-;'
CATATLOG No. 68                                  This paper is prepared fw
staff use And is not for
publicatioT. The views
espressed are those of the
author and not nwcesaarily
those of t1e Bank.
INTERNATIONAL BN FOR 1ECONSTRUC'IOi ANM DEVELO1PNT
Economics Department Working Piper No. 15
TM COFFE ECONOMY OF COMOMBIA
May 21, 1968
This paper is a contribution to and will form part of
the World Coffee Study, which is being sponsored by
the IED, the Food and Agriculture Organization of the
United Nations, and the International Coffee Organizat.
tion, and is being carried out under the direction of
Dr. Gerda Blau, Director for Special Studies of FAO.
Preared bys   George Kalmanoff



This report is based on the findi-ngs of a MisSion to
Colombia in September/October 1967, sponsored by the
Tripartite (IBD/ICO,/FAO) Coffee Study. The task of
the mission was greatly facilitated by the iniraluable
cooperation of the National Federation of Coffee Growers
of Colombia.  The assignments on the mission wekre as
follows:
George Kalmanoff (IBD) - Chief
Onno van Teuten (FA) - Agricultural Economist
Roberto Egli (FAO) - Agronomist
Keith B. Griffin (Consultant) - General Economist
Ralph A. de Rosayro (FAO) - Porestry Specialist
,



TABLE OF CONTENTS
.a                                  .
IPage No'.
SUMMARYAND CONCLUSIONS . . .g....          . .....      i   xi
I COLOMBIA IN WORLD COFFE TRADE . . . .....               1
Share in World Coffee E.xport:3.e...c.....e.*     .1
Destination of aports 3...e            * *  . *     3
Competitive Factors    . . . . . . . . . . .       10
I.ET.  COFiFEE ANiD THE ECONOMIC DEVILO PWNT  F dOLOIMBIA  1?
Structure and Growth of DomesLic PtodVict . .      17
Agriculture            .........           ec      20
Manufactuiring   . . .  .  * . . ..   i .e .       26
Population Growth and Employ,ment  . . . . . .     28
Coffee and Government, Reveruies,  . .  . .        30
Coffee and the External Sector   . . . . . . .     31
III.  STRUCTURIE A10D PRiOSPECTS OF TH13 COLOMBIA3i
COFFEE SECTOR     . . . . . . .   , 31 D~ 1-      3 e4
Trends of Overall Productiorn, Planted Area7
and Tree Numbers                                 34. e  cecoc e c  34
Reg.ional Distribution of Corfee Cultivation .     41
Conditions of Coffee Cultivwition  . . . .  .      . 46
The NationaRl Federation of Coffee Growers
and the M.arketing Structure       .......       53
Inputs, Costs, and Returns of CofCee
Cuiltivation  . . . . . .      . . . .           59
For7mation and Trend of Internal Coffee
Prices     . . . e                         .     65
The Coffee Adjustment Problem    . . . . . . .     75
IV. THE CO^6EE DIVERSIFICATION PROBLEM    . . . . . .     91
The Caldas Diversification Pi ogrn   . . . . .     91
Plans for Future Coffee Diversification    e .    101
Problerns of Coffee Diversification  . . .   .    106
Requirenents of a Diversification Strategy   .    116
Annex - Forestry in Colombia   .   . e  . . . .   122



No.                                                  P'age No.
1      Colombian and World Coffee Expor s, S
1930-1966, 5-Year Averages      * . . .  * e o
2      Colombian Coffee Exp-orts, Major Rtegional
Distribution, 5-Year Averages 194.6/47-
1965/66, and 1966/67       4** .* *          4
3      Dizitribution of Colombian CoCCefi Exports
'ty S,ecified Markets, and Share of the
Markets in World Trade    . . . . . e . * *     6
.4     Exports of Colombian (Coffee to Annex A and
Annex B Markets by Trade P'ractice,
1960/6l-1965/66     ............               8
5      Colombian Coffee Exports under Bilateral
Paymaents Agreements, 1960.-1.966     *         9
6      U.S. Inport Unit Values of Coffee, by
Main Types, 1953-1966   . a . .e. *. .  e a  I 12
7      Indicator Price Ranges for Selective Partial
Quota Adjustmients uinder InternLltional
Coffee Agreement, 1966/67 and 1967/68 . .      13
8      Average Annual Growth Rates and Ca,osition
of the Gross Domestic Product by Sec tor of
Origin, 1950-1966   .....       ..a            18
9      Growth of Agricultural Pl"oduction, 1958-1965.   21
1.0     The Use of Agricultural Land, 1960  ,   . *      2
1LI     Growth of Major Cities, 1951-1964   . . . * .    29
12      Cof fee Production, 1930/31-1966/67,
Selected Years      . , . . . . 0 . .   . . .  36
13      Area Planted to Coffee, 1955/56-1966/67,
Selected Years      . I . . . * . . s D *      37
lb      Age Composition of Coffee Trees in 1962/63  .    39
15      Regional Pattern of Coffee Prodtuction  . . .    43



TEXT TABLES (Cont'd.)
No.
16      Coffee Yields by Departments, 1955/56 and
1965  .  *  .  .  e  .  .  e  .  e  .  .  .  .  .  .  .  .  45
17      Co f fe Harvest Calndar   e00    * e . 000*        47
18      Distribution of, Coffea, Frms by Size,
National Totals     . .. ,     .... *   ,   .   50
19      Department of Caldas - Nuber and Size of
Coffee Farms and Area Planted to Coffee,
1965  *  e  .  .  .  .  e  e  -   .  *  .  *  *  * .  .  52
20      Proportion of CoffGe Exports by Federation
and by Private Traders, 1959-1966   e . . .     57
21      Inputs and Returns for Coffee Cultivation,
by Varying Yields per' Hectare, September
1.967   A     .  . c   .  .  .   *   * e   *  *  .  .  .  .  61
22      Coffee Baawice Sheet, 1959/60-196/67    . e   .   77
23      Colonbian Coffee Production and EVports,
and World Coffee' Exports   .  *    0* .    * .  79
2L4     Colombian Coffee Exports and Quotas under
the. International Colffee Agreement   . . .    80
25      Projected Sutrp1uses of ColombLan Coffee,
1967/68-1972/7.3  .   . . .0 *  0 0 o  It .. . .  83
26      Caldas Five-Year Regional Development Program
1964-;968, Distribuition and Prgress of
Investhdents   .            *   *      **   0   914
27      Caldas Cre4it Program: Destination of Loans,
1 964-m~id-1967  4*               **    C       96
28      Caldas Credit Programn:  Crop Area and Number
of Animals Financed, 1964-mid.4967    * . * *   97
29      Pilot Diversification Project - Planned
\  Substitution ror Cof fee            * a o   a e  * * 103
30      Pilot Diversificationi Project   Labor
Relu.irements for Selected Crops and Live-
stock Act       itie s      ..............     105
S s il a u M p li z s r s j & . ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ . ...... .jJ@;2mt>Swc-Atz .< az,  t .X  at



APPENDIX TABLES
No.            .Page No.
I     Colombian and World Coffee Expoits, 1930-1966.      15
II     Colonibian Coffee Exports,' by Principal
Destinations, 1946/47-7965/66    l . . . . .      16
III     Coffee lroduction in Colombia, 1930/31-1907/68      86
IV     Indices of Domestic Coffee Prices and Cost of
Living, 1930-1967    . . . .   & . *i - * * .    88
V     Indices of Coffee Prices, New York, and
Domestic Wholesale Prices of Coffee in
Colombia in Real Terms, 1.930-1967  . . . . e     89
VI     Colombia:  Cof fee Prices and Production in
Selected Periods      .          .                90
MAPS AND GRAPHS
Coloimbian Coffee Map   . a * * . e * e a  * . 9 a  . .     42
Graph 1     Colombia:  Indices of Domestic Coffee
Prices and Cost of Living    ... .   . *      70
Graph 2     Colombia:  Indices of Coffee Prices, N.Y.,
and Domestic 'iholesale Prices of Coffee
in Real Terms   . . . . . . . .. . . . .      72
Graph 3     Colombia: Coffee Prices and Production
for Selected Periods.                         740



SUJIMARY AND CONCLUSIONS
Colombi.a in World Coffee Trade
Colombia, with a level of exports of about 6 million bags per
year, is the second largest coffee exportliig count;ry after Brazil,
accounting for some 13 percent of the world total. This is about the
same as the share held by Colombia in the early 1930's. Its share
rose to substrantially higher proport.ions of total world exports during
the Second World War and in the post,-war period up to about the middle
195('s, buit since then Colombian export growth han faZ.led to keep pace
with that of world exports, In the pre-war period, Colombia's share' of
world exports advanced largely at the expense of Brazil, whic:h was
unilaterally holding exports back ill an effort to sustain the
price. With the weakening of world prices in the mid-1950's, Colomrbia
joined with Brazil in efforts to support the price by restraining exports,
and other coffee expo-ting countries expanded both production and exports
at the expense of the two leaders.
During the post-war periodi there has been a marked shift in
Colombian coffee exports from the United States to Europe, with the present
position being that the volume of exports to Europe is almost as large as
that to the United States. This occuirred in connection withl a shift in
United States purchasing to cheaper sources of supply, especially to
AUrican robusta coffees. In the face of these developments,, Colombia has
attempted to diversify its markets as broadly as possible in Europe. In
this effort, it has relied to a considerable extent on exports of coffee
under bilateral payments agreemtns. Exports under such arrangements have
increased more than four times since 1960, and now account for about 15



pereant of total exports, or close to 1 million bags per year.
As these ccnpetitive difficulties have been encountered,
there has been stnewhat of a narrowing of the price prenium over other
types of coffee which has traditionally been cbmmanded by Colombian
coffee   Colombia recently p.-rticipated actively in a movemtnt under the
International Coffee Agreement to narro the margins among the price
ranges established for its coffee and for other types of coffee, in
application of the selective price criteria for adjustments of export
quotas diff6rentiafly ror 'the' several types of coffee. Colbibia was not
prepared to suffer cuts in its export quota in order to sustain the price
for its coffee at substantial margins over those of the other types of
coffee.
Coffee and tli Economic Deveiopmgnt of Cqom~
Coffee cultivation accounts .for 7-8 percent of the gross domestic
product of Colombia, or about, one-fourth of total farm output. It
accounts for about one-sixth of the croplands under cultivation, and is
c;arried out on more than 300,000 farms with a population of clese to 2
million persons out of the total population, of 19 million. Excluding
taxation on coffee earmarked specifically to finance the accumulation of
surplus p,roduction, cof'fee provides about P-10 percent of general govern-
ment revenues.  Impoxrtant as coffee is in these relationships, its kcey role
derives from the fact  that it is the major source by far of the country's
foreign exchange earnings, accounting, even after substantial declines in
the share from the mid-1950's, for closei to two-thirds of the value of total
exports.,
With a maxirumi prospective gro-wth of coffee expof't easeinings of 2.5
percent per year as a result of the total world supply-demuad telationships
in this camuoxity, it is obvious that the maintenWace of ai *4kMrVe rate of



growth of the econcay is depelident on a ;substantiaL exansion of other
exports,  en March 1967 a basic change in the country's foreign exchange
reginie was enacted to stimulate such an exf,anoion df exports. Through
a flexible exchange rate which would depreciate in relation to the move-
ment of internal prices, in addition to subsidies, special credits, and
promotional services, it was hoped that the goal would be achiered of
raising "minor" exports, i.e., exports other than coffee and petrolewnu
from a level of US $128 million in L966 to $300 rr,lllon by 1970, ccnpared
to the projected value of coffee exports of $3140 .rillion in the latter
year. Even with such high sights for -export diversification, a current
balence of payment,s deficit was projected for 1967-1970 of P.bout double
the de:ficits incurred in  96L-1966, in order to support a national economic
growth target of 6 percent per year. Cvloinbia will depend on foreigon
aid to achieve the closing of this gap.
With a population growth rate of 3.2 percent per year, a
growth rate of 6 percent is necessary in order to achieve an adequate
increase in income per capita. During the pefriod since 1950 the growth
rate has averaged substantially less than that, amounting to 4.6 percent
per year both over the entire period 1950-1966 and during the recent
interval 1961-1966.
Agriculture, although declining relative to total national
*outputi still accounts for about 30 percent of the total. The growth of
agrLcultural output has been slightly lower than that of population,
though it has been held back by a very slow growth of output of coffee.
Excluding coffee, the agricultural. slhowinig has been slightly higher than
that of population growth. Many factors are responsible fGr the lagging
growth of agriculture, but among the outstandang ones have been the
patterns of land tenure and use. The larger part of landholdings is



accounted for by a small part of the population, and this has been
accompanied by a tencdency toward extensive rather than intensive
utilization of land. This has had an adverse affect with respect both
to expansion of agricultural output and the provision of employmerit
opporturities in agriculture for the growing population.  Symptomatic
of these conditions is the extent of under-aloyment in rural areas and
iinemplo3ment in the cities.  In recognition of this situation, a land
reform law was enacted in 1961 and an active land reform institute,
TNCORA, was entrusted with its adminictration, The extent of land
redistribution activities by INCORA has, however, been limLttdLn  During
its first five years of operations, fewer than 10,O000 families a year
benefited from INGORA's land distribution program, compared tp a growth
in farm families estimated at about hO,000 per year. INOORA attributes
its difficulties to ColombizQis poor soils and the unsuitability of many
of the large landholdings for intensive exploitation.
Althogh manu.factaring output has increased at a higher rate
than both agriculture and total outpuZt in Colombia, it has eqp4mded
employment at a much lower riate than output, because of its capital-
intensive nature and low level of capacity uti4Vization. The growth of
exports of manufactures is hampered 'by the hi4hcosts of many manufacturing
activities, which have been protected 'frtm for4gn ccmnpetition largely as
a by-product of balace of payments imort restrictions.
The employment problem resulting fram the nature and ipace of
agricultural and industrial growth i.n the country bears signiticantlr on
the coffee adjustmert and diversification prOloiop.  The problov of provid;ing
alternative employment to resources engaged in surplus coffee Mroduction
adds to an alr eady stiubborn general prob2em of creation of .emTh-oyment
opportunities in conne.tion with the rerall development of li*w (eponcmy.



ee                                      Coffee  ector
Colombian coffee production at a- level of about 8 million bags
per year has undergone virtually no change during the past decade. The
long-term growth rate since the early 1930's works out at slightly more
than 2 percent per year. There i8 little conclusive evidence on the
trend of area planted to coffee., which is how estimated to amount to about
1 million hectares, hat it has apparently been stagnant or declining over
the last ten years. There is similarly no evidence of significant change
during that period in the estimated 2 billion coffee trees in the country.
As of 1962/63, 60 percent of the coffee trees were over 15 years old, and
should therefco,e have declining yields. It is believed that the rate
of replanting is low, and therefore no significant change should be
expected in the number or age composition of trees.
Close to three-fourths of total output comes from the west
central section of the country, where coffee is cultivated on the slopes
of the western and central ranges of the Andes mountains,   Coffee was
first introduced into the country to the east and north-east of this
central region, but has historically shifted to the west central area.
This regiorn has the highest yields per unit of area, and is the best
endowed in the country ecologically, for coffee cultiv.tion, with
an optimum altitude range of 1,300-1,800 meters, and with much of the
coffee lands on steep slopes. With a considerable range of climates
in coffee producing regions, there is little annual fluctuation in output;
adverse weather conditidns in some regions tend to be offset by favorable
conditions in others.



-vi-
Coffee cultural practices are siple and highly labor-intensive.
The national average yield ts about ',00 kilos of green coffee equivalent
per hectare, but the potentials for increased yields are very great.  An
outstanding characteristic of coffee cultivation in Col,mbia is the small
size of many of the farms. About one-third of' the more than 300(,000
coffee farms have less than 1 hectare each, and there has b>on a trend of
deline in the average size of farm. The smallest farms cultivate coffee
over most of their area, with an  inverse re-lationship betwbh the size of
farm and the proportion devoted to coffee. The smallest fax~s also tend to
be on the steepest lands, and consequently ofter the most li'nited possi-
1ilities for alternative activities.
Monetary returns to toffee tultivation are estimated et some
3,800 pesos/ 'per hectare at the nationAl aVerbage yield5 cbuhting labor as
a return rather than a cost, and making no allowance for d-piSciation of
investment  'This works out to 29.2 pesos per man-day for 130 man-days
of labor requiYed to produce the aviera&  output per hectate i5 a year.
Margins increase more than proportionately with increase's ih yields. In
one extreme case of a fhrm visited by the miaaion, in which the most
adfraiced cultiyation technbques are applied, th'e yield is &8 muph as
7,000 kilos per hectare, and the return on invostment is estimated to be
as high as 66 perqent. Changing over to the advanced production practices
(new varieties of trees, high density per unit of area, heAVy application
of fertilizers, and growth without shade) requires investminnt, and is
held back because of the high cost of eredit Ai the risks iiolJAed in
sch investment.
1/  16.25 pesos   US $1.



The L0-year old National Federation bf Coffee Growers, a
private bodyv representing the interest or' doff6e producers, Is the
pivotal agency in the carrying out of the coffee policy of the country.
This agency constitutes a remarkable exaMple of organization and stabiLity.
It has had $nly two general manager.; duriiTh its entire history, and
employs some 2,800 persons. By arrangeiient with the government, it
car,ries out Colombiats obligations urnder the International Coffee Agree-
ihent. tts pur6hases 6f coffee, amounting to 60-70 percent of the total
crop, at support prices fixed in agreement with the government,, constitute
the basi c factor in internal price policy. The Federation is the agency
which buys and sicores surpluis coffee, which it finances primarily through
the proceeds of a 19 percent "coffee retention" tax, entrusted to it by the
government for this purpose.  The Federation's storage facilities for
coffee represent close to half of the country's total storage capacity
for agricultural products.
The Federation support piice for coffee is fixed by a cammittee
consisting of the Federation's General Manager, the Manager of the Central
Bank, and three Ministers of the National Govermnent - Finance, Agriculture,
and Development. The price is, of couirse, influenced greatly by the level
of taxation, which is applied at the point of export and amounts to
slightly more than 40 percent of the excp)ort value, including the 19 percent
"retention" tax. The principal element in taxation besides the retention
tax is anesport tax, which now (February 1968) amounts to 22.25 percent.
The export, tax was imposed in March 1967 at the time of basic modification
6f the exchange regime, in replacement of a penalty export rate that applied
to coffee; it was initially established at a rate of 26 percent, with the
provision that it was to decline by one-fourth of 1 percentage point per
month until it settled at 20 percent in December 1968. At the same time,



the export rate applicable to ooffee was established as the same
"certificate" ritLe applir;ab.l r to q-.orts; generally,, and the certificate
rate was left to fluctiur-ite in the market. Thus, while the export tax
rate was to decline steadily., the peso proceeds from coffee exports by
virtue of the exchinge rate were also to Increase; indleed the certificate
rate rose by 20 percent fromn 13.50 pesos per dollar in March 1967 to 15.70
pe6cs in November 1967. The Federation support price .Ccr coffee was
raised by scne 18 percent during that period. The rationale for this
poiLicy is to improve the terms of trade for coffee producers relative to
their position in 1962., and to keep private exporters fr,cm selling too
cheaply in international markets. The goverrnent is aware of the need,
however, to balance against these objectives the undesirable effect of
i.ncentives for expansion of coffee output. The net effect of the opposing
tendencies of supporting the inccrie of coffee growers and applying a high
irate of taxation on coffee exports, which have been at work for sane time,
has been the low rate of growth in coffee outout.
Colombia nevertheless has accumulated substan'tial surplus stocks
of coffee. The stocI% accumuilation is reported to havne begun in 1959/60,
and to have risen to 5.3 million bags at the beginning of the coffee year
1967/68, i.e., as of October 1, 1967, of which some 4.7 million bags are
c onpidgred to be purplus. The annual average accumulation during the
8&year period 1959/60-1966/67 was about 600,000 bags, though the rate
o±f accuinulation steped up during the last three years of the period
to aboyt 9.0,000 pO     per year. With only a modest rate of growth of
output, the surpluses have resulted frm restrictions on exports,
pnrticular1y unde4r the International Coffee Agreement.



P!rojections of Colombian 3urplluses through 1972/73, the
period for which the International Coffee Agreement is now being
extended, range fram a high estlimate ol' about' 80d,000 bags per year
to a low estimate of about 500,000 bags per year. By the end of the
period, i.e., in the year 1972/73, the high estimate surplus would still
be some 720,000 baLgs, but the low estimate su'.plus woulQ amount1 to only
some 200, 000 bags. The factor accounting for the difference in the high
and the low estimates is the assumption concerning production growtht a
growth rate of 2 percent per year is assumed for the high estimate and
1 percent for the low estimate. The e&sential reason for the lower
surpluses proJected in either case ccnpared to the last three years
is a .more satisfactory rate of growth of exports.
Solution of the long-run cof'fee adjustment problm is thus
heavily dependent on internal policy to keep the growth rate of output
within the limit 1f 1 percent per year,; At at growth rate of 2 percent per
year, a rate of annual surpluses of some 800,000 bags will most likely
continue. Such a rate of surplus, at the present national average yie7 i
of o00 kilos per hectare requiring somne 130 man-days of labor to produce,
would seem to make redundant the labor of some 50,000 persons, assuming
no change in existing under-employment, arnd the cultivation of some 100,000
hec-tares.
The Coffee Diversification Problem
The only concert,ed effort carried out in Colombia thus far aimed
at diversifying economic activity in tbe coffee sector is the Caldas
Regional- Development Program formllated for execution during '.he period
196l)-1968. However, this program did not have as a. goal a cut-back in
coffee production. Aimed at the principal coffee producing region of the
count,ry, its objectives were to inceroase the overall growth rate of the
region and to reduce its dependence on coffee by limiting the growth of



coffee output relative to that of other agriculture and industry. There
was a stated goal of r6c1ucing coffee acreage by less than 10 percent,
but none with respect to coffee output except to keep it "within the
limits of the domestic and export markets". By the middle of 1967 less
than half of the total investments formulated for the prograjn had been
carried out. As for agricultural credit, whith constituted a main
element in the program, some 3WOO farms, or saae 5 percent at most of
those in the region, received supervised loans,, eame intended fpr coffee
rationalization and others to finance 'livestock and crop  other than
coffee. While the incomes of the farmers benelt'ted may have increased,
the effect on coffee aears to have bee    soe increaze in owutut by the
farmers involved, precisely as their Incomes it-creased and were partly
plowed back into coffee. The problem of iaaking alternative activities at
least as attractive financially as coffee has not been met, though the
program has introduced sorie elaento to offset thi traditional coffee bias
in the region.
In recopgition of the need to do more than merely caftt the
"coffee mentality" in the contr-y axd diversify the sOurces of income
of coffee farrers, the "olombian Soveruent constituted a coeI1vttee in
1966 to formulate a pilot coffee diversiflcatio project to be undertaken
after completion of the Caldas prograr.m. In the work which has thus far
been done on this project, there is included a goal of reduction of coffee
output in addition to merely a reducti;cn ii acreage. The goal is a
reduction of smewhat more than 3OO,O.C bags per year, which would go only
part way in meneig the surplus problem under tho high estinate projection
cited above, though It would meet the problem on the low eetlAwtv basis by
1971/72.  However., the pilot project is once again aimed at t1e principal



coffee producing region of tht, country, znd it L's not 4et clear how coffee
yields will be kept f7ran increasing and not offsettinfl the reduced acreage,
particularly under the national policy of increasing the incomes; of
coffee faxmers.
The basic problem in coffee diversification in Colombia, in
the sense of replacesaent of same of the coffee output by other economic
activities, is that the policy of support of coffee prices, while holding
increases in output in check, has not resulted in an adjustment of otuput
to the total marketiig possibilities. .At the Same time, steps ro make
alternative activities at least as remunerative as coffee production have
not been pursued with sufficient vigor to be effective. The value of
coffee surplusos being accumulated amot.nts to about, 425 million, pesos
per year, which may be taken as a measure of the wasted effort fran the
viewpoint of the national economy.
I         The elimination of some coff6e output and its relacement by
otLher econ,ric activities have to be pursued Jointly. Irn ccasidering how
to achieve this, decisions have to be talken with respect to the regions
of the country and the types of farmers to be included in arWr diversification
program.
The pro'blem of relative efficiency in coffee production of both
regions and types of farmers has mrrre than one dimension. Consideration
has to be given not only to the relativlr  efficiencies in coffee production
itself but to the relP-tive opportunity coats of coffee product-Ion, i.e.,
the relative efficiencies in alternative activities, Looked at in this
way, it is not nocessarily clear that the principal coffee producing region
Of the country is the one where the diversification should be carried out.
This region appears to be at the same time the most efficient in coffee
prodicetion and the prospectively least efficient for alternatives. As



-xii -
regards the types of fanmers, where the most convenient distinction
th-t can be made is by .size of frm, it is by no means clear which are
the most efficient in coffee production. When introducing the opportunity
cost element, it appears Ptpriori that diversification should be aimed
at the larger farmers rather than the wiall ones . Such an efficiency
criterion coincitdes widh conaiderttions of equity, involving the undesir-
ability of eliminating for the small farmer even the present low level of
returns for his effort in exchange for still bleaker alternatives.
The cost of coffee drSrification Vould, of course, be
dependent on the decisions  iat have to be made with respect to the
most appropriate alternative activities to be induced. Short of estimates
which have to be made in detail on the basis of such decisions, a general
order of magnitude ray be obtained by applying very rough factors of
amounts of investment required per work-r for alternative activities.
Such a proceduire yields an estimate of l.5-2.O billion pesos to employ
the 50,000 displaced workers previously mentioned, which is equivalent
to about five years of the annual cost of buying and storing 5ukeplus
coffee and is about equal to the magrnitude ib present value of the
5-year Caldas investment program.
The basic need for a coffee diversification strategy is.sto
fomulate . program of ci.t-backs in coffee ard o1, alternati;e 6nploymqnt
for the resources displaced, by both region or regions and types of farmers,
to be worked out by application of opportunity cost principles. The
divfirsific#tion a?tivLtips could include alte.^native activities on coffee
lands as well as agricultural, forestry, indtlitfial, and public works
activities near the coffee regions.



In order to incluce cut-backs in coffee production, fiscal
rnsc$anisms might be considered, such as differential taxation on large
famers in the region or regions selected  for diversL .fication.  The
taxes could be related to such criteria as size and characteristics of
farm, or relative returns to coffee and piospective alternatives per
unit of area.
One of the weakest areas to date in coffee diversification
activities in Colombia, and indeed in agricultural development generally,
has been that of providing support through guaranteed prices and marketing
outlets. Much more will lhave to be done in this area to solve the problems
in both cases.
To meet the costs of a diversification progrmn, as a rainimum
.,resent taxes on coffee should be maintained, including the retention
tax, even if and as surpluses and the need to finance then decline. As
the need for firiancing of surpluses declines, a pool of funds would thus
gradually be created out of which to finance diversification. Furthermore,
there would be some added reveniues concurrently resulting frm the
differential tax imposed on the larger farmers, In addition, Colombia
would have resourges at tilable for diversification arising frwrt its
participation in the In,vernational Coffee Diversification Fund, which
has been agreed to by the International Coffee Organization.
The decisions to be faced in working out a coffee diversification
strategy in Colombia are thus difficult ones to malke and carry out.
However, they are no more difficult than those involved in the general
strategy required to meet the country'o stubborn problems of nployment
creation, income growth, and general economic diversif'ication, in connection
with which there is considerable satisfaction in the international camaunity
with the policy directions being pursued.  Furthermore, hope
that the country will be prepared to adopt the measures required for solution



-xiv-
of the coffee adjustment problem can be derived from the prudence and
responsibility with which it has managed its coffee policies in the past,
and frcau the relatively small magnitude of the problem conpa?ed to a
number of other countries and in the contoxt of its overall econonly.
, I'



CHAPTER I
COIW)MBIA IN WORMD COFDEE TM
~I
Colonbia currently accounts fr dpprbximately 1.3 perc'nt of
the volume of world coffee exports. This is about the same as the
share it had irL the early 1930's, but represents a substantial decline
fbrum appreciably greater shares it achieved in, subsequent periods.
As indicated through the 5-year averages contained in Table 1, Colombia's
share in world coffee trade steadily advanced into the late 1930's, and
into the period following World War II, until it reached peaks in the
mid-l950's when the sellers' market for coffee prevailed to the greatest
extent.' IWith the weakening of markets that set in after the mid-1950's,
Colombia's share started a steady decline to the current level.
Table 1
Colombian and World C ffe    orts 1    -166   -Year Av     s
(green coffee, millions of bags of 60 kilos each)
(1)         (2)          (3)
Colombia     World
1930-34        3.2       26.0          12.3
1935-39        4.0       28.0          14. 3
1947-51       5.5 1      31.*2         16 3
1952-56    VI5e7      *  34.0          16e.3
1957-61        5-.7      40.0          14.3
1962-66        6.1       47.0          13.0
SoCuuroe: Appendix Table 1.



s~' 2
-2-
This is 'shown even more dramatically by data relating to
individual years, contained in Appendix Table I A prewar peak share
for Colombian exports of 17 percent ob the world total is attained in
1938,  In a number of years during the abnormal period of World War II,
Colombia's she   was at 20 percent or above, reaching 29 percent in 1943.
T.a country's share declined somewhat during the early posta    years,
but later recovred, rewching a peak of 20 pereent in 1954h. Since then,
Golombia's exports have failed to keep pace with world exports,, and its
share has declined to the present level. Colombia's share in 1he basic
export quotas of the current Internatioral Coff-e Agreement has been 12.9
percent. At the Eleventh Session of the Intarnational Coffoe Council
held from Novsuber 20 to December 41, 1967, Colombia's share of basic quotas
for the now Coffee Agreement to extend.through 1972/73 w&s ti#d at 12.7
percent.
During the period of sharp dooline in world coffee prices in the
1930's, Colombia's prodaction and exports both advanced. Its siare in
world exports also advanced, at the expense of Brazil which
was carrying out a unilateral effort to sustain world prices by
restricting exports4 With the fall of peak coffee prices after the
mid-1950's, Colombia increasingly joined with the other majqr producer,
Brazil, in restricting exports in an effort to sustain prices.
Colombia and Brazil took the lead starting in 1957 in arranging several
anxrual :-nternatiozial 9offee agreements design9x to rqstrict eports and
support the market. The Brazil-Colombia leadership eventually led to
the existing International Coffee Agreement. The market support action
by Brazil and Colcmbia wits, of course, partly in,trumengal n causing the
increase in production and exports of other countriw which h*s occurred



-3-
over the last fifteen years. The cost to Colombin of acting to achieve
1czue stability of world coffee prices has thus been a gradual reduction
in its share of world exports.
Destination of E!rts
There have been several marked changes in the pattern of
export destinations for Colombian coffee during the postwar period.
Taking the entire period of some 20 years, there has been a sharp
shift in the dostination of sales from the United States to Eurape.
Since the begihnirtg of the present decade, th&re has also baen a
large increase in sales of coffee under bilateral payments agreements.
Sales under the provisions of such agreements now constitute about 15
percent of the volume of total coffee shilpments. Such sales appear
to be made at prices samewhat higher thant the New York spot
price, but Colombia must surely be paying higAer prlnes in return
for goods imported under these agreements than it would under more
freely campetitive circumstances. Markets to which coffee is sold
under bilateral payments agreements incluide both countries which are
"traditional" Annex A quota markets under the International Coffee
Agreement, as well as "new" Annex B non-quota markets. Taking the
Annex B exports as a whole, both to payments agreemnts countries and
to countries whose foreign exchange is convertible, Colombian eports
have increased to omeo extent, though less mnarledly than they have
under bilateral payments arrangements, and .such sales to non-traditional
markets now constitute only about 3 percent of the total volume of
Colonbian exports.



-4--
The broad geographic distribution of exports over the entire
postwar period is indicated in Table 2. The decline to the United
States has been marked in botih absolute and relative terms, ad the
reverse has occurred with respect to &irope. Exports to the United
States went down from  4.4 million bags per annum as an average in the
first five postwar Yars, or 94 percent of Colombia's total exports,
to 2.8 million bags in 1966/67, which represent half of total exports.
Exports to Europe in the lattfer year, at 2.5 million bags, were almost
equal to the total fcr the United States.
Table 2
Colombian Coffee Mgorts -Wjor Re ioral Distribution,
5-Year-Averagie9 192/4 7-1265'66 and, 166/7-
(green coffee, millions of bags of 60 kilos each)
1916/47-    1951152-    1956/57-    1961/62-
9O/L,     _l25/S6      1960/61,    126W6       1966/67
U.S.A.            4.14        149         4.3         3.6         2.8
Europe            0.2        0°7          1.1         2.2         2.5
Other
countries       01          0.2                                   2
Totals                      1jj j,
Source:  Coffee Federation
The increase in exports to Europe has been widely spread among
a mmber of markets. The principal outlets are: the Federal Republic of
Germany, Swgdep, Spain, the Netherlands, and Finland. Other significant
European markets are Belgium, Poland, East Germany, the United Kingdoan,
Italy, Denmark, Noray,, and Switzerland.
Outside of the Urited States and Europe, Canad.a, japap and
Argentina are by far the principal markets, Exports to Canada have
varied little over the postwar period, and continue at 1O0,00Or-lO5,000
bags per year. Exporta to Japan and Argentina, op the other    ad,,



which were insignificant through most of the 1950's, have recently
picked upy despite strong canpetition from other suppliers, and are
now same 60,000-70,000 bags per year each.
The shift in exports fram the American to European markets,
and the spread in sales in the latter markets are to same extent the
result of Colombia's deliberate marketing policy of distributing its
coffee as widely as possible and of gaining good positions in markets
with the greatest growth potential. While very ixnpcrtant, the United
States market has appeared t,o have a less favorable outlook wtvh respect to
growth than some of the others, and the conditions of the American coffee
trade have led to a shift of United States purchases to cheaper sources
of supply, with less concern for the quality of coffee produced. Colombia's
reaction to this situation has been to maintain insofar as possible a
substantial flow of coffee to the United States but to put the weight
of its export promotion on more discrimin.Ating and promising EiLropean
markets.
By the mid-1960's, Colombia had reached the situation in which
its coffee exports are distributed among the major markets to an extent
roughly comparable with the importante of the various markets in the total
of world imports. This phenomenon is illustrated in Table 3.
The classification of Colombian exports by quota and non-quota
markets of the International Coffee Agreement and by types of payments
agreements is indic,ated in Table 4, andi the principal countries with
which Colombia has bilateral payments arrangements for coffee exports are
indicated in Table, 5- The largest part of trade is conducted in terms
of convertible currency, mainly with North America and Europe, Most
of the increase in coffee exports under bilateral payments agreements



-6-
has been to Arnex A quota markets. The principal ones are  pain, Finland,
and East GoermaxW,, followed at sace distance by Denmaark, Argentina, and
Cswhoslovakia.
T Able 3      *
Distrbution of Calobian Coffe EEsA         § 9s1mifed 2a,el
and Shae of t ht in Wbld Total
COL0OBIAN COFF                MOTAL COPM
EXPORTS                    lDORTS
s * t ;
(l96V/65-1965/66           (1965/66 average)
2mw- - -- --
'000 bags of    ttal     000 bags cOLf  % of total
60 kilos each  exports   60 kilos each   w0rld imports
U.S.Ae                 3'260         55ls       20,169         42.6
E.&Co                  1,322         22.5       12,626         26.7
Othar Western
Eiirope                356          6.1        2,538          5.4
Northern Eurcpe          534          9.1        4,838         10.2
Oth-er        .                 ,,1,15.1
Thtal                              0
Sources: Coffee Federation and Tripartite Coffee Study.
In the Anmex B markets, Poland has p1qyed the maJor role as
regards exports under bilateral pajymeta agreesmts, and Japan is the
sole convertible cuirencymarket, Expwta to the U.S.S.R. awe not



-7-
carried out under any broad payments agreetaents, but on a barter
bass, is ., involexing an echange of coffee tor specific Sdviet
products>-
Taking coffee exports under bilateral payments agrements
as a whole, exorts amouAt to close to I million bags per year,
most of which goes to Spain, Finland, and East Germany. The Federation
has Pound these schemes useful as an instrument to stimulate coffee
wcorts.   The agreements are negotiated and ilpemented by the
Federation together with the Central Bank of Colombia. The increase
in exorts under these agreements has been more than fourfold since
1960.
For example, it was reported early in 1968 that Colonbia had
signed an agreement with the U.S.S.R. involving the ewhange
of some 18,700 bags of coffee for 50 Soviet trolley buses. The
price for the coffee was apparently 46.5 cents per lb., or
about 10 percent higher than the New York spot price. It was
also reported, however, that the Soviet buses, at a price of
some $23,000 each, cost more than si-milar buses available from
Canada and the United States. The 18,700 bags of coffee involved
compare with shipments to the U.S.S.R,. of 7,000 bags in 1964,/65
and 5,000 bags in 1965/66.



Table 4
EMorts oaf G   abian Coffee to Annex A and Annex B
Marketp 2Tade Pratl        -26e 61-1965/66
(green caffese, thiousands of bags of 60 kilos each)
*  .6/1 16/2     162/6     12963/6i  196V.65   1965/66
ANNEX AS
Convertible           5,827    5,9274     5,757     6,075     5;,.207   14,962
BM lateral pa
agreements-           241     357         L47       580'      644       678
Convertible7             46     52           63        59        35        69
BilateralJ paents
agreements,W           28     1J                     31        143       80
Barter9/                  3                             -         7
Totals              6,45                6,339     6,715    '5936      5,794
a   Thore am ame      c roencies between these figures ad those reported by
the Internatianal Coffee QrgaztLzation;  cf.. Chapter IQ ai, ble 13.
b/ Argentina,, I3ilgaria, Chile, Cze  eLIvakia, Dnmark, Sant   ermany.,
Finland,    in mand Yugoslavia.
c/  Japan.
A/ All Amex B c=mtries, excePt Ja          l    S UJ.S R.
5/ UeS.S.R.
Source: Coffee,Fe4aration.



-9-
Table 5
Colombian Coffee Exports under
Bilateral PA~Yments Age-ets 16-1966
(green coffee, thousands of bags of 60 k5.lo6 each)
Spain   Finland  East Germwn    Poland  Denmark   USM    Others Totals
1960   33.3      51.8       23.6          -     26.5     3.3     59.1  197.6
1961 100.2      89.3          -        !25.0                     64.4  278,9
1962 204.9      104X8       31.5          -     32.2             83.1   4565
1963 250.4      104.2       36.0        28.7     32.6     -      85.3   537.2
1964 235.1     146.5        80.0        34.6    37.1      -      47.6  580.9
1965  300.2    154.3        78.7        50.4    44,7    11.7     73.8  713,8
1966  333.6     206.1      121.8        84.2     53.3   33.3     46e2   878.5
Sourcei Coffee Federation.



-10-
coimetitive Factors
Colcmbia bas established a reputation in the world ooffee
trade for cons       good quality, steadiness in supply ava abilitiea,
and an elaborate and efficient marketing structure. The country is
reported to have ad no dcifficulty in filling its quotas under the
International Coffee Agreement. However, one reason for its need to
resort to bilateral pay mts agrements to market coffee even in
quota countries aw$ b that it has been' running into sone competitive
difficulties. This is also reflected in its sharp relative loss of
standing in the 1     U.S. market, as we3   as in some move twd
narrowing of the price preiums traditiorally enjoyed by Co2labian
coffee relative to other mild coffees, to Brazilian coffee, ard to
the robusta coffeesof Africa.
The absolute and relative declines in Colombian aports to
the United States have been accaManied by a substantial change in the
composition of imports into that market by types of coffees and geographic
origin, particularly since the early 1950's   . Colombia's share of U.S.
imports   minisbed from 21 percent as an average for 1950-52 to 12 percent
In 1966. A declin7 bf a similar order 9f magr4tude occurred in the share
of Brazil in U.S. imports between the two periods (from 52 percent to
30 percent). The declining shares of Colombia and Brazil were taken up
by a substantial IxWretse par ticularly in the ehare of African coffees in
that market,, and even by some increase in the share of otlher mild coffees
from the Western Hlisadphere. The share of robusta coffees in the U.S. market
increased from 2-3 perfcen in 1950-52 to as mch as 28 percent in 1966;
the share of un      aarabica fram Ethiopia increased from 1 pcrcent to 4
1/  "Develoment     U.S. Coffee Consumption, Analysis and Fcrecasts'",
preared for T#pqrtite Coffee 4'tudy by Ckertrud Lovasy aad George Kawata.



percent;   and the share of "other milds" frc* 22 percent to 25 percent.
The change in the caEosition of U.S. imports reflects the trend of
increase in the marketing of soluble coffees and a general 1trend of use
of less expensive types of coffee, even in reular blends.
The price premiums enjoyed by Colombian coffee over other
types of coffee, based on the relative unit values of iports into the
U.S., are illustrated in Table 6. Thus, the Iit values of imports
from Colombia have exceeded those of other mild coffees by a range of
2-9 cents per lb. during the period 1953-66; they have exceeded those
of imports frcm Bbazil by a range of 3-19 cents per lb; and they have
exceeded those of imports of robusta coffees by a range of 10-29 cents
per lb.
Under the scheme introduced into the International Coffee
Agreement in the coffee marketing year 1966/67 of partial adjustment of
quotas of fcotur different categories of coffee on the basis of price
movements outside a range separately established for each tpe, the
premiums fixed for Colombian coffee relative to the ranges of the other
three types.,as indicated in Table 7, were as follows:  3 cents per lb.
over other mild arabicas, 6 cents per lb. over Brazilian coffes and
other unwashed arabicas, and 13 cents per lb. over robustas. This
system resulted in a cuR in the Colombian, quQta for that 14ar of
360,000 bags because of declines in the price for Colombian coffee
below the floor established. This led to a great resistance on the
part of Colombia in negotiations under the AWe.ment to continuation of
the selective price ranges,particularly at these levels.  Though the system was
continued in the coffee year 1967/68 apd has bean incorporated in the new
Coffee Agreement, the premiums of the Colombian range over those of the



-12-
Table 6
L Am      -UhtYl              e, Lm
(cents per lb.)
Oter
Other
milds            Ro-uslas
1953    55.9       50.2      52.9       46.0       5.7      3'.       9.9
1954    71.0       63.5      64.8      55.4r       7.5      6.2      15.6
1955    62.6       53.9      4T.7       37,8       8.7     4.9.      24.8
1956    61.8       59.4      46.1      34.6        2.4     15.7      27.2
1957    63.6       54.8      44.9      3Jj.3      8.8     16.7       29.3
1958    51.8       44.8      41.2      36.0        7.0     ;.Q.6     15.8
1959    45.0       37,8      32.1      27.6        7.2     12,9      17.4
1960    43,4       37.;5     32.6      2Q.18      5.9      10O8      21.6
1961    41.9       34.8      32.2      l9,4        ?.1      Fo7      22.5
1962    39.5       32.9      30.1      p.9.1      6.6       9.4      20.4
1963    37.8       32.1      29,7      2 .2       5.7       #446
1964    45.5       41D5      39.6      32.7       4.0       5.9      12.8
1965    454        li..4     .         25.6       4.0      $,5       19,8
196     45.6       39.7      36.8      29o0        5.9      8*8      16.6
b/ Weighted awrp qf inport unit val1ees fork iatin Amerinn milds.
b/ Weighted awrage of iport unit values for ajor Afri¢.an and Asian
sourcea of  *pl1y,
Source:t Lovasy -anl K    $wat. op . cito .



Table 7
indicator Price R "es-for Select ye Partial Qu1 Va AdjuL5tGnts
(US cents per lb., ox-dock Now York)
1266/6.7                2L6
Floor   Celing         Floor   CGel ng
Mild .krabicas&            43.5o      47.50        38,75     42.75
Other :lld arabicasy       4o L5o    44W50         37.2;     41.25
Unwashed arabicasP/        37.50     41.o          35.25     39.25
RobustasW                  3o.5o     3J50          30550     34.25
Excess of range for hmild
arabicas over other
mild arabicas                 3.0                    1.0
Excess of range for mild
arabicas over unwashed
arabicas                      6.0                     3.50
Excess of range for mild
arabicas over robustas       13.0                  8.25-8.50
' Colombian "Mams".
b/ Arithmetic average of Sa.lvadoran Central Standard, O'atemalan
Prime Washed, and Mewican Prime Washed.
9 Santos 4                                       t
WV Arithmetic average of Angolan Ambriz 2AA and Ugandan Native Standard.
SOurces International Coffee Organization.
.                    .  .,  .



other tlree types were sharply reduced in 1967168, as follo%rm:
relative to otWr nild coffees, from 3 cents to L.5 cerdts 6er lb.;
relative to EPrabils and other umwthed 'sabimks, froi 6 cents to 3.5
cents; and rel&tiEe to robustas, fr-Cn 13 cents to 8,25'8.o0 cents.
The general potmon of cooimbia' has 'Nbn a;±nst any, sudh s*electivity
system at all, and for a situation of free price ccwnpet±tirn Abng the
different types of ooffee. OoloSnbia has apparebntly been unable to
caOpete effOeftiely with the Othdr f3tes of ddffee
at wide margts 6f.VIf-niumi  r it's 5n c"ffeeunder the p     k.ti.g
conditions of denad,, and it is not p~r*red ,o suffer cutis I  its
quota in an -ofetfrt to sustain-gsuh jdAitgins6



-15-
Tabl e I
Colombian and World Coffee Exports. 93-66
(green coffee, thousands of bags of 60 kilos each)
(1)         (2)            (3)
Calendar years     Colombia     Worldi   (1)         .L2)
1930           3,173       23,924       13.3
1931           3,039       25,756       11.8
1932           3,185       27,911       11.4
1933           3,327       22,624       14.7
19314          3,085       26,Ll1        11.7
1935           3,768       25s,297      14.9
1936           3,942       27,166       14.5
1937           4,178       27,686       15.1
1938          4l,273       25,357        16.9
1939           3,773       30,0,39      12.5
1940           4,3         29,041      15.3
1941           2.,912      23,390        12.4
1942 4,310                 21,135        20.4
19)43          5,252       18,219        28.8.
1944           4,923       22,765        21.6
19)45          5,150       26,111        19.7
19)46          5,662       27,623        20.5
19)47          5,338       29,304        18.2
1948           5,588       28 ,592       19.5
1949           5,410       32,116        16.8
1950           4,472       34.,164       13.1
1951           4,793       31,858        15.0
1952           5,032       32,328        15.6
1953           6,632       35,194        18.8
1954           5,754       29?,128       19.8
1955           5,867       3)4,0)46      17.2
1956           5,070       39,533        12.9
1957           4,824       35,872        13.4
1958           5, 441      35,976        15.1
1959           6,413       42,241        15,2
1960           5,938       42.,763       13.9
1961           5,651       43,227        13.1
1962           6,561       45,906        14.3
1963           6,134       48,579        12.6
1964           6,412       46,268        13.9
1965           5,652       43,717        12.9
1966           5,566       50,339        11.1
Source: Tripartite Coffee Study



Table II
Colaobian Cd:tfee Eports, by Principal Deatnattiom, 194647-1965/66
(green coffee, thousards of bags of 60 kilos each)
1946/iL7  1947/b8   1948/49  1949/SO   1950/5-1  1951/52   1.952/53  :1953/5h  1954/55  1955/5'6  1956/57   1957/58   1958159   1959/60   1960i61  1961162   1962/63   1963/64   1964/65   1965/6 6
lUnited .3tates          4,923     3,082    5,138     4,253     4,430     b,1403    5,080    5,942     4,150     5,109     3,601     4,1oiL    4,823     4,413     4,545    3,721     4,058     3,913     3,452     3,068
C*cnads                    177      192       211       154       130       131       183      161        81       126       130        82        82      134       111       136       126       133       117       121
Fed.vep. of Gerr'.gny                           2        69       167       30)       191      395       296       42b       163       3ld       629      621       597       727       842        S90      882       869
FAst .'ermany                                                                                                                                      8        25       13         7        49        Io       101        67
selgium                     32        30       43        36        33        38        55        80       55        56        53        66        is        19       51        62        66       143       113       145
Spain                                                                                            18        35       81        33        38        29        50       67       118       204       269       367       234
Piuland                               58       87        54        12        17         4       10         66       39        39         22       43        51       62        98       113       171        75       205
Netherlards                 87        3         10       1C lb                5        68      118        63        9i        39         73      109       199       96       118       161       266       239       232
United Kingdcd               7         1                            1                  10                  3        11         4          6       19        55       117       50        49       116        33        63
I taly                     11         6        10        11         5         2        27       54        19        71        55         20       23       32        47        94        59        89        42        57
rorway                                                                                                     2                                      26        15       19        21        27        62        16        142
Polend                                                                                                                                                                12                 29        17        26        76
S%eden                      56       26        39        jO        53        68        86       89       116       131       116        149      163      185        180      202       223       299       263       282
De-mnirk                                                            1'                 17         8        8        12         9         25       26        27       28        29        34        40        33        55
Sv± tz"rland                13        12    &            44        27        11         i        1?         4       l         3         22       10         9        17       102      104        58        26        33
Japan                                                               3         3         3         4         a       13         8         11       23        38        46        52        63       59        35         69
Argentina                              1       12                                                                                                            i         2         4        13        32        21       56
Other countries             34         10       u       18        27         4        41        29        23       62        11         33       83        86       135       156       119       147       .95      120
Totals                 5,3VC     3,421     5,564    4,679     4,903     4,982     -,779     6,925     4,865    6,247     4,264      u,991    6,166     6,023     6,145     5,697    6,339     6,745      5,936     5,-91
Soviice: Coofef FedeiAifu
.6



-17-
CHAPTER II
COFFEE AS THE ECONOMIC DEVEWP?8,IT OF COLOMBIA
Coffee cultivation, transport, and marketing constitute the
single most inmortant activrity in the Coloihbiari economy. Coffee culti-
v;rtion alone accounts foi: about a fourth oC tctal fam output, or about
7-8 oercent of tota-L gross domestic prod1uct. The role of coffee in
toW-tJl ouitpu' ol goods and servtces i., o1 course, even greater if the
related processing, storage, transport, nmarketing, and financial
activities are taken into account. Coffee acreage accounts for about a
s;ixth of the total croplands (annual crops including fallow land, plus
tree crops), and consists of more than 3(0,000 farms. The farm farilies
engaged in coffee production comprir:o clo:;e to 2 milLion persons, i. e.,
10 r,ercent {or slightly more of the total population. The really crucial
role of coffee in the econony derives, howover, from the fact that it
constitutes the country's major source of foreign exchange.
6tructure and Growtlh of Domestic Produc-t
The slow long-term growth of coffee production has, nevertheless,
resuxlted in a reduction in its role in total national output. In this,
it has shared in the relative decline of agricultiire. The proportion of
gro;3ss domnestic produict originating in agriculture declined from 38 percent
in l])riO to 30 percent in 1966. Agricultuire's loss of 8 percentage points
was accompanied by gains in manufacturing (from 15 percent of gross
domestic product in 1950 to 19 percent in 1966) and services (transport,
cummunications, electric power, gas, water, government, and other services
rose fr<n a cambined total of 25 percent of GDP in 1950 to 29 percent in
.1966).



Table 8
Average Annmal Growth Rates and CGoMosition of the
Oross Domstic Product kESector of Ori&n 1     -16
(percentage)
Average Annual Growh Rates                  Coosition
1950.,56  1956 -61  1961-66  1965  1966   1950-66    1950    1966
Gross D  sc Produet           5.1      I41        4.6    3.2    5.3             100,0   1000
Agriculture                   2,7       3.6       2.9   5O41 5.h 34              37.8   29.9
Fishing and Hunting           7.3      17h4       883    6.6    5.1     10.7      0,1     0.3
Foresty                       7.6       3e6       3.8   10.0    3.3      5.1      a.3     0.3
Mining                        5.1       3.6       4s.9   7.s4  -2.6      4.6      3.5     3.5
Manufacturing                 7.1       5.9       5.9     4.7   7.4      6.4     14.8    19.3
Construction                 108       1.7      -.6    -5.8   -1.0     14.3      2.7     2.5
Commerce                      5.1       3.8       4.9    14.2   5 .1    14.6     15.7    15.7
Transport                     9.2       2.7       4.4     2.8   6,3      5.6      5.0     5.9 ?
Caunicatis5- 9.9                        8.4      10.8-  l.2    7.2      9.7      0.14    0.8
Eletric Famer Gs, Water      10.0        9.3      9'.D    92    2.8      9.5      0.5     1.0
GovermeTt                     5.5        4.3      5.5    15     3.9      5.1      4.6     5.0
Other                         5.2       6.8       3.6     5.9   5.6      5.2     114.6   15.8
f With rwspet to previos, ytew or over Indicated period. eksed on data in constant 1958 prices.
Source- Based on data fran Banco de la Republica.



-19-
Total gr ss dmnestic product increased duiring the period
1950S1966 at an annual rate of 4.6 percent in real terms, or only slightly
more than the high rate of population increase of some 3.2 percent per
3ear. The years 1550-1956 of sharply increased prices and exchange earnings
of coffee recorded the highest anrnal rate of growth o.f national output
during the entire period, viz. 5.1 percent. The subsequent period l956-
1961 of rec&uced1 coffee prices and readjustment trom the policy mianage-
ment of the mnilitary dictatorship in 1953-1957 recorded a decline in the
annual growth rate to 4.1 percent. The period since then, from 1961 to
1966, was one of slight revival, to an annual growth rate of 4.6 percent.,
or the same as the long-term average. The new Administratlon which took
office in August 1966 has formulated as a goal an anmnal growth rate of
GDP of 6 percent in real terms.
Agricultural output has Increased at a lower rate than total
domestic product. Over the entire period 1950-1966 it has barely kept
pace with population growth, and the annual growth of agricultural output
in the most recent period 1961-1966 has been somewhat slower still (2.9
percentY  . Manufacturing output, on the other hand, has substanitially
outstripped total output. Its growth rate was 6.4 percent over the entire
period 1950-1966, though it slowed down in recent years. Still higher growvh
rates characterized the sectors of transport, communications, electric power,
gas, and water supply, reflecting the emphasis on these sectors of public
investment in the postwar period.
1/ However, the growth of total agricultu.ral output has been held back by
the slow growth of coffee, which accounts for about a fourth of the
total. Agricultural output ex-coffee has grown somewhat faster than
population.



-20O
ricu:lture
The agri;ultPiral development of Colombia has bden influenced
to an unusual degree by topography. The three ranges of the Andes,
stretching northwards, divide the coumtry into distinct regions,
communication between which Is dift-icult and costly. Over half the
country, comprising the Llanos and the Amazonian hinterland to the
East and Southeast, contains only about two percent of the 19 million
people who comprise the nation; soils and climatic conditions are rather
poor and'access is diffLcult. The forested narrow coastal belt on the
Pacific with its very heavy rainfall, and the upper levels of the
cordilleras, rising to over 15,000 feet, also are very sparsely populated.
It is in the northern coastal region, the central valleys-,and
slopes of the Andes, and the high plateau of Bogota wher the population
is concentrated and most agricultural activity occurs. These agricultural.
regions represent a small proportion of the total Landl area. Geography
has thus imposed a distinctive regioal pattern of developmeqt, anid has
impeded mobility ithin the country.
Although the share of agriculture in gross domestic product
has declined to bou t 30 percent, the proportion of the total population
dependent on agriculture is still close to half. Despite its falling
share, agricultur, and the associated econcrlc activitie st4ll largely
determine the pattern and pace of growth in Jolombia. Agricultural
production, marketing and processing together may account -for close to
60 percent of GDP.
During recent years, livestock production has grown considerably
faster than the population, while crop production has laggei seriously
behirid. Production of food for donestic consumption, in particular, has,



-21-
in many instances (beAns, corn, potatoes, wheft), become virtually
stagnant. In other words, the poor performance of agriculture as a
whole r,3flects largely the slow increase In output of basic foodstuffs1-I
As a re-ult, imports of food and other agricultural c  hmrodities in recent
years have amounted to 10-15 percent ot thb value of total imports.
Table 9
Growth .f Agricultural Prduti0-28l
(percent increase in volume over the period)
Total agricultural output              20.5
Livestock                               27. 4
Crop production                        16.4
Domestically consiumed food crops
Rice ,                                39.1
Beans                                 4°0.3
Corn                                  5.8
Potatoes                               8.9
Plantain                              22d5
Wheat                                -28.6
Population                              23.0
Source: Banco de la Republica, Cuentas Nacionales.
1/ This is reflected in retail price indexes. During the period
1958-1966 food prices increased 155 percent, whereas the cost of
living iLndex, although heavily weighted by food, rose by 142
percent.



-22 -
The food supply problsn, in turn, is intimately related to
the way the land is used and the closely associated land tenure
conditions- . According to the agricultural census of 1960, over
half of the agricultural land in the Andean and Caribbean regions
is fallow or in natural pastures, while only 15.7 percent is culti-
vated. In other woirds) nearly fourtLmes as much land is devoted to
extensive grazing of cattle  , as is devoted to crops. Moreover,
the cattle frequently occupy the relatively flat lands In mountainous
areas and, in company with the large-scale ccnmercial agricultaral
enterprises producing for export (e.g. sugar), the broad river
valleys; food production for domestic consmVtion is largely confined
to the steeper slopes of the hills and mountains on suall farms.
Agricultural land is highly concentrated in a few hands.
Data in a report by the Inter-American Ccmnittee on Agricultural
Development referring to 1960 indicate that in the Andean region 5.3
percent of the agricultural exploitations account for 67.9 percent of
the land, while at the other extreme, farms of under 5 hectares which
represent 63.5 percent of the exploitations, account for only 1.7 percent
/ There fare, of course, numerous other factors which have impeded
Colambian agricultuire development, including inadequacies in
officii,l plan iing, research anct extension, marketing and infra-
structure 3upports, scarcity of credit, and antiquated methods and
poor management. See IBRD, A Review of TNCORA and its Program in
Colombia,, Octpober 26, 1967.
2/ On average, there is only 0.9 head of cattle per hectare.



-23-
of the land- 1  The distribution of land in the less densely
populated Caribbean region is equally skewed. Even these figures,
howevers greatly understate the extent to which the land is
controlled by a relatively small number of indivicluals.  The data
refer to the distribution of farm size, not the distribution of fann
ownership. Many persons may own more than one fa:rm or exploitation.
Furthermare, tha d&ta on the distribution of land by farm
size fail to take into account the large and growing number of
landless agricultural workers.  Some of these laborers become
pernr&anent hired hands on the larger exploitations while an increasing
number are forced to becone migrants, movlng from one harvest to
another. The authors of the CIJ)A report estimated that in 1960 about
8.6 percent of the agricultural labor force was landlessV; the
proportion is probably higher today.  Some observers estimate that
300,000 families, i.e., close to 2 million people, are in this
position. These people have a starndard of living roughly canparable
to a smallholder or "minifundista" .
The influence of the tenure system upon the way land is used
is seen in the table below. The "'miniCundistas" tend to cultivate a
very large proportion of their land, in spite of the fact that their
land frequently is located on very steep and eroded terrain, and
leave little over a third in pastures and woods. The large landholders
1/ Comite Interamericano de Desarrollo Agricola (CIDA), Tenencia de
la Tierra   Dsrollo Socio-E;conomic                =  =
.edel Sector Agricola en
Colombia,,1966, p. 72,
i/ ~~ Ibd  135.



or "latifundistas", in contrast, cultivate less than 6 percent of their
lnd, and two-thirds of their land is fallowordevoted to livestock.
Thus, most of the food produced in Colombia is provided by a large
number of small farmers who between then own only a fraction of the
agricultural land. Over 70 percent of the agricultural surface is
held in the form of "latifandia", yet these farms account for only
28 percent of the land under cultivation.
Table 10
Th    eofAiculturalLd.
(percent)
Total Agricul-  Cultivated      Fallow and    Woods and
tural land      surface   natural pastures mountains
Minifundia             100          62.5            32.0         5.5
Family farms           100          29.1            49.4        21.5
Latifundia             100           5.7            65.8        28.5
Source:   Based on data fron CIDA, op. cit., p. 139.
In social terms, the land tenure reflects great inequalities
in income and wealth. In economic terms, the tenure systen has led to
inefficiency and slow growth, to both underutilization of most of the
land and to unelcyment of labor. The poor utilization of land an(d
labor is further aggravated by the concentration of nearly all other
resources in the hands of those least inclined to make productive use of
them. For example, a aisproportionate part of the agricultural credit



goes to only a minority of borrower3s.
It is generally recognized that additional agricultural output
could be achieved by redistributing the land held in large holdings to
thiose who could make best use of it, viz., to the growing mnmber of under-
employed rural wori.ers. Thus., through the maciinery established by Law 135
of 1961, a land reform is in the process of being executed. This program
is being adninistered by the Colombian Agrarian Reform Institute, INCORA,
a new agency estab.ished for this purpose.
Progress in the redistribution of lan,d) has, however, been slow.
After five years of operations ,INCORA had. acquired 1.9 million hectares
and given land titles for 1.7 million hectares to 147,688 rural familiesz/.
In other words, fewer than 10,000 families a year benefited frcm the
program, ccnpared to a growth in farm families estimated at about 40,OOQ
per yeari.
INCORA attributes its dif.ficulties to Colombia's poor soils.
According to its investigttions, only 2.8 million hectares or 2.4 percent
o0f the total area of the country, are suitable for intensive farming; the
remainder, apparently, is suitable largely for pasture or is in forestsaW
The expansion in agricultqral productLion wh-Lch it has been possible to achieve
appears to have cczne primarily from an increase in productivity on existing
land, rather than frcn an extension of area. This has not been accampanied
1/  C.A. Fernandez, La Estructura del Credito      icola  Colombia  1I,
pp. 4l-142, estimat 2perct o       f      the credit goes to only 10 per-
c&'nt of the borrowers,
/   INCORA,   jo    eReformaSocialAraria      Informe de Ac4i.vidades en 1966,
P. 13.
,   INCORA estimate; INCORA further est4rmates that 800,000 of the country's
2 mI.llion farm families are "surplus" in the sense that the present level
of production could be maintained without ther.
).s/ INCORA,Informe de Actiividades en 12)



-26-
by a corresponding increase in labor requirsnents, and at the same time
has widened the gap in productivity between large and snall farmers, thus
putting the latter into an increasingly less favorable ccnpetitive position.
M4anuf'acturi4
The relatively rapid growth of manufacturing outnut has occurred
large.ly as part of a process of import substitution in consumer goods
under protectionist policies. The share of consmer goods in imorts
declined frcm 21 percent in L950 to about 5 percent in 1966. As a result
of the protectionist policies and the orientation toward import substitution,
most industries - with the outstandirg exception of the old and highly
developed cotton textile industry - are relatively inefficient and high-cost
producers. For this reason as well as becauise of the maintenance until
recently of overvalued exchange rates, little has been accomplished by way
of expansion of exports of manufactures. In 1963-1965 such aports averaged
only about l percent of total exports. Great anphasis is, however, now
being placed on the expansion of exorts of mafactured go      . Te policy
instruments being used to this end include the recent move toward more
favorable exchange treatment, discussed in greater detail below, negotiations
for the broadening of markets through the fomnaaioon of regionil and sub-
regional more highly integrated economic blocs within Latin Amrerica, expansion
of credit availabilities, and public-private consultations aimed at increasing
industrial efficiency.
The expansion of indistrial output has not beenp nutched by a
corresponding increase in industrial stployment. The proportion of total
employment ijn manufacturing industries declined frm 15 percent in 1960
to 13 percant in 19643/. Between the two cenius periods, 1951 and 1964,
a/ F.H. Cardoso and J.L. Reyna, "Industriali~acion5, Estructura Ocupacional
y Estratificacion Social en America Latina", mimeo, August 23, 1966,
Santiago, Chile, p  1.3; D6partamento AdMinistrativc Na-ional de Estadist. ca,
(DANE),        of P          Jul ,        ogc,ta.  I i



-27-
ihdustry was able to create only 196,000 now jobs, as ccapared to the
current anniual increase in the labor force of about 190,000 persons.
In the last five years the industrial labor force has increased,
on the average, on.ty about 1 percent, per aMaum. Enploynment in several
manufacturing activities has actually declined, e.g., tobacco products,
,shoes, iAood products and furniture, rubbor products and chemical prod cts-
In other words, iniustrial employment has tailed to keep up with the population
increase; even less has it been able to reduce the backlog of uneployed
work3rs and absorb the influx of labor from the countryside.
The lag in industrial eployment relative to increased output
Of manufactures reflects a trend toward capital intensity in industry.
This has beon responsive to overvalut,¶ exchange rates which have cheapened
the importation of capital goods, to the problems associated with the
managemient of personnel and with the r equiremients of labor legislation,
to the transplantation of capital-i.ntensive techniques fran industrial
countries,, and to interest rates which have been low relative to the
2/
degree of inflation- e The capital-intensive nature of manufacturing
coupled with a low degree of capacity utilization means that it would be
possible for output to expand steadily for a number of years without a proportion-
ato effect on the level of employment.
1/ DANE, Boletin Mensual de Estad�stica occasionally publishes data on
employment for about 20 manufacturing incdistries.
/ The last-named factor is beginning to be realized by Colombian authorities.
Recently a measure was adopted to increase credit to industry through
the use of reserves of the Colombian Social Security Institute.  Constant
value bonds are issued in exchange for the reserves. A part, of the
reserves is channeled through the goverrment-owned Industrial Development
Institute (IFI). To compensate for inflation, IFI charges interest of
15 percent per anmnm on loans out of these reserves, of which 8 percentage
points are an adjustment for inflation.



-28-
?~uation (3rowth and, L1olMent
In recent years the rate of growth of populaeion has been about
3.2-3.3 percent per aaruml. The rate of increase still appaer to be
accelerating and instantaneous rates of 3.5 percent are sometimes citedi'.
Moreover, in cosnnction with the problem of achieving an adequate grcmth
of income per oe;ta,, the age structure of the population is extremely
unfavorable, i.e.,, thwe is a high ratio of dependents to members of the
aotive labor fore>.
The high rate of population growth creates serious problems of
giloyment. As regards agriculture., the eployment problems can be
attributed in large part, first to the slow rate of growth of ag)cultural
output relative to the increase of the rural populat.0on and, teond, to the
inadequate temre sytem which encourages labor-ertonsive tsque       of
Droduction. The population increase occurs predominantly in the "nminifundian
areas where population density Already is extremely high. 14l1yment opport-
unities are strictly limited in the vast "latifundia" areas. Thus the
growing rural population leads, first, to incrwead fragmentat±on of holdings
in the "minifundia" areas, second, to an increase in the landless rural
proletariat and the migrant labor force. and third, to incraeod rural
undereoployment.
Studies qf the Roldanillo-La Union are  in the Valle del Cauca
pravide smne indication of the extent of undere.loyment on mallholdLngs.
I/ For example, in a Mliistry of Health dozument, "'Caracter~sticas Generale.
de la Poblapidn Colmbiana",, by H. YAndoza Hoyos, mimeowp Botota, August 9,
1967,9 Pole.
2/ According to tk4e population census of 1964, 44 percent of the population
is under 15J yer of age and 5 percent is above 65,
-   +Aa x w E. sriNk.i1 A  .i4 -  L ,gWLbi@iSNWsiSSl'frt''t4~>  ma '"-CaifN172<'sz:'/ia  v*'=ul1>ws-4/<ss  +



-29-
Surveys inLcate that in this re.gion only 1e.. percenit of the available
family labor can be eMployed on farms df 1.6-'?5 "plazas", and 42 per-
cent could be used on farms between 7.6 and 15 "plazas"-/.
The declining employment opportunities in agriculture have led
to substantial migration to urban azreas. BetWeen 1951 and 1964 the
fifteen major cities with a population greater than 100,000 persons
increased at an average rate of 5.15 percent per year. This is very
rapid even by Latin American standards.
Tnble 11
Growth of Major Cities, 1951-I1964
(percent per annum)
Armenia          6.69        Ibague        7.18
Barranquilla     6.52        Manizales     6.66
Bogota            3.76       Medellin      4.98
Cali             4.68        Pasto        10.30
In many cases, those migrating to the cities became openly
unenployed2/; more often, they entered service activities and increased
the amiount of underemployment in th.at sect.ar. Only a minority was able
to frin( employment in industry.
1/  CIDA, op. cit. p. 156; one "plaza" = 0.64 hectare.
2/ The Universidad de los Andes undertakes periodic sample surveys of
unemployment in Bogota ("Resultado de las encuestas sobre desempleo
en Bogota", mimeo). Results for recent years are as follows:
1963      7.9%
1964      7.3%
1965      8.9%
sept.  1965      9.7%
1966     10.1%
Sept.  1966      9.6%
Sept. 1967      10.6%
-                                      25. i t  . .   iA 4 4  . . .  S i l!d ! E2  2 i2 , ,b2s,jt4l1,.2t:t ........................ b 4<1'*,1-- .,A:;. t ..............9E>9,.:> ........ -,A t0 At... w&,r ... zX4 . 2 4 e:. .  .v  . .a . . . . . Q



Coffee and Goverrnment Revenues
As explained in Chapter III, internal coffee prices in
Colombia are kept well below wprld market prices by a systbt 0bf
taxation that amounts to about half of the external price. A large
part qf these tax proceeds, however, is used to finance the retention
of coffee as well as investnents and services in coffee region-s carried
out through the National Coffee Growbrs' Pederatibn. The portSUbf
retained by the Certral Government consists priatily of all but 4 per-
centage points o1f the export tax on coffee (only bne of the  %*ra1 taxes
on coffee); the Central Government portion thus amounted to 19 percent
of the value of exports in December 1967 and is scheduled to de61ine
to 16 percent by December 1968. The export tax, addpted inA m&6 1967
at the time the general system of exchange rate6 wa's changed, Jr'llaced
an earlier system of a penalty  exchange rate fo± coffee whi£h -b ilded
tax proceeds through a differential between that rat6 and iV*i't rates.
The proceeds of the itfferential were used in ptrt to subsidiAe dcbktain
imports, to defray sme expenditures of the Cential Bank, and ta Yray
sane debt of the Central Government to the Central Bank. ThesX dcpen1d-
itures were ilot included in th'3 curresit operatitLg budgets of the tentral
Goverrnent, and the proceeds that were arailable ror general governmental
expenditures constitute  only a part of the totala This accoufts for
the substantial variations in the amounts of tho dbffee diffe?ential
proceeds that appear 4n the current operating accounts of thd Cehtral
Government., ranging frcwi 17 percent of tot4 curlrent account itflonvi in
1963 to 6 percent ir) 1966, By the same token, -he t'atio of thSe
proceeds to investment expenditures of the Central Goverrment aibb
ranged widely,T fomn as much as 49 percent in 1963 to 19 pqrcent in 1966t
IM.,.-.  .  .L....



It is understood that in the future the antire proceeds of the
export tax on coffee allocated to the CentralGovernment will be available
for general expenditures. On this basis, Central Goverrmient revenues
from that tax are estimated to amount to 8-10 percent of its total current
account incone, or somewhat more than the share of coffee cultivEttioiA in GDP.
The sIgnificance of the revenue from the coffee tax is greater in relation
to investment expenditures of the Central Government (about a fourth) or
to  total public invest1nent (about 15 percent). Total public investment,
however, accounts for only slightly more than a third of total gross fixed
investment in the economy, which is estimated to amourAt to about 17-20 per-
cent of GDP, so that the role of coffee taxation in the financing of total
investment is quite limited. As stated at the outset, it is as a foeign
exchaiige earner that coffee plays the most crucial role in the economy.
Coffee and the Excternal Sector
Although the value of coffee exports irn relation to total
exports has declined during the postwarperlod, primarily becauise of the
combination of a slow growth in volume and a decline in prices particularly
duiring the last 10-15 years, it stlll accounts for close to two-thirds of
the total. The ratio rose from about 77 percent in 1948-1950 to a peak of
about 82 percent in 1954-1956, and by 1964-1966 it had declined to some
65,6 percent. Thus, exports still largely depend on coffee, and imports -
in spite of considerable foreign aid - still must be financed largely from
export earnings. Increased investment depends in turn upon increasing
imports of capital goods, and growth depends upon the rate of accumulation
of capital.  Furthermore, increased output is to a large extent dependent
on increased inmports of raw material.s and intermediate goods. Thus, economic
development in Coloiabia, as in many other less developed countries, is
highly sensitive to the trend of exports.
- - - - - - - v l v g Sa e A s Ll h S S @ ar S : s rk b S r ; sl l ; E Ro i - s a bA c  ... -2..2  -- -  >  :29;- & t  *> --P$----  ¢



Frcii 1958 to 1966 exports rose by only 11 percent., frcau $497.1
million in 1958 to $551.5 million in 1966,, or less than 1.4 percent per a=nm,
This means that exports per capita declined by 1.8 percent'a year, and the
consequence was a decline in the per capita availability off imp orts.
During the pEt"iod 1961~-1966 Colombia ran persistent balance of
payments defioits on total current account, averaging about $l35 million
per y-ear.. These deficits were financed by an inflow of capitaI and by
utilization of ifiternational rederVes. Mu~ch larger deficits., a'mounting to
an anmnal average of' about double'~the average deficits of 1961-1966,
or slightly more'than $270 million., are projected for 1967-1,910.
These balance of pa~yments projecti-ons were rtiad?e to s'upport ithe
goal of 6 percent annual growth of CIDP. The ilo-ots were p~ro'jected
on the basis of the historical imported c*np6ri is of the principal macro-
economic variables, such as the ratio 'of imports of' interiviidiat6 goods to GD?
and of imports of oepital goods to gross fixed investment. The' projections
of exports asm=6n a considerable -effor~t 'of promotion of wi&iA-b 6ther than
coffee -and the maint~eftince of an ex6h6hge rate poli'cy that Vill ~ioduce,
ra;;es thau equld be conisidered as "lequilibri~uml rae. Coff*b exports were
projected at an annrual voltume growth :~ate' of 2.5~ percent and at the average
1967 New York price. It is thus obv-iouis that, lunder the m~ost favorable
cireunstances, Oplombia will be dlependent on a p,onisiderably stepped uip
volume of capital impdrts to maintain a satisfactory rate of economic growth.
Exchange rate policies designed to stAimulate exports have beeni
introduiced. The sytstm adopted in March 1967 is a step toward a unified
exchange rate system, On that date two exchar*6 'Mi"kets were e.gtablished:
an exchange certiftoate market and a capital market.
The capita~L market rate is fixed at 16.25.-16.30 pesos' per U.S.
cdolar and applies to trade in invisibles and capital transactions.s



-33-
The exchange certificate rate, however, is a fluctuating one.. It
started at 13.50 pesos per U.S. dollar in A*iil 1967, and by the end
of November 1967 had moved to 15.70 pesos.: It applies to mrw.jor imports.
Exports other than coffee, petroleum and cattle hides also qre traded
at the certificate rate,, and in addition, receive a 15 percent subsidy.
Coffee is traded at the certificate rate less the export tax.
This is clearly an attempt to ilncrease the volume of minor
exports. Minor exports have, in any case, been rising rapidly in the last
few years, viz; fram $70 million in 1960 to $128 million in 1966,, or from
14.0 percent of total exports to 23.2 percent. Moreover, studies have shown
that the minor exFI)orts are quite responsive to variations in the exchange
rate, provided devaluation does not result in rapid inflationr . Thus,
one can anticipate that the combinatJ. on of a flexible exchange rate and
a 15 percent subsidy should lead to a fxurther rise in these exports. The
Government's target is a value of $300 million for minor exports by 1970,
compared to a projected value of $340 million for coffee exports.
/   J. Sheahan and S. Clark, "The 1Response of Colombian Exports to
V:ariat:Lons in EffectiveExchange Raitesl", Research Menorandwu No. 11,
Williams College, June 1967, mimeo.



CHAPER III
STRUCTURE MI PROSPECTS OF TIM COIDM'WIAN COFFEESECTOR
frends of Over-all Production,.     d--ra,gTe
The available data on Colombias cofree production show a
current level of oetput of close to 8 million bags of greep coffee per
year. The long-te iM growth of output. to this level frba  the early 1930's,
when output amounted Po about 3., nmillion bags., has proceeded at an annual
rate of approximately 2.3 peroent~. However, there appears to have been
virtually no chance,in the our^r*t ;evel during the last dqcsde. There was
a sharp increase in the level of output from the early 193Q's through the
Second World War. COtput in 19647 was approximately 6.2 million bags,
ccpared to the 3.5 million bag level of the eprly thirtiesl. Thereupon,
a new plateau appears to have beena maintwt4inqO for the aucee%Ung ten years,
after which another jump seems to have occurred to the current plateav
which has prewailed for the last tan years.
The basic series for prodpotion data in Colombia is t~hat published
by the Coffee Federation. The Fedqra#.on figures are arriye at es:tential)y
as a summation of exports, domestic consumption, and stock accumitlations.
Long-term series on' olombia'u coffee production are also published by the
U.S. Department of Agriculture.  The latter seies diffmre frcx the
Federation's series for certain individual years, but it beprs out the
observations ionerning the trends of output g4ich have been noted above.
Fcr cert4n indiv.idual years, there .re still other data available,
based on special investigations which have been carried out. In 1955/56,
a joint UN tconomi6 Commission for Latin Amner1p#/A0 coffee siwvey
was carried out in Colcmnbia, and the production figure il. yiploed
for -that year was 6.1 million bags, or about midway between th,



5.6 mi.Aion bag estimate of the Federation and the 6.8 million bag
estimiate of     the U.S. Department of Agriculture,          In 1960
the Statistical Department of the National Go4ernve6nt of Colombia carried
out an agricultural census, and the figure it yieelded for coffee output
was 6.8'millioni bags, or substant'iLly less than the 7.8 million bags
est,imated by the Federation and the 8 millLon bags estimated by the
U.S. Department of Agriculture.          In 1965 the Statistical
Departmiient; carried out a sample survey of Colombian agriculture, and the
figure for coffee output was once again well. belo* the o&her estimates:
6.1 million bags compared to the 7.9 million bag estimate of the
Federationi and the 7.6 million bag estimate of the i.S. Department or
Agricultuire.             These varying estimates for individual years
provide no basis for puttiing into question the observations concerning
long-term trends since they provide neither canparability nor continuity.
E3lements that conceivably might be omitted fram the Coffee
Federation's production series, which is essentially based on output
reaching commercial market channels, couldtinclude consumption 3n
c(offee farms themselves and amounts of crops not harvested but simply
left orn the trees or lost in the field. Coffee Federation technicians
estimate  that unharvested cof1fee together with iunregistered local
(onl:urption Or Varn f'aini ies may amoiinb to up to 5 percent of the registered
orop in any one year. This would mean approximately t,00,O0O bags at the
c.nrront level of output, consisting perhaps of 150,00O bags of unregistered
fann consumption I, plus unharvested coffee to the extent of 200,000-
300,000 bags. This perhaps provides some measure of the extent to which
1/  By 2. 5 mLllion persons at a level. of per capita consuxiption of 3.8
ki Los per year, i.e., 9,5 million kilos which., divided by 60, equals
).58,333 bags.



-36-
Table 12
Coffee Production 1     -1    67 Selected 1ears
(thousent of 60-kilo bags of green coffee)
Federation         U SDA
1930/13            3,374          3,367
1936/37            4,369          4,197
1946/47            6,158          n. a
1950151            5,038          n.a.
1951/52            6, 711         6,711
1952/53            6,Jo5          6,405
1953/54            6,719          7,083
1954/55-          6, 285          6,405
1955/.5156         5,             6,800
1956/57           6,086           6,5oo
1957/58           7,806           7,800
1958/59           7,h42           7,700
1959/60            7,684         8,000
190/601            7,500          7,700
1961/62           8,035           7,800
1962/63            7,500          7,500
196 3/64           7,800          8,200
196W/65            7,900          7,600
1965/66            8,200          8,200
1966/67           7,,800          7,600
Sources Appendix Table III.



-3.7-
outou  couLcld be deliberatoly expanded in the short run il response
to Lncentives, though these coffees are probabLy of a relatively low
quality. Under the best of circumstances, the margin for cuch short-run
ex,ansion thus see.ns to be quite limrited.
The official Federation data show a total area planted to
coffee of some 810,000 ha. in 1966/67, a level which has remained
unchanged during the last five years. Indeed, over a scunewhat longer
period, there is reported to have been a decline i.n acreage, fran
840,000 ha. in 1958/59. There was nevertheless a reported increase of
about 8 percent in acreage between 195/156, the year of the UN ECLA/FAO
survey, when it was reported as 777,000.ha., and 1958/59.
Table 13
Area Planted to Cofee2,   /   l561,16-7, Selected Years
(thousands of hectares)
1955/56           776.8
1958/59           840e0
1959,/60          836.0
1960/61           831.5
1961/62           824.1
1962/63           810.0
1963/64           813.l
196LV65           812.0
1965/66           811.4
1966/67           810.6
Source: Coffee 2ederation
It may be noted ths t the reliability of data on planted area
is subject to even greater doubt than that, on output, which at least
can be built up from the somewlat more easily measurable data on
commercial marketings and stocks. Indeed, area data from the censuses
i.nd ,l iirveys o C the National G(overnment's 'Statistical Department are
seriousl at variance with the data resported by the Federation.



Thus, the area figure from the 1960 census of the Statistical
Department is 968,000 ha., comared to 832,000 ha. reported by
the Federation for d960/61; and the area figure from the Statistical
Department's sample survey of 1965 is 965,000 ha., campared to the
Federation figure of 811,000 ha. for 1965/66.
One -. ason for the difference betlween the Statistical Department
and Federation estimates of planted area may be more adequate coverage
by the Statistical Department. The Federation data are based on the
1955/56 master sample designed for purposes of the ECLA/FAO survey,
whereas the Statisttcal Department samle base has been revised more
recently. The official Federation figures may, therefore, not cover
some areas of reoent planting. Recent limited field cheokt by lederation1
personnel tend to a*nfirm the Statistical Department f-igures cth planted
area, rather than the Federation's ocn publidhed figures. The plamted
coffee area in Colombia may thus be approximately 1 million hla. at present.
There appears, nevertheless,, to be little conclusive evidence
to contradict the Federation estimate of stagnant or declining area
planted to coffee during the last decade. The SCWA/FA0 survy tbound
some 85,000 ha. of new plantings for 1956. The 1960 census dots not
repoart new coffee plantings, and the recent national sanple surveys of
the Statistical Department give 56,0CO and 64,00J0 ha. of new plantirgs,
respectively, for 1964 and 1965. Even though t.he latter two figures may
be correct, it is believed that any expanded coffee plantings that took
place were in lower yielding areas of the country such as the Departments
of Magdalsna, Boyac9., and Caqueta, so thLt any increase in production
that might be forthaoming would not be proportionetely as great.
In order to provide more reliable data on coffee 4  , the Coffee
Federation has started a project involving the use of 'rial photography techniques,
which it is hoped may be completed by 1970. This would also provide the basis
for a better estimate of tree ambers,



-39-
The latest estimate of tree rnxubers by the Federation is
2,025 mdillion treea3 for 1962/63, and the fragmentary data which are
available on this subject indicate no change ih tctal trAe population
during recent years. The ECLA/FAO survey ln 1955/56 estimated the national
ligure at 1,938 million trees, and the Federation subsequently estimated
the rnwnber at 2,100 million tree;3 in 1958b9.
The latest figures available on the age compositioh of the tree
population also date frcn 1962/63 and are as follows:
Table 14
Age Conposition of Coffee Trees in 12/6
0.)              (2)                (3)                (4)
Ae      k       Yearslanted       Percentage of     Average yearly
.otal           replanti'ag rate
during each
:Zarpegod
*(3)                                        3
1- 3 years        1961-63           8.0                 2.6
4- 6   "          1958-60          11.0                 3.6
7- 9              1955-57           7.5                 2.5
10-12   "          1952-54           7.5                 2.5
13-15   "'1949-51                    6.0                 2.0
over 15 "          befcre 19419     60.0                  --
Source: Coff ee Federation
These figures show that 60 percent of the trees were over 15 years
old and could be expected to have declining yields per tree. For the
predcxninant variety of coffee tree i.n Colombia, namely the "typica", yields
incrqa.se gFradualiy up to an age of 12-15 .rears and then decrease, more or
less rapidly depending on the care given to the plantation.



No direct data are available on the yearly replanting
rate or on changes in the rate over time. If, as has been suggdsted
above, the total trw population did not change significantly frcm
1949 to 1963, it can be ir.ferred fraxn the figures cited 'oh ate
composition in 1962/63 that during the period 1949-1963 the yearly
replanting rate averaged 2.6 percent, since 40 percent of ihe trees
were replaced over a period of 15 years. In considering the data by
successive three-ye   periods since 19449, as is done in the table, there
is no- indication of any signifioant deviation frcn this average during
any of the thme-ysear perio ds.
IEt may be noted, furthermore, that there is no clear corre.spond-
ence between the nmmber of coff'ee trees and total output. There is
both a negative and positive correlation in t}he country, depending upon
the nature of techniques employed and ecological conditions, between
tree density and coffee yields per tree. Coffee plantings in the
optimum producing areas in the country average a lower tree density but
have a much higher yield per tree than those in the outlying areas.
On the other hand, the most modern cofPee farms in the optift ptoaucing
areas use a high tree density in combination with other advanced coffee
production techr.iques io achieve very hith yields per tree.
The fraVentary nature of the tdata available makes it
difficult to relat. trends in production to those in acreage and tree
numbers. One benaeimar) that may be used is dat& available ftan a
coffee census carried out in 1.932 by the Coffee Federation carared to
the most recent data cited above. Such a coiparison shows that production
over the long rtn has grown at a lower rate (2.3 percent) than either
area (3.0 percent) or tre  ntunbers (3.9 percent) indicating a de.line in
yields both per heat.re and per tree, as indicated below:
----=..........................-..-,..|. . M  ..  - -.:  --|-.- . -t-  -tv.!  -:,-!-  --.S.--  .  .-  - --.  .J-. -. - ' . -



Production        Area         Tree Nos.
g                             (thousands)
1932 CensIus                3,461           356*5            531
1966/67 data
(best estimates)          7,800         1,000.0         2,000
Annual growth rate
during the 35 years
1982-1967                  2.3%         3.0%              3.9%
Regional Distribution of' Coffee Cultivation
Coffee is grown in Colombia almost exclusively on the slopes
of the three cordilleras which cross the c6untry in a north-south direction.
The main coffee growing area is located on the slopes of the Central
and Western Cordilleras with the Departments of Caldas, Quindio and
Risaraldab/ as its center.  This central area extends into the adjacent
Departments of Antioquia, Valle and Tolima, and produces currentLy 70-75
percent of the total coffee crop of th6 country. The area is rather
homogeneous in climatic and soil conditions.
Much less important coffee growing areas are located in the
southern part of the Western Cordillera in the Departments of Narino and
Cauca and on the slopes of the Eastern Cordillera in the Departments of
Cundinamarca, Huila, Boyaca, Santander and Norte de Santander. An
isolated coffee growing area is situated on the slopes of the Sierra
Nevada de Santa Marta in the Department of Magdalena. More reoently
coffee has been grown to a small extent in the Department of Meta (see
map attached).
Coffee was introduced'into Golombia from Venezuela during the
middle of the 18th century. It was first. cultivated on a canmercial scale
in Norte de Santander in the beginning of the 19th century and fran there
1/ In 1966 the Department of Caldas was divided into these three depart-
ments.



42-                                            ,
i   tter ip to be produced by 1¶ripartite Coffee Study: date and legend to be added) t
-A                        . .   -I. ....- .....,..-
-                -       -    ,,,,,,-       \  STRIUUCION  PORCENTUAL  O A
< i                       i-  >      - v RODUCCION  NACIONAL DE CAFb
POR DEPARTAMENTrS           :
ANTIO  IA          I 8, 4 6
"          *OYACA                     /   i,
K  ~                  CALOAS
00ISARALD                       3 O,
.... .  L  4  /       '     /   \    OUINDlO    I/
C AUAG0AIUe                 U C A             3,0 a
CUNOINAMAC          ,9S
'   ) J                UILA    /   \      359
SUCEN~                      ' s,MAGDALEN         1 4 4
CqQ'00A                            N A R I N  O        1 I 9
- -                                     N C S     ER         3, 1 '
T . 0S -.;,L: I                              AN -1           , .. 4 .+ . -- .XH\9
,,             ,,' /  /  "'Ifif- ,          .
IA                                         I-.
(vJRfGI  CAFETR  PRDUC  IOb  .   .               r
/  /
\ YN                                               OOC POR          1
\4NOSA CAETRO 1906  M6
V                                         A              I .6,I  4
9,2  2 . 6 as .0ooo                                          r AP l:  *
T              A
LA A          'AAr                    VA  ILA
ss2c                  ( - g  s00. 000u |rYA                     I 0 P0,0R0'
A U C A                                                L,~
C H ~   ~      .......e ......
I  c)~  "A                         C     ATA
P L  *                         A U T U Mo  A I
/.0                                                 U     R 'A's I L A
CENAMUCAO PO,Hli7EA                   Z0id'
fr-Kr200  Si$ I'
1 AR,NO-                      N 12--000.*'*:0-         '---- .
tso  1.400 ... ..00. ...   ..    ...    ...    ..
4944  I 6  40 10  0  7



-43-
it spread south along the cordilleras.
Available data on the regLonal phttern of coffee production
f'or various years are sunmmarized in the table below.
T able 15
RPattern of Coffee Production
(1)         (2)        (3)           (4)              (5)
Census     ECLA/F.A0   Cens'as    Sample Srvey     Federation
195515,      1960          1965      1966/67 Estimate
(percent of total)
Antioquia      17.9       16.2        15.0         14.0            17
Boyaca          0.7        0.8         2.5          1/5             1
Caldas         29.1       31.8        28.8         35.6            3La/
Cauca           1.6        2.6         4.8          4.4             3
Cundinamarca   11.7        7.8         8.0          5.3             8
Huila           1.5        4.2         2.9           4.9            4
Magdalena       0.6        1.3         1.3          4.1             1
NJariSnio       0,5        0.7         2.0          0.5             1
Santander N     7.8        2.8         4.4          2.,7           .3
Santander S     4.3        2.3         3.8          4.1             2
Tolima         13.0       15.2        12.4          9.7            16
Valle          10.3       13.6        13.5         12.4            13
Other areas     1.0        0.7         0.6          0.8
Colombia      100.0      100.0       100.0        100.0           100
g/ Caldas, Risaralda, and Quindio combined.
Sources:  (1) International Institute of Agriculture, The World's Coffe,
(1947), p. 221.
(2) UN and'FAO,        in Latin America, Vol. I, (1958), p. 23,
(3) Departmento Administrativo Nacional de Estadistica (DANE),
Directorj-o Nacional. de,Eclotaciones Agroecuarias (Censo
Agropecuario - 1960).
(4) DANW,      sta_ruariaNacion2L!16
(5) Coffee Federation.
Although the reliability of the data in Table35 is questionable,
they give some indications of a long-term pattern in coffee development.
The center of production is moving increasingly toward the relatively new
and ecologically well-suited central western coffee area and away from



the eroded and markinally too dry regions in the est and northeast.
The Caldas-Valle area, in particular, seems to ha"e
strengthened its relative position in the past 35 years, increasing
its proportion frmri some 40 percent to close to 50 percent of the
national crop. On the other hand, there seem to have been consistent
declines in older producing areas such as in the two Santgnder depart-
ments and in Cundinamarca. The 1932 weght of the latter three areas
of almost one-fourth of total output declined to about one-eighth by
the mid-1960's. * mall but significant production gains are shcin by
the Departments of Boyaca, Cauca, Huila, Magdalena and Meta (the
latter is not shown separately) which, together, increased their share
fran about 5 percent of total production in 1952 to 10-15 per-ont in
recent years. The long-tem decline in the proportion reported for
Tolima through 1965 has a special explanation., in thalt it it probably
related to the political unrest which plagued this region for many
years. The latest reports indicate a substantial recovery of coffee
growing in Tolima. These shifts signify roughly a displacemnt of
coffee growing in Colombia from the east of the country to th West
and also from the north to the south.
In very general terms it can be said that the centrol coffee
growing areas have the best natural conditions for coffee grt*ing.  They
have good soils,grainfall is above 2,000 mm and is seasonally well
distributed. To tNh rwth and south of this areao climatic .c.ditions
become less favorable mainly due to the seasonal rainfall distribution.
This is documented by the very low yields in Santander and Norte de
Santander in the north and Cauca and Narino in the south.
Within the central coffee growing area, vcological marginality
for coffee growing is determined mainly by altitude. Elevations fr.an
1,300 to 1,800 m are most favorable for tbe produotigon -of orSd quality



coffee.  'lhis optiLal altitude belt often coindidebi with the steep
slopes of the Andes and many of the plantations are on steep land
ofteri with gra'dients up to 60, percent.
Coffee yields by departments are mumtarized in the table
below for 1955/56 and 1.965. 'rhese data should be taken as illustrative,
and do not represent any significant changes over time since the bases
of their calculation are not comparable.
Table 16
Coffee Yields byDepartmnents, 1925/56 and 1965
(kilos of green coffee equivalent per hM.)
1955/56        1965
North
Magdalena                  360           580
Norte de Santande          377           288
2,antander                 411           328
Cen-ter
Antioquia                  565           472
Boyaca                     309           511
Caldas                     680           738
Cundinlamarca              330           447
Tolima                     517           337
Valle                      621           593
South
Cauca                      420           376
Huila                      397           550
Narieno                    1'91          254
'0ources: E'C,LA/FAO Coffee Survey for 195/56;DANE, Encuesta Agro-
pecuaria Nacional, for 1965.
.. .. . .. .   .   ....



-46-
Conditions of Coffee Cultivation
The physical conditions of coffee cultivation in Colombia are
favorable, and lend themselves to potentially large increases in output
over the long run, as well as to considerable stability in the output
from year to year. Though the best coffee soils are those which are
dominant in the central coffee area (except perhaps Tolirna), soils
are nowhere a limiting factor in the entire coffee growing area.,which
stretches from north to south over sane 1,000 kilometers. The
location of the coffee growing area, from latitude 1  north to 110 north,
produces a pattern of rainfall distribution which reslts in two harvests
of coffee during the year in most of the regions and consequently a
harvesting of coffep virtually all year round. As indicated in Table 17*
the main harvests in Antioquia and Caldas *ewr during the last quarter
of the calendar year, and in Quindio, Tolima and Valle in the second quarter
of the year, with the secorndry hsrsts anDroximately the "everse in
these two grm*s of departments. The pattern of climatic
conditions is  alsQ responsible tor the very small fluctuations in
yearly coffee production; generally when weather conditimu are
unfavorable in one area, they are favorable in others, and ttotal
prodyction tends t9 be stabilized. During the ten-year ptriod ended
in tl,.e coffee year 1966/67, the largest crop was oily about 10 percent
greater than the smallest one (see Tablel2),,with only negligible changes
in the area planted to coffee. In Baz:Ll, for a comparable period of
time, the laageat crop was 150 percent higher than the smallest one.
The normal planting distance on coffee farms is 3.2 x 3.2
meters, wbich resuXts,in a density of abaut 19000 trees per ha.



-47-
Table 17
Coffee Harvest Calendar
Secondary
Departn,ents           Main har'vest                harvest
North:
Magdalena          Novembelr-Jal2uary
Santander          Sept mber-Dec ember
Center:
Antioquia          October-Jarnii*ry            March-June
Caldas             October-Dec ember            April-June
Quindio            March-Iay                    October-December
Tolima             March-June                   September-January
Valle              March-May                    September-October
South:
Nari!nio           May-July                     January



Maintenance consists usually of two to three weedings per year.
The plantations are rather heavily shaded. Guano (Inae1.5)
serves as permanent shade. The shade has to be constantly adapted
through removal and replanting and pruning of shade trees. Most
coffee plantations are interplanted with plantain which provides
additional shade, trash for imulching, and an additionaL crop. The
use of cover crops for erosion control and -green manure is not caonnmon.
Formation and pruning of th6 tree con8ists of cutting the ybung tree
at a height of about 1.6 m "which le.ads to an umbrella 'sh   All
suckers a,e regularly eliminated and on oldi' trees ths d   wood is
renoved every year.
The small amount of manure availablle in the fam  of canpost
fran coffee pulp and other soarces is givbh to newly plart*6 trees qr
used in the nursery only. It is 'eatitnated that about onei4ghth of all
coffee plantigs receives some kind of fertilization, but i&Ost of it is
fran the organic sources indicated. The use of chemical fertilizers is
very limited, because of the lack of' working capital and Skilled manage-
ment among small faMers, probably reaching'less than 3 pedceant of the
planted area. Colombia has no corfee pest or disease problems of
econaoic significance and the exponges for plant protectiolr in coffee
plantations are intgnificant.
These traditional cultural practices, simple and highly
labor-intensive, are still in use bn the great majority ot coffee farms
in Colombia. The Coffee Federation has developed cultural methods
which give substantially higher 3tields. Theie Are discusSed ifl a
subsequent seetion dealing with the inputs* costs, and r#turns of coffee
cultivation in



-49-
the country. The traditional methods give a zlatibnal average yield
estimated for 1966/67 at 577 kilos of' green coffee per hectare,
based on Federation estimates of total output and acreage (see Tables
12 and 13. The yield would be lower (468 kilos) if the suggestion
noted above that total coffee acreage amounts to perhaps.1 nmillion
hectares is correct. In any case, the gerneral order of magnitude of
yield appears to be in the neighborhood of abdut 500 kilos per hectare.
Fitting in with these characteristics of coffee cultivation -
high labor-intensity and low, ai.;^d not increasing, yields - is the
pattern of size of coffee farms- Coffee production in Colombia has to
a large extent always been the activity of small family-operated and
family-owned farms. Coffee cultivation he.s characteristically been
externded throughout the country through larid settlenent by small farmers.
Wherever climate seemed to permit it, coffee was planted as the firs-t,
cash cron in the place of original forests. Under various laws the
settlers could claim ownership of the lands thus inproved and, even
today, one of the important activities of INCORA,     the agrArian
reform agency, consists of the definitive allocation of title to
agricultural, including cof.fee, squatters.  The Colombian coffee farmer
has traditionally lived on the land he works, and,' considering t}e
pattern of settlement, the size of coffee faxns has to a considerable
extent remained small, determined by the working capacity of the farmer
and hi,s I'amLly.
Only inadequate data on the di.stribixtion of coffee farms by size
throqghout the country exist, and these are presented in Table 18, Thn
Coffee Federation estimates tVhat there were sacme 300,000 coffee farms in
195/66, and estimates the aver&. size per farm to be u.der 3 hectares.
As mary as one-third of the total lnumber were farms of less than 1 hectare each.



-50-
The number of such very small farms has increased over the last decade
at the same.rate as the total number of coffee farms, despite the
injunction since the adoption of the current agrarian reform law in
1962 against the subdivision of Ifarms of less than 3 ,hectares. As
indicated in the table, there has apparently beern s(one decrease in the
average size .of 1coffee farms over this period.
T-A-be 3,T
Distribution of Coffee Fa:ms ~brSige, National Totals
Less than 1 ha.  mno.. of farmas          77 ,26 41;09, 7-0
1 - 5O ha.      -       "135,148                     -19 ,1 ?17
Over 5O ha.             "577
Total           -                        212,970      301,530
Total area -pr.anted to coffee (0(.0 Iha.)   777        .en
Average sise Per fam   (ha.)                3.65         2.69
Source3 Coffee Federation
MatsAsd data on f am size -exist Tfr tho leadiw. 2 6fee
Department of .a1das (;. e,, the 9-amer Department of Calds, encanpassing
that fDepartinent as well as the two mew Departments of Risaralda and Quindio*'*Kh,
together aci*mmted fov about one-third of *oal coffee output in the countey),,
c   .np,.led remnU1  y t ie Pevelopment Pr ,gra  for Caldas. The conditions
prevailing in Caldas are similar to those in the neighboring important
coffee Deparbents of Antioquia,, Valle, and Tolima so that these data can be
consid"ved as represeetat1ve for -the central coffee growing area as a whole4
AsI ndivated in Table 19,half of the coffee fams in that area
had an averav size of 3.5 ha. or less and as many as 90 pecent of the



total had an average area of 15 ha. or les. The farms up to 15 ha.
in size account far only somewhat more than 40 percent oIf the total
area of coffee farmas, but have as much as aboit tWo-thids of the area
planted to coffee on these farms. The ratio of area plahnted to coffee
to total farm area generally declines quitb regularl3t as the size of farm
increases. Smaller farms, up to 3.5 ha., devote about two-thirds of
their land to coffee; farms averaging between 4.5 ha. and 15 ha. devote
abaut half of their land to coffee;  arxd the ratio continues to decline
with the very large farms devoting only a small minority of their total
lands to coffee.
The small farmers in fact grow virtually no other crops than
coffee, not even food crops for their own consumption, except for plantain
which serves as shade for the coffee. In this respect the small coffee
growers in Colombia differ from smallholders in other coffee growing countries,
such as Guatemala, where the snallholders are essentially subsistence
farmers but graw scne coffee in addition to food crcps. Dependent as they
are on purchases of food, these small Colombian farmers are highly vulnerable
to coffee price fluctuations. On middle sized farms9 land not planted to
coffee is devoted mostly to sugar:ane, cassava, or pasture. The very
large farms are essentially livestock fanms which also grow saoe coffee.
The mallholdings are located mostly on the steepest land on
the mountain slopes. The middle sized farms, occupying the best coffee land
with respect to both topography and altitude, are to be found onr the foothills.
The largest farms are mostly in the valleys or plateaus on relati}vely flat
landi and same are either too low far good coffee production or are
situated above 1,800 meters, the upper altttude limit for coffee growing.



Table 19
Department of 0aldas   Number and Size of Coffee Farms and Area Planted to Coffee, 1965
Average size     Number     % of total      Total area     % of total     Area planted     % of total    Average area      % of farm
(ha.)           of          farms          (ha.)           area         to coffee         coffee       planted to       planted to
farms                                                      (ha.)             area          coffee          coffee
(ha.)
*.25       2,640          4.4              660           0.1              458            0,2           0.17              68
0.75       3,660          6,2           2?4S            o.4            1,759            0.7           0.48              64
1.5        7,920         134             188a0           1.s            7,67?            3.2           0.97              65
2.5        8,400         14.2           21,000           3.1           13a374            5.7           1.60              64
3.5        6,961         U1.7           24,364           3.7           17,358            7.3           2.49              71
Sub-totals      299581         49.9           60 649           9.1           40 621           17.1           1.37              67
4.5        5,221          8@8           23,495           3.5           13,340            5.7           2.56              57
7.5       12,422         20.9           93,165          14.0           50o863           21.5           4.o9              55
-5.0        6,879         11.6          103,185          15.5           47,221           20.0           6,86              46
Sub-total      2,522          41.3                           33.0 9U45                       47.2           4.54              51
25.0        2,081          3.5           52,025           7.8           18,856            8.0           9.o6              36
35.0        19172          2.0           41,020           6.1           18,511            7.8          15.80              43
45.0          ?59          0.4           11,655           1.8            3*898            1.7          15.05              33
Siub-totals        5.9                       104,70           15.7          1,265            17.'          11.75              39
75.0          942          1.6           70,650          10.5           23,806           10.3          25.27              34
150.0          392          0.7           58,8oo           8.9            9,166            3.9          23.38              16
350.0          301          0.5          105,350          15.9           7?385             3.1          24.54               7
750.0           30          0",1          22.4500          3.4            2,o74            0.9          69.13               9
1,750,0            6           -            10 -51           1.6             '215            0.l          3.83                2
2,50oo.o           5           -12 5                         1.9              200            0.1          40.00               2
'7ub-totals                     2.9          280,3Mo          42.2           42,846           18.2          25.56              15
rTrand totals   59,291        100.0          665,494         100.0         -236,156          100.0           3.98              35
Source: Developmept Program for Caldas.



The ?T--,tional Federation of  ffee Growe  and the Marketi  Structure
The National Federation of Coffee Growers is the pivotal agency
in the carrying out of the coffee policy of the country. Though it does
not have a monopoly of internal or external marketing of coffee as in the
case of the British-type mrketing bard, it approaches such status.
Through a network of agenol.os in the. o:nt'e departmnents, the Federatt.on
stan4s ready to bdiy from farmers coffee4i 4.tctil meets its specifications, with-
out limit at fixed prices.     The Federation in practice exports only
a minor portion oe' total coffee exports from the country, but it shares
in internal ourchases of cof 7rte to a muLch greater extent; much of the
coffee exported by private parties is puirchased by them from the
Federation. The Federation is the only agency which handles the storage
of surplus coffee) and is virtually the only agency which carries out
storage to any extent. The Federation is daminant in the marketing of
cofEfee for domestic consumption. I't also has been designated by the
gy'overnment to carry out Colombia ts obligation,s under the International
Coffee Agreement. In this connection, it issues certificates of origin
for coffee exports, handles the export st-uns, reports in detail to the
IC(, and, in general, supervises and regulates coffee exports within the
allotted quotas. Furthermore, the Federation has a monopoly for aUl Anxx B
(non-quota) exports , for exports under bilateral payments agreements, and
in the export of aged coffees.
The Federation was established in 1927 as a private body, rere-
.;enting the interests of all coffee producers.  It has, nevertheless, close
r elatlions with the goverrnment. Under contract with the government, besides
sAipervising exports pursuant to the provisions of t,he International Coffee
Agreement., it operates the National Coffee Fund, the device used to finance



the purchase and storage of surplus coffee. The internal price for coffee
paid by the Federation is fixed by a committee consisting of the General
M   g     of the Federation, the Manager of the Central Bank, and the Ministers
of Finance, Agriculture, and D8velopment. The top executive.body of the
Federation is an eleven-member National Committee, which meets every week.
The Committee, besides six members elected by the Coffee Congress of the
Federation, includes five govermental representatives ex ofticio: the
Ministers of Finane, Agriculture, Development, and Foreign Affairs and
the Max4agar of the government-owned agricultural bank, the Caja Agraria.
Despite its close links to the government, the Federation has
achieved a remarkable record of stability and continuity of its personnel,
outside the vagaries of the political arenLva. During its lif of sane
L40 years, it has had only two gerral managers. Its personnel corps,
consisting currently of some 2,800 persons, is operated on a career basis,
under conditions more favorable than those whiph prevail in the civil
service.
The general assembly of the Federation is a National Coffee
Congress, which normally convenes every two years. The    Gress elects
the six members of the Federation who serve on the executive National
Cpmrd,ttee., citeI abovee  The National Conmnittee names 12 dqartmental
ccumdttees, throgh which the Federation' s functions in marketing as well as
in extension work, and in the construction of infrastructure facilities,
are carried out. 1esidoes its network of agencies in the principal cities,
tcwns, and ports in, the country involvd in the marketing of coffee, the
Federation maintains offices abroad, in Brussels, Romne, Madrid) Copenhagen,
Tokyo. New York, azd Benos Aires, as well as in key embassies in other
ccuntries.



The level of the onperating budget of the Federation is close
to 175 mLillion pesos (about US $11 million); it amounts to 173.6 million
pesos for 1968. Of the total, 124.1 million pesos were budgeted in 1968
for extension work and infrastructure through the regional coffee growers'
committees. The extension service of the Federation is the largest of its
kind in Colombia, comprising a total field personnel of 512 (in 1965), in-
cluding 77 agronomists, 7 veterinarians, and 31 home economists, The infra-
structure program in the coffee regions involves the cofistr'uction'of penetration
roads, school.s, and water supply facilities. A total of 13.7 million pesos
was budgeted in 1968 for the Chinchir( experiment station of the :dration,
The sources of funds for the cperating budget of the Federation
are principally the proceeds of sales of low quality "pasilla" coffees, which
are contributed as a. tax by private exporters (6 percent of the volume of their
sales), the proceeds of 4 percentage points out of the export tax on coffee,
and contributions from the National Coffee Fund for contractual services rendered.
In addition to this operating budget, the Federation administers
the National Coffee Fund under contract with the government. Using principally
the proceeds of the coffee retention tax, currently amounting to 19 percent
of thq value of exports,and having recourse to borrowing from the CentraJ
Bank, the National Coffee Fund finances the purchase and storage of surplus
coffee. In addition, the Fund has over the years entered into other activities,
directly or indirectly related to the coffee industry. Thus the Federation,
through the Fund, fully owns a camercial bank, the Banco Cafetero; holds
80 percent of the capital of a shipping company, the Flota Grancolonbiana; and
it has substantial interests in the government agricultural bank, the Caja
Agraria, and in several of the regional industrial finance corporations
recently established in the country. Other capital interests include Cafe'
Colombia S.A. of Buenos Aires, a coffee roasting plant; and the Colombian Center
of New York, a public relations agency.



The Federation appears to have becme more active as an internal
coffee buyer over the past decade. In 1966, the Federation bought 70
percent of the crop, and in 1967 the proportion was expected to be over
60 percent. No precise infoarmation is available for the 1950's, but
it is thought that the Fedleration, on average, purchased much less than
half of the crop during that period. The maximum prdportion of Federation
purchases is *stimted at sonm 85 percent in any yea so f6.
tn ad&Itioon to private U't8Ars, coffee mark6ting cooerat.ires in
the main producing areas hAve gaif1ed some inportance in r6cent years.
It is esti*ated that almost 20 p&dPet oI primary coff'lb Wikketine? and
processing was handled by ooap&Wtes in 1967. Alth4h        ti1.- handling
only a wall part of total coff6e MPrketingb   these 6    %ttives nevertheless
account   for the vast majority of all coapi&tive miftetift activities in
the country; they are estimatod tos represight about 80 perdent of the turnover
o=' all rWket±ng oooperatives. The growth of aepswatives has been actively
stimulated by the Federation. Th    are initially fiMfthed by it and their
mangement is   upervised until t,h  oan azry on thedAM41#66  The cooperativos
halve thus far not eaged in *port peratiOns;      thdief AttiVity consists
for the most part of the purchase of coffee of below Fl'deration standard
and the upgrading of such coffee for sale to tWe Federation.
Private exporters play a preponderant role in the export prodd§si
an the average
Since the beginng of the present docado, they have hflled/same 70 percenit
of export sa3les  1 thpre is no ttend of dh6nge in this proportion, as
indicated in Table 20. The Federat&oa has handled 22-33 pelcent of coff6d
exports in the years since 1959. 0% the Vhcle, sales to the U.S. mar6t,
which cons±str of mpre uniform typedt of cdffade, are effeotdd by private
exporters, whereas the Federation pla;ys a doanant role int the more oon2pl e
and heterogewnous !urTpean    kete. The Federation's adtite tole in European



-57-
Table 20
Propertion of Coffee Exports by Pederation
and by Private Tradeors  29-1266
(percentages)
Federation      Privatt Traders
1959        23.5              76.5
1960        31.8              68,2
1961        33.0              67.0
1962        25.6              74.4
1963        25.1              74.9
1964L       22.7              77,3
1965        31.6              68.4
1966        31.8              68.2
Source: Coffee Federation
markets derives from its policy of spreading coffee sales as widely as
possible among national markets and of establshing its product in the
areas of greatest consumption growth. Furthermore, the Federation
handles all exports effected under bilateral payments agreeints and
all exports of aged coffees. Exports of coffee under bilateral payments
agreements have risen sharply during recent years, and currently amount
to close to 1 million bags per year, mostly to Spain, Finland, and East
Germny.    Similarly, the export of aged coffees, drawn from retention
stocks, is reported to amount to the considerable sum of 800,000 bags
per year, both to the United States and to certain Western European
countries.
The domestic market f9r coffee, which mounts to some 1.2
million bags per year, is controlled by the Federation to the extent
of gome three-fourths, Domestic consumption is du.rrently caosed to the
extent of 75 percent of exort quality and 25 percent of "pasilla" or
inferior grades. In order to kbep a clear cistinotion between the inferior
grades aLnd the exort grade, the former type of coffee circulates In a



-58'.
partially roasted fom. The retail price o t coffee destined for
dmaestic consumption is fixed at a level which corresponds approximately
to the Federation support price for parchment coffee, i.e., about half
of the world price.
The Federation presently owns eleven coffee hulling plants,
with a total rated two-shift capacity of approximately 1.5 million bags,
or about one-fifth of total annual hulUngs. Since the Fedration handles a
much larger share of internal purchses and a somewhat larger share even of
eports,, hulling services are contracted to a considerable extent with
privately owned mills. The Federation is in the process of *VandLng
its own coffee hilling aepacity.
The Federation is practically the only agency in Colombia which
stores coffee. In 1965 a wholly-owned subsidiary of the Fedeation was
established to manage the atorage operation, previously  arnied out directly
by the Federation. The subsidiary, Almacafe S.A., is owned jointly by
the Federation, the Banco Cafetero, and the regional coffee growers' ccmnittees.
Almacafe has warehouses in 57 different localities in the countryr,
with a total storage capacity at the end of 1966 of 6.2 million bags of green
coffee equivalent, or sOme 400,000 tons of parchnent coffee. The enter-
price's own premises account for 5.6 million bags of this capacity, and
the balance consists of rented space. The storage capacity used by Alma-
cafec rePr'esnts. clqse to half of the countryt' total available storage
spac9 for agricultural ommoditieso as shown br the following estimates
of storage capacitr at the end of 1966:
Tons
State-ownvd              152,000
Privately olwied         278,000
Almacafe
Total        8)0,0000



459.
With coffee stocks of over 5 million bags of green coffee,
Almacafe currently operates at near capacity levels. It plans to keep
adding storage capAcity to handle annually produced surpluses, which
amount currently to about 800,000 bags. Storage capacity may, however,
be used for other commodities besides coffeet)
In October 1967, the Federation inau&urated the world's first
large coffee silo, for bulk storage rather than storage in bags. This
installation, located in Medellin, has a capacity of 30,000 tons of
parchment coffee, or alternatively, of 42,000 tons of wheat or 38,000
tons of corn. Its cost, equivalent to close to US $2 million, of which
about 30 percent was far the site, is expected to be recovered in less
than 5 years by savings on bags and handling charges.
Imuts, Costs, and Returns of Coffee Cultivao
Comprehensive data on inputs, costs' and returns of coffee
cultivation in Colombia, by size of farm., location, or ecological conditions,
are not available. Some idea of the pattern of inputs and returns may,
however, be obtained from partial data supplied by the Coffee Federation
for production of coffee at different yields per hectare. These data are
summarized in Table 10.
The data summarized in the table give the situation as of
September 1967. Variations are indicated for levels of yield of green
coffee equivalent per hectare, ranging from 300 kilos to 1,500 kilos. The
national average yield, it will be recalled, is estimated to be about
500 kilos per hectare. PLeld maintenance, consisting of weedng.,
replanting and pruring, and shade management, is assumed to be constant
per hectare at the varying levels of yield.  The inputs which var with
the yield are harvesting and processing., and the application of-fertilizaer.



-60-
farvesting and processing require 1.5 mar/days for each 10 kilos of
gree  coffee.  Fertilizer starts being applied only at yields of 600
kilos, i.e., at yields above the national average. The
labor requireient for application of fertilizer cosists of 5 man/days
per application. Each application amounts to 200 kilos.
Labor inputs are translated into labor costs at the wage
rate of 17 pesos per man/day. The factor fcr the cost of fertilizer
(12-12-17) its 1,640 pesos per ton.
The unit price for calculation of revenue is the support
price paid t1 the Federation for parchment coffee meeting its
specificatims. This price was 767.50 pesos per "cargaw of 125 kilos
in September 1976 or 6,14 pesos per kilo. This price is aplied to
yields 25 percent greater than those indicated in the table for green
coffee equivalent; the ratio between units o.f parobmeit coffee and
green coffee, is 5:*4.
As can be seen from the table, the gross margin of coffee
cultivation at the national average yield of 500 kilos of green coffee
is about 1,500 pesos per hectare. These calculations make no allow4zce
far depreciation of investmnt.   On the other hand, the cost of labor,
which represents the remuneration of the farmes' own labor on the many
smal coffee farms prevalent in the Qountryp amounts to some 2,300 pesos
per hectare aid should be added to the gross margin to obtain total returns
to the farmers an    l  farms. The total qf aj)out 3,800 pesos amounts
in dollar terms to about VzS $235 at the presenr rate of exchange for 130
maVdays of labcr, or about USl,8O  per an, day. This assumes, of course
productim of coff e of Foder ion grade.
It can also be noted that within this range of yields gross
wargins incroee more than proportionately in relation to the increase
in y1lds. Thos, 4s yields incres     five-fold from 300 kilOs to



¢61-
Table 21
Inputs and Returns for Coffee Cultivation, by V;rying Yields per Hectare, September 1967
Yields in green coffee equivalent
300 kilos              40C kilos               600 kilos              900 kilos              51,50 kilos
man-days     pesos     man-days      'esos     man-days     pesos     man-das       pesos     man-daY       E808
Field maintenance                 54         918         54          918         54         918         54         918          54         918
Harvesting and processing         45         765         60       1,020          90       1,530        135       2,295         225        3,825
Application of fertilizer                -           -                -           5          85         10         170          25          425
Total labor                       99       1,683        llb       1,938        149                     199       3383          304
Cost of fertilizer                             -                       -                    328                    656                    1,640
Total costs                                1,683                  1,938                   2,861                                           6,808
Revenue                                    2,303                  3,070                   h,605                  6,0                     11,513
Gross margin                                 620                  1L132                                                                   4L705
Source: Coffee Federation



1,5OO kilos, gross margins increase almost eight-fold,fram
620 pesos to 4,705 pe6os.
The most important input by far in coffee cultivation, in
Colombia, as has ben suggested, is labor. Fron information gathered
in the central coffee growing area, it appxard that in farms with up
to 1 hectare of coffee, most of the labor is provided by the farm family.
In farms with. 1-3 hectares of coffee) most af the fiald maintenance iE;
done by the farm family, but hired labor ib used for about one-half to
two-thirds of the. hrvesting. Beyond that, tke proportiom o-f hired
labor in relatioz to family labor increases rapidl,r with the size of the
coffee plantationq.. Farms of 4 he-tares and above have to dmnd largely
on hired labor, Permanent labor is enloyed frr field miintevnce in such
cases,, bt lhoe ts naturally a large amount of seasonal labcr employed
for harvesting. In the central coffee growing area, migrant workers come
mostly from the. lo,lnds of Antioquxa,, Choco, andc Valle, Because of the
variations of the harvest seasons prevously noted, migra.zt workers find
oployment as eoffe pickers for 5-6 months of the year. Tb.     m  labor
force provides the  easonal labor requirements for cane and cotton harvests
in the Cauca Valley.
It is technically feasible to raise coffee yielcd in Colombia
substantially. The potential is great as a rsult not only of methods
developed by the Federation through its research facilities, but also
because of its Larn*  staff engaged    in extension work   The advanced
agricultural methods which have been developed involve mainly.: replacing
the "ltypical? variety by "bourbon"t or llcaturra??, which have higher yields;
inereasing the dwnsity of trees per heotare from 1,000 to 2#50O or more;
removal of shade; and heavy application of fertilizers. On a priwately
oper4ted conuercial coffee farm near MwAzales, a yield of 7,(QW kilos of
-                       -   , ,   ,.



-63-
green coffee equive1tt per hectare is obthinId, i.e., almost 15 tiaes
the national average. Coffee technicians in Colombia feel that yields
could be raised economically even to about 10,000 kilos per hectare.
A rough estimate of the return oh the 7,,000 kilos per hectare
farm near Manizales yields the very. high figure of about 66 percent on the
value of investment,and 37 percent on the gro$s revenue. The estimate
has been made on the following basis (per hectare):
Pesos
Replacement value of land and inmrovanenes        30,000
Annual production cost, including dep'reciation   34,000
Revenue, 7.,000 kilos green coffee equivalent
(8,750 kilos parchment coffee) @ 6.14 pesos
per kilo of parchmet coffee                     53,725
Margin in pesos                                   19,725
Margin as proportion of gross return                 37%
Margin as proportion of investment                   66%
It should be naoted, of caarse, that this high-yielding fam
repres8eiXs truly exccptional conditions of climate, location, technique,
and management* Such conditions do not prevail generally. ThovSh there
are no conclusive data on thie not economic returns at various ]hevals of
coffee yield, it neverthqless appr-rs that at more manageable levels
above the national average, say, up to abaut 1,500 kilos p  hectare,
it is profitable to ralse coffee yields substantially. And thiii
is the situation at the current level of farm prices, after making allowance



for the substantial degee of taxation of coffee in Colombia,which
insulates the internal price frcn the international market price so that
the internal price is only about half the international price.
The question then arises as to why there has not been a marked
trend in Colombia toward higher coffee yields per hectare. One of the
reasons is the need for capital and the risk imrnlved in changing over
to the new cultivation met;hods, It is estVated that the ccrversion
of traditional plantations to the new cultivation methods takes about
five years and involves an investment cost of about, 12,OO pesos par hectare.
Such investments are clearly beyond the capabilities of small coffee farmers,
and there is little credit available in the country at reasonable cost
for such investment, as a result of official policies which have kept
bank credit for coffee to restricted levels-/.
Furthermore, the combination of increasing labor costs in the
highly labor-intensive activity of coffee cultivation and national policips
which have limited increases in the internal price paid fo'r coffee,
has also discouraged inteisifieAtiozi of coffee cultivation. The cost of
labor at 17 pesos per nar/day as of September 1967, v.sed in the calculations
, It is estimated, on the basis of data available in the anmual and'
monthly reports of the Central Bank, that outstanding credit for coffee
cultivation of the cnmercial banks comtiixed plus the Caja Agraria
amounts to about 5 percent of their total, portfolios at the end of
1966, caiared to the share of coffee cultivation in total gross
dmxnestic product of £'-8 percent.



above, was about two and a half times as great as the averagpe cost
reported by the Federation of 6.61 pesos per man/day for 19611 the
support price of 6.14 pesos per kilo of parchment coffee paid in
September 1)67, on the other hand, was only somewhat more than one and
a half times as great as the price of 3.78 pesos per kilo in 1961. Also
as a matter of deliberate policy, the extension service of the Federation
is nht .ctively' prSmoting coffee rationalization, so as not to aggravate
the problem of surpluses which exists even under the traditional methods
of cultivation.
Formation ard Trend of Internal. Coffee Prices
In the last analysis, the price paid to coffee producers is a
CcNnc-ion of the export price less the taxation applied - which is all
impol'3ed on 'cxperters - less, of course, the costs of transport, hulling-
handling, and the margins of exporters. The principal variable under the
control of the government is taxation. The support price paid byr the
Federation is fixed in the light of these elements, except for tb" extent
to which the goverranent is prepared for the Federation to pay higher prices
than warranted by these eleaents, in order to affect favorably the inceme
position of producers as well as to sustain the exort prices at which
private exporters are induced to sell. When the government has been
prepared for the Federation to incur such losses, as occurred to a
considerable extent from 1962 through 1966, the losses have been
financed largely by credits frao the Central Banks which have an inflationary
inpact and result in a transfer of inc'me frao other sectors to the coffee
sector of the econny,
The prices received by coffee growers from private traders are
generally about 5 percent below the Federation support level for coffee of



-66-
Federation quality. The reason is that private traders offer services not
available from the,Federatort, such as credit, transport, and grading.
The present (February 1968) support price of the Federation is
847.50 pesos per "carga" of 125 kilos of parchment coffee, while private
traders pay 820 pesos for the same grade. Assuming, as has been previously
suggested, that Federation purchases account for 60 percent of total internal
purchases (about the proportion ea;ected in 197>, the weighted average
price received by growers is 8224.50 pesos per carga of parchkent coffee
of Federationgrade. An equivalent wight of green coffee-, which is
worth 25 percent more than parchment coffee, would be valued st 1,030.63
pesos, or 8.25 p..oe¢ per kilo. This is equivalent to about IUS $0.23 per
lb. or slightly more than half of the current 1New York spot price for
Colcmbian coffee of about US $0-.43 per lb. Part of the reason for the
difference, is of course, transport, handling, and hulling Costs as well
as traders' margins, but the main 'eason for the difference is the level of
taxatiorn-
The le've' of taxation of coffee exports currntly 'awunts to
slightly more than 40 percent of the export value. Taxatiorit consists of
the following caoponents:
1) An ad valoren export tax applied to the local eurrency equivalent
of the export price. This was first imposed in March 1967, when
the Colmnbian exchange system underwent a basic modification
designed to bring about a greater uniformity of exchnge rates
and to allow them to fluctuate in relation to increases in the
internal price level.
g The marketing margin within Colombia is estimated to amount to 14 percent
of the wcport price; maritime transport is estimated to amount to 5
percent of the Ner1 York price.



-67-
The export tax replaced an dxchAnge differential which
previously prevailed., between a penalty xcohange rate
applied to coffee exports and highbr rates that applied to
imports. The tax was first inmosed at the rate of 26
percent, with a provision that it kas to decline by a quarter
of 1 percentage point each month, until it reached a lewl
of 20 percent in ;Decomber 1968, at whidc rate it was to
remain fixed. At present (Lrtdry 1968), the rate is 22.25
percent.
2)  A "retention" tax of 19 percent of the volume of green
coffee exports, paid by exorters in the equivalent
volume or cash. value of parchment coffee. This tax has
as its essential purpose the financing of tbe accuimlation
of annual coffee surpluses. It is paid only by priatte
exporters, and corresponds ap7- ?-mtmtly to the volume
of the surplus, If private ewporters account for about
70 percent of total exports, i.e., if they account for
about 4 million bags of gren coffee per year, the 19
percent rate comes close to 800,000 bags per year.
3) A "pasilla" tax of 6 percent of the voliune of exports, paid
by exporters in the same volume of "pasillall or low grade
coffees. Since the value of such coffees is less than the
value of export coffee, this tax represersts correapondingly
less than 6 percent of the value of exports.
4) A tae of 25 Colombian cents per 7'0)-kilo bag of coffee
exported, which amounts at presenht levels to only a very
amall fraction, about 0.03 peentc,, of the value of exports.



-68q
5) Another small tax that arises frm the requrement that.
private wxpcrrtrs surrender exschange for coffee weports
to tiw Gentral Bank at a fixed price, which gewraly
slightly exceeds the actual market price for coffee.
The purpose of t44s provision is to prevent undrinvoicing of
coffee weports. To make up tk2 difference betwen the
aurredr price and the actual mrket price, the     xorters
ma1t go into the foreign exchange market and purcbase
dollars at the crital market rate (16.30 pesos per dollar),
which has ben higher tham the certificate rate of exchage
applicable to coffAe xports (15. ?0 pesos per dollar in
Navber 1967).
The cipnanta of taxatiwn tbu    omamued to somewhat more than
45 percent whe  an mw   sytem w.as iestiuted in Marc& 1967, *  d will.
declin6 to soms,Mt more than 3,9 pwesnt When the ad valorm Aport tax
settles to its p     ent level in Dleoobor 1968. At the       ttme that
this decline in tafation has been ocurring, the peso proces. spr dollar
of exports hav .bee Increasing under the new exchange reglm      virtue
of the depreciatiow of th  cert±fictate rate of exchange. The -noW system
started with a certificate rate of 13.S0 pesos,,per dolla;r,  d this
had increased by about 20 percent to 15.70 pesos per dollar in.November
1967,
The inrosep in the peso earnimgs per dollar of .Vorts under
-the new exchange 'egii6e has made it possible to increase tho prices paid
to coffee growrs, both by the FedexM<jon and    'private traders. It has
alsc improved the ;ftincial position oI' the Federation ar4 .go.a=quently



L/'
eliminatied the inflationary iripact of its operations  O The Federation
support price has been raised from the 1e"el 6± 715 pesos per carga
of parchment coffee which prevailed before the adoption of the new
exchange regime to 847.50 pesos by February 1968,or by some ;1& percent.
This policy has had the twin objectives of reAtoring the real valus of
earnings by coffee growers, taking into accovunt increases in the cost
2/
of living, to the level that prevailed in 1962  ; and of inducing private
eorters to sustain the export price at which thesr sell. The government
recognizes that these objectives must be balanced against the undesirable
effect of providing incentiras to growers for expanding coffee output.
The long-t&m evolution of the internal terms of trade of
coffee, on the other hand, has been more favorable to coffee, as shown
in Graph 1. Since 1930, internal wholesale prices for coffee have
advanced more than the general price index for all camnodties. StartinW
both indices on a pre-World War ZI base, coffee prices in 1966 were
appro,imately 50 percent higher than they wxuld have been if ther had
merely followed the country's general price trend, as reflected by the
2/ By the end of 1967 the Federation had been able to reduce by almost
25 percent its end-of-1966 indebtedness to the CentraliBank of
1 billion pesos, and it had also been able to red-ce somewhat its end-
of 1966 external indeb6edness of scme US $90 million.
2/ -The present Federation support price of 8h47.50 pesos per carga of parchment
coffae is about 75 percent higher than its average support price of
483.75 pesos in the coffee year 1961/62; on the other hand,9 the
national index of consumers' prices (to workers) in Decanber 1967
was 108 percent greater than the average for the calendar yea
1962.



..   .- .   .    .
'''  1;  !  ' ,(   i
I.      * ./
,  .1
.;  .   . .
I   '/
* ,l,.1  ....
I I,
I       I.>,          ,                9
,~ ~ ~ ~ ~ ~ '    ' r '' '/



cost-of-living index. The relative losses suffered by ooffee since
the mid-195'0s have not wiped out this advantage despite the fact
that. about half of the maximum advarntage reached in 1957 was lost in
subsequent price developments.
'Graph 2 shows how izrkernal prices, adjusted for increases in
the general price level, iave been searated frm the world price trend
for coffee in dollars, at least since 1940e This comarison shows a
gradually increasing "gap" between the inter4al and the externl prces
of Colombian coffee. It is particularly instaructive that the extreme
fluctuations of world coffee prices in the mid-1950's were less sharply
reflected in their dcaestic impact. In the absence of internal support,
damestic fluctuations wouldaknost certainly have been even greater than
external price fluctuations-' because of the stickiness of freight costs
and of trade marginsi/. Similarly, the sharp fall in world prices
between l956 and 1958 was absorbed in a much more moderate way on the
internal market. The most recent experience is that neither the 1963-
196L4 rise, nor the subsequent decline in world prices has been passed
on fully to the Cblombian coffee grower (it is assumed that producer
prices follow the internal wholesale price trend for coffee).
1/ If the world price were to fall from 100 to 80, and freight and
trade margins were, as is most likely,, to reiain unchanged at 20,
the total decline of 20 would affect the farmers. The lattpr
would then experience a price fall frao 80 to 60., or 25 percent
as a result of a 20 percent fall in world prices.



(e'tz copy beLng produced by Tripartite Coffee Studyr)
iC
V       'a'I
'-I,-,.l
;t C        U,  C;'
j   'I ,             'i'
,N,                    '-,
9         *5s
Q     ,l  ' Li  n                               \



It is of interest tV note the cyclical effect of price
movenents on production, depicte;d in Graph 3. Duking the period
of declining prices in the 1930's, production tended to advance
steadily, presumably because the numberous small producers wre
attempting to oompensate for the falls in pride by eaanding volume.
This is, of course, not an unusual phecnmehodn in agriculture, ahd
indeed has constituted the rationale for government action in
developed countries such as the United States to support agri-
cultural prices in an attempt to improve the terms of trade of
farmers vis-a-vis the rest of the economy. The rising prices of
the 1950's appear to have induced, with a time lag, an increase
in output to a new plateau reached toward the latter part of the
decade, a plateau which has persisted to the present. The time
lag is, of cvutse, plausible on the basis of the number of yeaxrs
before newly planted trees begin to bear. The declining prices
since the late 1950's have been accompanied by only a small growth
of production.



aed^                 -41--d m '-
7-I7 77 7                    I
[  :'I t  i A lllltt'  ;'1[!1   t  I ' 1  ,' :  : Io   '| !   [
9 t
I' tO'          liit'{. ;1 1I:I2       z
.  ;               '' .*  K ''  # ' .2 | | !tl i 'l - ' '!ttX ! j ,   , , , $ , t ,, ,
I I
K1'  i-  llSjJ,4i.l'            ;jjR'  .1  4
!' 1                                '7£ 1   ; I i ! '' |  l ..
|~ . !  j .,.it,j                        1h's.  | i * ' ;l §eg r  : |I
I_ -                          J: i;
|-u>;i4,  1  '|'  |r W  ~~~~~~0  f t 1 || t1v~' |  t 1



The policy of aiming at some degree of stabilization of
internal coffee prices and of incomes of coffee farmers may have
constituted soae encouragemenft to prod&Action,, but the extent of
taxation and of other policies to keep the expansion of coffee
output in check such as restraints on credit, cannot be ignored.
The net effect of the two sets of factcTrs working in opposite diroctions -
policies of income stabilization on the one hand, and policies of limited
incentives on the other - has been a growth in output which has been slow
at best. It should, furthermore, be borne' in mind that the grcwth pattern
of coffee in Colombia .Js', to a considerable extents the result of natural
and structural factors, such as the predcminance of smallholders using
primitive techniques, and the absence of areas where production. could
be increased rapidly and spectacularly, as occurred in the Parana region
of Brazil.
The Coffee Adjus   nt Prioblwe
The Coffee Federation reports that it started accumulating
stocks to an appreciable extent during the coffee year 1959/60. It
opened that year, on October 1,1959, with stocks of approximately
250, 000 bags. The corresponding figure for the opening of the 1967/68
coffee year, i.e., for October 1, 1967, was 5.3 million bags. The latter
amount is said to include working stocks to the extent of 600,000 bags,
equivalent to 1 month's exports plus domestic consumption. Thus, the
suirpluis stocks said to have acc.imulated amoun# to some 4.7 million bags,
compared to the levels of annual output of about 8 million bags and of
exports of close to 6 million bags. During the 8-year period 1959/60-
1966/67, the average annual accumulation of surplus stocks therefore



amounted to about 600,000 bags. The rate of accumulation, hwever,
stepped up daring the last three years, 1964/65-1966/67, to about 950,O0
bags per year.
The reported movement of stocks is comrpared with proliction
and distribution in a coffee balance sheet presented in Table 22. The
balance sheet has been worked out on the basis of irnformation supplied
by the Federation, It may be noted that the reported addition to stocks
falls short of the apparent addition to stocks over the 8-year period,
based on a confrontation of the figures of production, dcmestic consumption,
and exports, to the extent of close to 1 million bags.
Assuming that the figures on production are correct, those
reported for damestic cow wnption and for exports must tberefore be too
low. The series on dcuestc consumption does, indeed, present some
peculiarities; no increases at all are reported for the first five years
of the pqriod, and a sharp decline i:3 shown for one of the years (1960/61).
As regards exports, the figures, of course, do nQt take into account
contraband ewports. The substantial taxation on coffee exports and the
requirement that foreign exchrange earned frc?e weports be surrendered to
the Central Bank have provided a strong incentive for such illeal ecports.
It is thought that, despite serious official efforts to control the illegal
traffic, contraband exports still amount to some 200,000 bags per year.
This would explain the discrepancy on the average between the reported
addition to stocks "dring the 8-year period and the apparent addition,
based on the figures for production, damestic consumtion, and registered
exports, of close to 1 millio   t. though there are marked peculiarities
which defy such an explanation for a number of individual years.