLublic institLutions has been targeted towards the ocorest 5 deciles.' Overall, of the 60,0Q0 Lunits provided in 1 931 by the +orma± sector, only 16i was X Ii, rted at the poorest hal ± c, 1^-1t t- :C. L .. c1 ti cm, r ti ie abv, :I ',out 2 . was iea, r-ed to the ri c he- . hOUseChbi dc. The mr:ain di++icultv for this type of e-stimates is inTormatic- aboLut income and aCtLual rental payments. To caiv e atn idea o+ the affordability of some of the programs, an approximatiorn *o+ bot, utrban income distribution and rental by income was obtained fro. infc-mation related to Bogota. Fortunately, it was iound th,:- Bogota according to the last national household Survey made by DANE in 1979 had a very similar income distribution to the whol- urban area. Since the most reliable estimate for Bogota refers t: 1978 it was necessary to estimate th*p actuLal level of income. UsinQ the i931 credit terms for ICTI the 199S; minimuLm wage and the estimated income and rent distribution it was possible to derive the results that appears in Table 39.. The popula - programs such a:- site and services plots. core units, minimal soluttions and basic solucions are affordable by the lowest half of the population. Th; two more esxpensive programs are ornly affordable by the top 2u r Cz the population. A similar exercise made for CAVs plans showed that even Lunder the most favorable conditions. on interest rate. graduated payments plans, and payback period housing can only be affordable byv amilies above the 6b peraentile. Table 40 shows the effect of variations in payback period and thG vaILe o, the hat-tse in re;iative a±f±ordab litv. It shos that i. ti-l TABLE39- Affordability of ICT prograss by type of program. (1981 credit conditions for largest cities) HOUSING DOWN LOAN INTEREST ANNUAL FIRST INCOME PROGRM VALUE PAYMENT PERIOD RATE GROWTH PAYMENT PERCENTILE Serviced plots 123480- 0.0% 14 16.01 15.0Z 841 3 Core Units 200655 0.0% 14 16.0% 15.0% 1,367 5 Minimal solutions 463050 0.0% 14 16.0 15.0X 3,155 21 Basic solutions 661,235 15.0% 13 18.0% 15.0% 4,573 40 Intersediate solutions 1,021,797 20.0% 15 20.0% 15.0% 6,630 *66 Maximal solutions 1 1,784,286 25.0% 15 23.0% 15.0% 12,748 86 Maxiual solutions II 1,784,286 25.0% 15 30.0% 15.0% 17,586 90 Estimated with the 1985 minimum wage of colS 9261 per month SOURCE: Table 24 and 25 100 TABLE 40 RELATIVE AFFORDABILITY BY PAYBACK-PERIOD AND VALUE OF THE UPAC SYSTEM Value of house Y E Au R S (UPAC) 5 8 10 12 15 O - 1,300 231 156 132 116 100 1,300 - 2,800 151 129 114 100 2,800 - 5,000 218 151 129 114 100 5,000 -10, 000 214 149 128 114 100 10,000-15,000 195 140 122 111 100 Notes: Monthly payments increase at an annual rate of growth of 23/c Source: Appendix Tables 1 0- 1 pa,yback per1 Cal i r; redLkcCd from 15 to 5 y ear s the *i.; LLIe c. tt-' first paymeTrit in z, gQ-adLuated payvmert plan dOLUblet. The ef+ect i- 1 arger the owei- the Val Lie of the hoLise. lrnterest ra-tes a.--- directl- related to house valLues). It car, then be sai-d that the tendency o'r the CAVs to decrease the ja ybacck: period wi 11 have a negative effect in affordability and that this effect will be stronger the lower the incofme. 'D. Rental lMarket for Hous Inc 1/ Another important -aspect in the definition of the housing problem relates to the functioning of the rental markets. Rereital hous nc already shelters 40Y% of the urban polpulation and more in lower income deciles. In spite its importance, Colombia has a history o-f several decades of different forms of rent control, freeting rents in particLular structLlres or allowing limited adjustements.. Although rent control mechanisms are justified on eqcuity gruLinds, they are unlikely to work in a market where vacancy rares are low-. The poor, in particular, are unlikely to exercise their leg.-i rights under rent control, especially 'in the informal hoLusig. martets. Even among the more probable middle incom.e beneficiarie-e of rent control, evasion is likel'y to be a rule. However, Since January of 19S3% some measures of the Government tends to redLuc / 'This and the following two sections are based in the Colombian rirb:n Sector Report (Wosld Bank: IJ,:3 C! . I oz cc.ntrcI c':st trWe rentol mar.e-t. ;411 uni ts t.nit are 3nanced bv the Ci*)s are +ree of rent corntrol. Other units ha,;e their rents Linke: to the catastral ValLie the mMaXimUr, mornthly rent is 1.3%," of thF:: catastr ti value. Since the catastral v-ales USare s-ometimes wel. below the market value rents are in some sense controlled - an: produLce SmMe distortions. E. Fublic Sector Housing 8Clnci es As we have seen before, there are several positive aspects of the various roles that agencies sutch as ICT, the Ca!as and F,NA ir. housing prodLuction that shouild be preserved and even strengthened. There are also several institutional issues related to the waS. these agencies have evolved whi ch may pose constraints, i r. reaching the increased targets and in fulfilling the prominent role assigned to +ini for low-income families in the emerginc shelter plans. Overall financial viability and financial managment of the public sector agencies have increasingly come into cuestion during the past several years because4 in the context of increasing f scal constraintsP they have continued to reLy on significant transfers +rom national and mu;nicipal treasuries. In 198'., these aimounted tc about a qumarter of ICT's aross re venues. and a greaaer proportior o+ those of some murnicipal Caias. [he reasons for its dependenc. on larce scaie SUbsidies are not r-ecent and the re-asons relat4e tc d Cteri.' cr- a t i n-tnci al managment o+ public sector et qonC * ieM Linreliable cost accouLntinrg; high arrearage rate, and poc- collection records; negative interest rates: resist ance t¢- irn*lation-corrected instr-utments (sLIch as UFPAC system irs the private sector); large administrative overheads4 LunecononmiCral larnd purchases; .-and issLues of noncompetitive biddicng becaLuse of butreautcratic procedures. and delayed payments to pr i ate cont ractors. The issue is whether> in the medium term. puiblic housinc aoer,ciee at the national and local levels can be e:-oected to improve their financial management and become self-sustaining without a basic transformation of their legal operational strLuctures. The overall efficiency of public sector agencies as icr and the Cajas has come into question in the recent ppst becaus5 o+f thei.r broad objectives and possibly conflicting institutional mandatee in relation to those of other public and private sector hotsing agencies. They are, simultaneously4 policXy, financial 4 design. ano constrLuction agencies for land development hCsUingr often dwelegating few of these mul3.tiple functions. AlthoLugh in the case of design and constrL1Ction. ICT and the Cajas have begUn tc relr more on private firms and bLtilders, mnLuch o+ ICT's decision r,incr, continLtes to be centralized in the -ocota o+fices rather th-an relying on its regional offices or on locaLlly accoutntable interm,ediares, as. irn the case of+ INSFLfPL or E9CH. In terms o-.- iLts 104 d c.i qrn cFnL ccrnstrlcti on tLinccti ons. one opti on couLd inv -! converting l;CT' s r egiorl4 J. olt+ces into autonromuts, loc- .i accoLuntable hoLusina ror por at i onr, equli val ent to the wat-- Empresas), or transitormation into Cajas +or munici.pal housing urban development insti tutes where they e:::ist. PI possible solLution +or streamlining their fUnctions WOLlld be t- have them concentrate exclusively on servicS land and development. and/or housing uLpgradiing. This streamlining w4ouIld avoid, in the case of IiCT the present overl.apping o target poputlations wJitr BCH and CAWs. And, as an extension o+ their land development role. public housingsagencies could promote large scale "land pooling . s&hemes jointly with tne private sector. Thereby helping increase the supply of serviced urban land +or the augmented sheiter^ targets proposed in the new development plan,. SuLch lanc de>velopment ftunctions could be complemented by an extension o5; their housing improvement credit and core hoLusing programs. F. The Public and Private Mort age Interest Rates in aggregate, and especially in providing lower cost housing, the CAVs (includina CAV o+ the EDCH) have been assigned a crucial roe. in meeting shelter plan targets. Accor4ing to the proposed shelter targets. ICT is sLu;osSed to prodLuce aproximately UO. 800 lnits vearly; the CAVs are to provide 40,)20) Lunits and BCFN is assicned - t Oaet ! 18."0) Units. ,i,he renainia 6,fiO- be-inco co vered bt t h e Caia'? ano1 the I-NA. P;s we hztve seen above4 The CAVs and EClH arre reqLtired to compets with other financial instruments irn the capital mari::ets, inclLucin: those of the TreasUry4 for f+unds wiith which to carry ou-t their activities. Under these circuLmStances. loans must be mrade at rates that cover costs of money, inclLding a retLtrn to orivats stockholders or depositors. With the cost of fUnds rurnning a* around 30 percent per annum, loans cannot be made at rates beloo: that, un+vrtunately, the main supplier of I Ow val Lie housi no i intended to be the pLublic sector agencies of ICT, Cajas arid FNA. which operate with little regard for cost recovery in real terms and which place most of the pLublic sector Lunits on the market with mortgages set about 18 percent per anrnUm. This one-si dec compettion was avoided in the past ' by/ the CAVs:they did r,ov completely comply with legal requLirements that mandated ever increasing portions of their portafolio be made up of lower value properties, and they rationed loans throtgh the Lse of larqe dowr payment provisions and other administrative means. Thi s wl 1 t become increasingly difficult Lunder the new shelter strategy anc the new legislation (Decree 1325/83), which sets new value tai-cets for the CA Vs. The CAVs now face the possibility o+ havina to tie LuP significant portions o0 their funds in mortgage of-ferings that COLld possibl%y find no takers, as the pc:tol ofT potential cLustomers sirimnpl-y quLeLtue Lt- Lo LIvit tor the availabiliity 0± lLI mort+aoe ^uncis; wh2:- presentlv have neaative real interest rates. LeCaLise th:- discrepancy can be wastetuil and counterprodUc:tlvee a poiicv optic- for eliminating utnequtal competition among financial institutiOn.- tor similarly priced uni ts could be considered. Either all 1c.. val uLe hoLuSi ng shoul d be offered at rates suLbsi di - ed by tne Treasury, which woulld have an even areater impact on t;he fisc, . si tLation. or all such mortoapes shoLld carry r-ates higqr, enoLlgh t: permit appropriate cost recovery, using alternative means +c- channelling sLubsidies, if this were desired. G. Subsidies A final issue that can be discussed is about subisdies. (is i t wa= mentioned before. the 'new .government established differer , interest rates for different house values. 'One of the mai,- arOLments for doing this was that it will compensate tht-e reoressive nature of the implicit subsidy caused by the deducti:n of interest payments for income ta.: pLtrposts. As it is evident From what we have seen in the previOLS chapters interest rates for most of the government programs are t ll bel c.- the market rate of interest or even what is charced to Ci84*r,s cuistomers. As conSequence of tnese low int erest rates qovernmier-, program beneficiaries are receiving also an implicit subsidy sf nc- the present valt.te of -LttLre payiments descoUnted at the market -a: o+ -interes is J ower than, the Y/alLte of the loion. I-or exz?,mple lev/el payrffent plan at 16Z +cr' 15 years W11l carry an irr,pliiCi- subsidy c+ 5o percent when the market rate o± interest is equLal t- 35Y% (If the interest rate parity theory is applied and a +oreiar rate of 1( /. i 5 assLmed wJi th a 25/. devaluati on one obtain= approx-imately a domestic rate of 357. In order to compare SLubsi di es or l ow income and high income households it is necessary to consider both the interest and the ta>: subsi di es. For l ow i ncome househol ds assu-mi ng that they do not pay taxes total subsidy is equal to the interest rate subsidy. For high income houtsehords they get the tax SLubsidy and if the price paid reflects the real pri ce they will get the.interes rate subsidy.. It coutl d wel l happen that low interest rates will be reflected in larger nominal housing values as it is the. CadE uLnder free market conditions (The use of points in the U.S. couLld be an example of this). Table 41 shows the implicit subsidies for selected conditions of the CAVs plans. It can be seen that even with the di+ffrerentizal interest rates total sutbsidy is reg-essive since the taxt^ subsidy is more imprortant than the interest rate SLubsidy. It can be observed that reductionrs in the payback period redLuce both the interest suLbsidy5 that is probaby what the lenders want, and the tax suLbsidy that is what the TreasLtry wants. Reiuctions in the rate or growth o4- the payments have a clear e-f+ect i n redLci nq tte 108 TABLE 41 IMPLICIT SUBSIDIES IN THE UPAC SYSTEM BY LOAN PERIOD AND RATE OF GROWTH OF THE PAYMENT o- -----o - - a-------- intrest growth Loan marginal interest tax net I rate rate aicuat tax rate subsidy subsidy price subsidy 6.01 0.01 978,969 201 85,735 129,131 664,103 24.4? 6.0? 15.01 879,969 20? 111,488 135,09% 632,395 29.1% 6.01 23.01 878,969 201 124,559 129,646 624,764 2a;9z 7.5% 0.01 2,704,520 45? 158,269 9479656 1,599,596 40.9? 7.51 15.01 2,704,520 451 209,005 995,3l9 1,501,176 44.51 7.5? 23.0a 2,704,520 45t 233,373 92,345 1,509,902 44.2? 9001 0.01 4,732,910 491 216,57B 1,940,876 2,675,456 43.51 9,0? 15.0Z 4,732,910 491 281,892 1,934,615 2,516,403 46.8? B.01 23.01 4,732,910 491 321,175 1,876,103 2,=,631 46.4Z 11.01 0.0o 7,099,365 561 -229,238 3,502,463 3,926,140 46.1? 11.01 15.01 7099,365 561. -309, 532 3,715,200 3,693,697 48.0? 11.01 23.0L 7,099,365 562 -347,372 3,656,480 3,790,257 46.6? NOTE: Market rate of interst 351 loan period lyears)* 10 interest growth Loan marginal interest tax net I rate rate amount tax rate subsidy subsidy price subsidy 6.01 0.0? 878,969 201 97,073 146,085 635811 27.7? 6.01 15.01 879,969 201 141,190 140,461 597,328 32.01 6.01 23.0Z 879,969 20Z 16U,041 131,856 579,072 34.1? 7.5? 0.01 2,704,520 45? 179,337 1,070,381 1,454,802 46.2? 7.51 15.01 2,704,520 45? 262,744 1,049,609 1,393,167 48.5? 7.51 230Z 2,704,520 451 316.889 11Oo,?001 1,386,730 49.7? 9.01 0.0? 4,732,910 49? 243,603 2,075,154 2,414,153 49.0? 8.0? 15.0? 4,732,910 49? 359,039 2,047,232 2,326.640 50.8? 8.0? 23.01 4,732,910 491 436,185 1,964,694 2,332,031 50.7? 11.0? 0.0? 1,099,365 56? -258,346 3,928,533 3,429,178 51.Z7 11.0? 15.01 7,099,365 56? -383,011 4,022,955 3,459,421 51.3? 11.0? 23.01 7,099,365 56? -479,710 3.977,672 3,600,403 49.3? NOTE: Market rite of interest 35? Loarn pnrriod (years) = 1 109 interest rate subsidy while trney have an ambigous ef:fect in th_- tax subsidy. If the 50%. figure mentioned above for the subsidy for.-ICT's plans. is combined with the figures in t'able 39 it is possible tc conclude that middle income cla'sses are getting the worst part. since both poor and rich are getting almost hal-f of their houses for free A related issue about the di++erential rate of interest has been mentioned above. Who is really getting the interest subsidies?. And also, who is paying.-for this subsidy? Another important issue related with differential in interest rates is tne real possibilities of cross subsidization. h4ill the hi.gher interest rate paid by the richer households. benefit middle income families .or only improve the lending institutions performance. Since we have seen that there are only 1C) CAVs and that they already are exercising their monopoly power it is very likely that the reform will end up helping them. EI BIOGF:APH ArrcelLis, F. and lMielt-erq f.. (197-.) . "The Markets for HOLISlrn an: Housing Services". Journal of Money, Credit and Banking. Februzary. Arango, S., et al., (1983). El Sector Financiero Colombiano.1MIimeo. Asociacion Bancariaq (1980). Sistema Financiero y Politicas Antiin- flacionarias. BoQota: Asociacion Bancaria. B.C.H.q(1981). Informes de Gerencia. Bogota: B.IC.H. Caballero. C. et al ,(1979).El Sector Financiero en los Anos Ochernta. CAMACOL4 (1983). Informe del Departamento de Investigaciones Eco- nomicas. Bogota: Mii eo. D.N.P. (1981) "La Edificacion de la Vivienda en Colombia." Revista de Planeacion y Desarrollo. Volumen XIII No. ; Bogota: DNP. D.N.P. (1982) "Dinamica Demografica y Provecciones de FoblaciBn" Revista de Planeacion y Desarrollo. Volumen XIV No. 10 Bogota: DNP., I.C.T. (1982) Informes de Gerencia '. Bogota: I.C.T. IngramS Gregory (1982) "Demanda de Vivienda en las Metropolis, en Vias de Desarrol.lo: Estimativos para Bogota y Cali, Colombia" Revi.sta de la Camara de Comercio Nos- 45 - 46 Bogota. Renaud, Bertrand ('1982) Housing and Financial Institutions in Developing Countries. Urban Development Departmnent, The World Bank. Sandilands, Roger J. (1,980). Monetary Correction and Housing Finance in Colombia. Brazil and Chile. England: Gower Company Limited. Sistemas de Informacion (1982). El Sistema Colombiano de Ahorro y Vivienda. Bogota: B.C.H. Strassmarn, .Paul W. (1982). The Transformation of Urban Housing. Baltimore: Johns Hopkins. VJillate, Eduardo. (1981) Experiencia Colombiana en la Aplicacion del Sistema de Reajuste Monetaria. Bogota: ICAV World Bank, (1983 a). The Colombian Investment Banking System and Related Financial Sector Issues. Mimeo. wor1d Bank'. (1 YB' b) Colombian Economic Development and PolicX Report. Mimeo. WSorld Bank. (19633 c). Urban Policy in Colombia Selected Issues and Some Directions f or Change-. Ilimeo. 4. ,} 0