Citation
Note on market liberalisation and agricultural performance in Central America

Material Information

Title:
Note on market liberalisation and agricultural performance in Central America
Series Title:
Working paper ; no. 71
Creator:
Weeks, John, 1941- ( Author, Primary )
SOAS University of London. Department of Economics
University of London. School of Oriental and African Studies. Department of Economics
Place of Publication:
London
Publisher:
SOAS University of London
Publication Date:
Language:
English
Physical Description:
15 p.

Subjects

Subjects / Keywords:
Agriculture -- Economic aspects -- Central America ( lcsh )
Agricultural productivity -- Central America ( lcsh )
Produce trade -- Central America ( lcsh )
Temporal Coverage:
- 1997
Spatial Coverage:
Europe -- United Kingdom -- England -- Greater London -- London -- Camden -- Bloomsbury
Coordinates:
51.52205 x -0.129

Notes

General Note:
VIAF (name authority) : Weeks, John, 1941- : URI https://viaf.org/viaf/46818818
General Note:
Working paper (University of London. School of Oriental and African Studies. Department of Economics) ; no. 66
General Note:
Working paper (University of London. School of Oriental and African Studies. Department of Economics) ; no. 71

Record Information

Source Institution:
SOAS University of London
Holding Location:
SOAS University of London
Rights Management:
Copyright, the author and SOAS University of London
Resource Identifier:
295215 ( ALEPH )
Pam Agriculture D /740291 ( SOAS classmark )
A338.109728 /721331 ( SOAS classmark )

Downloads

This item has the following downloads:


Full Text
Department of Economics

Working Paper Series

MARKET LIBERALISATION AND AGRICULTURAL
PERFORMANCE IN CENTRAL AMERICA

by

Professor John Weeks

WORKING PAPER SERIES NO. 71

SOAS

School of Oriental and African Studies

University of London




Department of Economics
School of Oriental and African Studies
University of London

This working paper series is intended to disseminate research in progress by staff members and
associates of the Department of Economics at SOAS. If you would like to receive copies of future
papers in the series, or obtain extra copies of a particular Working Paper please write, giving full
details of your name, position and postal address, to:

The Department Secretary

Department of Economics

School of Oriental and African Studies

Thornhaugh Street

Russell Square

London WC1H0XG

Phone: 0171 323 6180 Fax: 0171 323 6277

Previous Papers in the Series (most recent titles):

43. 'Open Economy Adjustment and Transition in Vietnam'

John Weeks, September 1994

44. 'Secular and Cyclical Movements in Employment by Establishment Size in
Developing Countries'

John Weeks and Cristiana Letteri, September 1994

45. 'Liberalisation, the Exchange Rate and Latin American Agriculture since 1980'

John Weeks, September 1994

46. 'Financing Post Apartheid South Africa: Theoretical Aspects'

Laurence Harris, September 1994.

47. 'Financial Fragility in the Transition to Market Economies'

Laurence Harris, September 1994

48. 'International Financial Markets and National Transmission Mechanisms'

Laurence Harris, September 1994

49. 'Public Debt and Redistribution'

Giancarlo Marini and Pasquale Scaramozzino, September 1994

50. 'Structural Adjustment and Employment in the Middle East and North Africa'

Massoud Karshenas, November 1994

51. 'Do Other Firms Matter in Oligopolies? '

Jonathan Haskel and Pasquale Scaramozzino, December 1994

52. 'Assessing Adjustment in Africa'

Paul Mosley, Turan Subasat and John Weeks, January 1995

53. 'A note on African Growth and Exports'

John Weeks, February 1995


54. ’Uncertainty and Endogenous Selection of Macroeconomic Equilibria’

Pasquale Scaramozzino and Nir Vulkan, June 1995

55. 'Will East Germany Become a New Mezzogiorno? '

Andrea Boltho, Wendy Carlin, Pasquale Scaramozzino, July 1995

56. 'A Fuzzy Snapshot of Some Poor and Invisible Women: Farm Labourers in
South Africa'

John Sender and Deborah Johnston, November 1995

57. 'Structural Adjustment and the Prospects of the Iranian Economy'

Massoud Karshenas, November 1995

58. 'Transition and Crisis in the Japanese Financial System: An Analytical
Overview'

Costas Lapavitsas, January 1996

59. 'Social Security and Intergenerational Equity'

Giancarlo Marini and Pasquale Scaramozzino, April 1996

60. 'Money and Capitalism: The Significance of Commodity Money'

Costas Lapavitsas, May 1996

61. 'Classical Political Economy of Money and Credit'

Costas Lapavitsas, May 1996

62. 'Endogenous Growth and Social Security'

Giancarlo Marini and Pasquale Scaramozzino, May 1996

63. 'Stages of Economic Development and their Impact on the Effectiveness of
Foreign Direct Investment'

Yi Zhu, September 1996

64. 'Machinery and China’s Growth-Exports Nexus'

Die Lo and Thomas M.H. Chan, September 1996

65. 'Social Time Preference'

Giancarlo Marini and Pasquale Scaramozzino, February 1997

66. 'A Note on Market Liberalisation and Agricultural Performance in Central
America'

John Weeks, April 1997

67. 'Re-appraising China’s State-owned Industrial Enterprises'

Die Lo, April 1997

68. 'Rural Poverty & Poverty Alleviation in Mozambique: What’s Missing from the
Debate?'

Christopher Cramer and Nicola Pontara, May 1997

69. 'On the Theory of Decentralisation: A Critique of Mainstream Economics and
Towards an Alternative Political-Economy Approach'

Rathin Roy, July 1997

70. ‘Latin America and the ‘High Performing Asian Economies’: How great were
the differences and how do we account for them?’

John Weeks, July 1997


MARKET LIBERALISATION AND AGRICULTURAL
PERFORMANCE IN CENTRAL AMERICA

by

Professor John Weeks
Centre for Development Policy & Research
School of Oriental and African Studies
Thornhaugh Street
Russell Square
London WC1H0XG

August 1997

Draft: Not to be quoted without the author’s permission.




Market Liberalisation and Agricultural Performance
in Central America

John Weeks

Central for Development Policy & Research

School of Oriental & African Studies

London

Introduction

One frequently encounters the argument that trade liberalisation and
deregulation of domestic markets result in increased incentives for agriculture. If a
proposition could be made true through repetition, the hypothesis that agriculture
gains from deregulation would be beyond challenge. We consider this proposition for
the Central American countries, all of which passed through fundamental policy
change either in the 1980s or 1990s. First, the standard argument is stated. There
follows a characterisation of the policy regimes in each country over various periods.
Armed with this, we move to an inspection of agricultural trade performance. The
evidence suggests that deregulation of trade and markets has not shifted incentives to
agricultural production; or, if so, the sector has not responded as one would
anticipate. An attempt is made to account for this.

The orthodox argument views deregulation as favourable to agriculture largely
as a result of trade liberalisation. While state interventions in domestic markets may
have a negative affect on agriculture, they can also have a positive effect, through, for
example, subsidised credit and fertiliser. The net affect of domestic market
interventions on agricultural growth is an empirical question. Such is not the case for
trade policy in a neo-classical world. Economies consist of tradable and non-tradable
commodities, with the former divided between exportables and importables. If one
considers only tradables (i.e., the economy has no non-traded goods), trade policy
must either foster exports or encourage import substitution; it cannot do both.1
Protectionism, if it is effective, changes the composition of aggregate production from
what it would be under free trade. If one makes the strong assumptions that yield the

1 Liang (1992) has demonstrated that import substitution and export promotion are not
mutually exclusive categories. In a three good economy, the production of both exportables
and importables can be increased by reducing resources allocated to non-tradables.

1


Heckscher-Ohlin trade theory, free trade results in a specialisation pattern dictated by
factor endowments (comparative advantage). If, as is common, one assumes that
agriculture is more labour intensive than the economy as a whole, then protection, via
tariffs and/or non-tariff barriers, will shift resources out of agriculture.

Agricultural Performance in Central America

As a first step to investigate empirically this argument for Central America,
Table 1 summarises the evolution of policy regimes in the region. In the early 1960s,
the governments of the region agreed on a series of measures, including a common
external tariff, to create the Central American Common Market, which officially came
into operation in 1963. The 1970s brought the fullest development of the CACM, not
withstanding that Honduras abandoned the enterprise at the start of the decade. To the
extent that the four remaining Central American countries engaged in a purposeful
policy of import substitution, it occurred in the 1970s. For all of the countries except
Honduras, the period 1970-1979 was characterised by a policy framework designed to
shift the structure of aggregate production and exports towards industry. This desired
outcome was achieved, though perhaps not as dramatically as policymakers had
hoped. In 1960, manufacturing value added was fourteen percent of regional GDP,
and rose to eighteen percent in 1975; and over the same period within manufacturing,
food and textiles declined from eighty percent of total output to sixteen-one percent
(Weeks 1985, pp. 135-136). In trade, five agricultural products accounted for
seventy-seven percent of commodity exports during 1960-1964, and sixty-two percent
during 1975-1979. With the exception of Honduras, import substitution regimes ruled
in Central America during the 1970s.

By the late 1970s, the CACM was on the wane, due to domestic, regional, and
international factors. The overthrow of the Somoza regime brought a new
government to Nicaragua which was ideologically committed to a high degree of
intervention in markets. Roughly contemporaneous with this, civil war broke out in
El Salvador. The response of the Salvadorean governments was to implement policies
which were in many cases as heavily interventionist as in Nicaragua (albeit from a
different political perspective). During 1980-1985, both countries can be
characterised as ‘strongly interventionist’, especially in their trade policies. In Costa

2


Rica and Guatemala policy evolved in the opposite direction, toward deregulation and
liberalisation, and more so for the former than the latter during the first half of the
1980s. In the second half of the decade, both countries continued to deregulate, and
we characterise them as ‘liberalised’. The second half of the decade also brought
liberalising moves in El Salvador and Nicaragua, though from a considerably more
interventionist starting point. Honduras was a special case. While its policies
changed less than for any of the countries in the 1980s, Honduras began the decade
with the least intervened economy of the five. By the time it initiated purposeful and
broad deregulation in the early 1990s, perhaps all four other countries had passed it in
the race for liberalisation.

After the early 1990s, the Central American countries were overwhelmingly
deregulated/ Table 1 summarises the discussion of policy regimes, with the shaded
cells indicating a high degree of liberalisation. No country in practice achieves the
ideal of a laissez faire economy, but during the shaded periods the countries were as
close to this as any in Latin America. As noted at the outset, the orthodox hypothesis
is that liberalisation favours the agricultural sector: first, because market interventions
are typically assumed to turn the terms of trade against agriculture; second, because
the sector’s output is overwhelmingly tradable; and third, because agriculture is
presumed to be labour intensive, and the countries to be abundant in labour.
Certainly, the tradability criterion applies to Central American agriculture, where
output is almost entirely either export products2 3 or internationally traded staples.4

[Table 1 goes here]

Prior to considering the impact of liberalisation on agriculture, it is relevant to
review the performance of agriculture in the Central American countries. Table 2
shows rates of growth of agricultural value added at constant prices, with the
liberalisation periods shaded (as in all subsequent tables), and the five year periods of

2 A good survey of policy changes in Latin America is found in Bulmer-Thomas (1992).

3 The most important of these for the region were coffee, bananas, sugar, livestock, and
cotton, though the latter declined dramatically in the 1980s. In the late 1980s and 1990s non-
traditional export crops, such as fruits and market vegetables became important in Costa Rica
and Guatemala.

4 The region’s most important staple was maize, followed by rice and beans. A fourth
internationally trade crop of some importance was sorghum, used both for animal feed and
human consumption.

3


most rapid growth highlighted in bold. For all five countries, the most rapid rate of
growth was during the years of most intense implementation of the import substitution
strategy, 1970-1974, or in the previous decade when such policies were being put into
place.

We look next at overall export performance, relevant because the majority of
each country’s exports were agricultural. One anticipated outcome of the policy shift
towards a policy regime with less government intervention would be the general
fostering of exports. Table 3 provides a crude test of the impact of trade
liberalisation, by calculating the trend in constant price exports, with a dummy
variable corresponding to the liberalisation years in the first table.5 The table
confirms the null hypothesis that the shift in policy had no impact on real export
growth, with the exception of Guatemala, for which the coefficient on the shift
variable is negative (significant at the .05 level). Only for El Salvador and Nicaragua
is the shift variable of the predicted sign. These two positive (but non-significant)
coefficients give no evidence for or against the liberalisation hypothesis, since any
export improvement for these countries might be explained by the end of armed
conflict. While these Tables 2 and 3 should be taken as indicative rather than
definitive, they suggest that a need to re-assess the impact of import substitution
policies on agriculture in Central America.

[Tables 2 and 3 go here]

Turning to a detailed consideration of the performance of the agricultural
component of trade, we can summarise the hypothesis we to test: if interventions
discouraged agriculture, if the sector’s output consisted of tradables, if the sector were
labour intensive, and if the countries were abundant in labour, orthodox theory would
predict liberalisation to reveal a strong comparative advantage in agriculture. This
comparative advantage would manifest itself in an increase of net exports for the
sector. Table 4 suggests that quite the reverse was revealed. Part A of the table shows
that in absolute terms (millions of US dollars) the net exports of the agricultural sector
declined in four countries: El Salvador (from the early 1980s), Guatemala (also from

5 Since the purpose is to test the hypothesis that policy change impacted on exports, no
attempt is made for a general search for structural breaks in the data series.

4


the early 1980s), Honduras (from the second half of the 1980s), and Nicaragua (from
the 1970s onwards). Only in Costa Rica does one observe an increase (in each
period).

A more relevant measure is presented in the second part of the table. The ideal
relative calculation would be net exports as a proportion of the gross output of
agriculture, but time series data on sectoral gross output is notoriously difficult to
obtain. The standard proxy is to use the value added of sectors as the denominator
(Wood 1994), which imparts no bias if the proportion of intermediate cost in total cost
is constant over time. For all five countries there is a common pattern for the period
1965-1979 (import substitution period), high and rising relative net exports. Then, El
Salvador, Guatemala, and Nicaragua entered into sharp decline. Decline occurs also
in Honduras, but by considerably less. Only in Costa Rica is liberalisation assoicated
wiht a rise in net agricultural exports. Below, we consider whether Costa Rica could
be considered a statistically different case.

[Table 4 goes here]

One might explain the decline in net exports for El Salvador during the 1980s
by the effect of war, largely fought in the countryside. In the 1990s the war was over,
yet liberalisation was associated with a massive decline in net exports, from an
average of over forty percent of agricultural value added during 1985-1989, to eleven
percent during 1990-1994. Similarly, armed conflict might be summoned to account
for the sharp decrease in Nicaragua. Net exports represented over seventy percent of
agricultural value added in the 1970s, then fell to thirty-eight percent during 1980-
1984, and to eighteen percent during 1985-1989. From an orthodox perspective, these
declines would also be attributed to the strong price and non-price interventions by the
Sandinista government However, with the liberalisation of the 1990s, net agricultural
exports continued to drop, to below ten percent. Net agricultural exports in
Guatemala remained the same from the first half of the 1980s to the second half, then
dropped by a full ten percentage points in the period we describe as full liberalisation.

While the evidence presented so far suggests that liberalisation was not
associated with a revealed comparative advantage in agriculture in the Central
American countries, this has not been analysed in a rigorous framework. To do this,

5


we develop a model of the determination of net exports from the sector. By
definition, relative net exports of agricultural products are:

Xnt = (Xt - Mt)/[GDPagrict]

Agricultural exports are taken to be a function of the real exchange rate and
the relative domestic price of agricultural products to other tradables. The former
price determines the relative return between tradables and non-tradables, while the
latter measures the return to resources in agricultural production relatively to their
return in producing other tradables.

Xt = Xt(RERt, RPAMt), with the first derivatives of RER & RPAM positive.

The real exchange rate is measured as the nominal exchange rate divided by
the GDP deflator,6 and relative tradable prices by the agricultural sector price deflator
divided by the manufacturing sector price deflator.

Imports of agricultural commodities are treated as a function of the real
exchange rate and national income (GDP):

Mt = Mt(RERt, GDPt), with the first derivative of RERt negative,
and the first derivative of GDPt positive.

Therefore:

AXnt = Oto + aj [ARERt] + ct2[ARPAMt] + ct3[AGDPt] + et

With it predicted that:

oti > 0, ct2 > 0, and ot3 < 0, and statistically significant.

The two relative price variables are measured as the first logarithmic
difference; e.g., ARERt = ln[RERj7RERt_ | ]. Experimentation showed that AGDPt
conformed best to theoretical prediction when measured as the square of the first
rleative difference in GDP (retaining the original sign, see Annex for detailed
discussion). Use of the first relative difference would result in extremely large
changes in AXnt, especially in the few cases in which the variable changes sign from
one year to the next. Therefore, the absolute first difference is used. It is to be noted
that this is not an equilibrium model, since there is no a priori stable value for AXnt.

6 This implies that an increase in RERt reflects a real devaluation. The RER is not adjusted
by international prices because the adjustment would be by the same price index for each
country (e.g., the US wholesale price index), involving only a linear transformation of the
variable.

6


To test for the impact of liberalisation, we introduce a dummy variable which
assigns the value of unity to the years in each country defined as ‘liberalised’ (see
Annex). As discussed above, in two of the countries, El Salvador and Nicaragua,
major armed conflicts disrupted economic activity, especially in the agricultural
sector. To accommodate this, a ‘conflict’ variable is introduced, with the value of
unity for conflict-affected years (again, see Annex). Finally, the time series data,
1970-1994, are pooled using dummy variables for countries (with Costa Rica
omitted).

The results of modelling exercise are presented in Table 5. The model
accounts for about a quarter of the total variation in net exports (the degrees of
freedom adjusted R-square is .234), and the overall relationship is significant at a
probability of less than one-thousandth (significance of the F-statistic). The
explanatory variables conform to their theoretically predicted signs and are significant
in all cases at less than five percent probability. With the relative price effects and
GDP growth accounted for, the liberalisation variable shows a negative sign and is
statistically significant at less than two percent probability. The coefficient indicates
that each liberalisation year was associated with more than a six percentage point
negative shift in relative net exports, slightly less than half the negative shift
associated with the conflict variable.

None of the country dummy variables was statistically significant. This result
is quite important and contrary to conventional wisdom, which portrays Costa Rica as
a particular success of liberalisation among the Central American countries, because it
initiated its deregulation of markets prior to the other countries and with more vigour.
Whether or not it is the case that Costa Rica pursued liberalisation with greater
commitment, the statistical results reveal no significant difference in the behaviour of
net agricultural exports when one accounts for the effects of relative prices and
national income, in a context in which the liberalisation dummy is negative and
significant.

As interesting, and unexpected, as this result is, it raises the question, why did
the policy shift toward liberalisation not have its predicted outcome on the agricultural
sector in Central America? The orthodox approach would turn to the movements in
relative prices in our model for the answer to this question: if a comparative

7


advantage was not revealed for agriculture, then it should have been the case that
relative prices did not move in favour of that sector. The orthodox relative price
mechanism develops in three parts: first, with liberalisation of the external account,
the exchange rate should find its equilibrium level; second, the exchange rate
adjustment should increase the prices of tradables relatively to non-tradables; and,
third, if agriculture were labour intensive and the countries labour abundant, then
within tradables agricultural prices should have risen relatively to other tradables and
relatively to the general price level.

Table 6 provides the basic relative price data used in the regression, and are
shown graphically in Figures 1 and 2. Part A of the table gives the purchasing power
parity exchange rates for the 1980s and 1990s, calculated such that an increase
indicates a real depreciation. The numbers show that for Honduras and Nicaragua
liberalisation of the current account had the predicted result: the exchange rates
depreciated during the periods of liberalisation. For Costa Rica and El Salvador the
anticipated outcome was contradicted: after appreciations during the 1980s, the
exchange rate showed a further appreciation during the 1990s. In the case of
Guatemala, the real exchange rate depreciated during the first five years of
liberalisation, then appreciated slightly in the 1990s. Overall, the hypothesis that
exchange rate liberalisation results in real devaluation is unproved in Central America,
for it is sustained in only two of five cases. Further, Nicaragua must be considered a
special case, for the real depreciation of the exchange rate may have had more to do
with the end of hyperinflation than with liberalisation as such.

In part B of the table we have calculated the ratio of the agricultural value
added deflator to the GDP price deflator. First, it is interesting to note that during the
import-substituting 1970s, relative prices moved in favour of agriculture in four of the
five countries (Costa Rica, El Salvador, Guatemala and Nicaragua). Relative prices
moved against agriculture only in Honduras, which was the country that implemented
import substitution least. For the liberalisation periods, the result was the reverse: in
Costa Rica, El Salvador, Guatemala and Nicaragua relative prices moved against
agriculture, and in favour of agriculture only in Honduras.

[Table 6 and Figures 1 and 2 go here]

8


In pursuit of an explanation of the decline in the revealed comparative
advantage in agriculture (net exports), we can disaggregate the output of the sector
between those for domestic consumption and those for export. Table 7 provides
relevant price statistics. The first five products listed in the table are the major
exportables, coffee, sugar, cotton, bananas, and livestock, which together, the five
accounted for over eighty percent of agricultural exports during 1980-1984, and about
seventy-five percent during 1990-1994.7 These are followed by four products for
domestic consumption, maize, beans, rice, and wheat (‘importables’). For each
product, the trend in the absolute price level is calculated using simple regression
analysis, for the years 1980-1994.8 For comparison, the US wholesale price index is
given in the last row of the table. These price trends suggest an explanation for the
decline in the net exports of the agricultural sector in the region. For two of the
exportables, coffee and sugar, the price trend is negative and statistically significant,
for two more the trends are non-significant, and for only one (bananas) is there a
significant and positive tendency. If the trend in the US wholesale price index is
taken as a measure of the ‘world price level’, one can conclude that relative prices
moved against all the region’s major exportables except bananas. As one would
expect, only bananas showed a consistent increase in its share in total agricultural
exports.

For the major importable agricultural products, three of the four show a
significant and negative price trend over the period (with the trend for the fourth,
beans, also negative, but non-significant). Thus, on the import side, producers of
basic food commodities in Central America faced declining international prices; i.e.,
competition with cheapening imports. Table 8 shows that falling border prices of
importables was associated, as one would expect, with declining net exports of these
products (increased net imports). For the region as a whole, net exports of maize
when from minus six percent of national production during 1970-1979, to minus
twenty-five percent during 1990-95 (with a lower proportional change if Costa Rica is

7 There was a dramatic change in composition, however, In the earlier period, bananas
accounted for less than twenty of agricultural exports and near forty percent in the latter
period.

8 For exportables the price is the average (trade-weighted) export price across the five
countries. For the importables the average (trade-weighted) import (border) price across the
five countries is used.

9


excluded). For rice, the deterioration in net exports was greater, from near self-
sufficiency during 1970-79, to almost forty percent in the 1990s. While the change
for the region as a whole was massive for beans (minus three percent of national
production to minus twenty-three, form the 1970s to the 1990s), almost all of this was
in Costa Rica. When this country is excluded, the shift is a relatively small ten
percentage points, plus six percent in the 1970s to minus 3.5 percent in the 1990s.
This relatively small shift is consistent with the non-significant trend in beans prices
(see Table 7, ‘importables’, number 2).

On the basis of the price information in Table 7 and the net export statistics in
Table 8, one can construct a story to account for the negative coefficient on the
liberalisation variable in our model. Declining world prices for the region’s major
exportable and importable agricultural commodities tended, ceterius paribus, to
undermine the return to most of the regions major agricultural commodities. On the
export side this resulted in strong disincentives, such that in real terms agricultural
exports grew slowly in every country but Costa Rica. On the import side, declining
international prices for the major importables also generated disincentives for
producers in the region, with the result that agricultural imports grew rapidly. Trade
liberalisation exacerbated these effects, by bringing internal agricultural prices more
closely in line with international prices. The relative price shifts might have been off-
set by exchange rate management (i.e., managed devaluations), but the liberalisation
packages included as key elements exchange rate unification and non-intervention in
foreign exchange markets. Rather than depreciating to off-set international price
trends, exchange rates in all of the countries but Honduras appreciated in real terms
under liberalisation. Why this occurred is beyond the scope of this study, but
appreciation of the market-determined exchange rates may have resulted from
financial inflows facilitated by capital account deregulation.

[Tables 7 & 8 go here]

io


Conclusion

Thus, our analysis concludes that trade liberalisation shifted incentives away
from the major agricultural commodities in Central America by facilitating the
transmission of international prices to domestic producers. An advocate of free
market mechanisms might argue that this result is not a bad thing: producers should
shift from commodities with declining world prices to ones with raising or stable
prices. None would deny that the Central American countries needed export
diversification. Table 9 shows that in the ealry 1980s five products accounted for
toward ninety percent of agricultural exports in Costa Rica, El Salvador, Honduras,
and Nicaragua, and over seventy percent in Guatemala. Inspection of the table
suggests that liberalisation was associated with some diversification, eleven
percentage points for Costa Rica and nine for Guatemala, and over ten percent for
both El Salvador and Nicaragua.9 For Honduras the change was small and merely a
return to the degree of concentration of the early 1980s. However, the diversification
was not sufficient to prevent a deterioration of the agricultural trade balance in four of
the five countries. This suggests that a purposeful trade policy, combining exchange
rate management and subsidies to farm inputs, might have produced an outcome more
favourable to the agricultural sector than deregulation and trade liberalisation. If
fostering agriculture was a policy goal, a judicious combination of sector export
promotion and import substitution could have been more successful.

9 These percentage changes overstate the degree of diversification, since the share for
bananas rose substantially. Further, much of the decline was the result of the collapse of
cotton production throughout the region.

11


References:

Bulmer Thomas, Victor

1992 Life after Debt: The New Economic Trajectory in Latin America
(London: Institute of Latin American Studies)

Comision Economica para America Latina (CEPAL)

1996a Information Basica del Sector Agropecuario, Subregion Norte de
America Latina y el Caribe, 1980-1994 (Mexico: CEPAL)

1996b Istmo Centroamericano: Series Historicas Macroeconomicas
(Mexico: CEPAL)

Food and Agricultural Organisation (FAO)

1996 FAOSTAT (Rome: FAO) (on ‘floppy’ disk)

Liang, N.

1992 'Beyond Import Substitution and Export Promotion: A new typology of
trade strategies,' Journal of Development Studies 28,3

Rodrik, D.

1991 ‘Policy Uncertainty and Private Investment in Developing Countries,’
Journal of Development Economics 36 (June)

Rueda-Junquera, Fernando

1997 ‘Costs and Benefits of Agricultural Trade in Central America,’ paper
presented to the 1997 meeting of the Latin American Studies Association,
Guadalajara, Mexico, April 17-19, 1997

Secretaria de Integracion Economica de Centro America (SIECA)

1981 Compendio Estadistico Centroamericano (Guatemala: SIECA)

1988 Series Estadisticas Seleccionadas de Centroamerica (Guatemala:
SIECA)

1990 Series Estadisticas Seleccionadas de Centroamerica (Guatemala:
SIECA)

Weeks, John

1985 The Economies of Central America (New York: Holmes & Meier)

1995 ‘Macroeconomic Adjustment and Latin American Agriculture since
1980,’ in John Weeks (ed.), Structural Adjustment and the Agricultural
Sector in Latin America and the Caribbean (London: Macmillan and the
Institute of Latin American Studies, University of London)

1996 ‘Regional Cooperation and Food Security in Central America,’ in Jaime
Behar (ed.), Regional Integration and Economic Reform in Central America
(Stockholn: Institute of Latin American Studies)

World Bank

1995 World Development Indicators (Washington: World Bank) [on CD-
ROM)

Adrian Wood

1994 North-South Trade, Employment and Inequality (Oxford: Clarendon)

12


Table 1: Characterisation of Policy Regimes by Period, 1960-1995

Country Period 1960-1969 1970-1979 1980-1984 1985-1989 1990-1995
Costa Rica Shift towards import substitution Import Substitution interventions Moderate liberalisation Lil^ralised. (from 1983) Liberalised
El Salvador Shift towards import substitution Import Substitution interventions Strong intervention Moderate liberalisation Liberalised -
Guatemala Shift towards import substitution Import Substitution interventions Moderate liberalisation Continued liberalisation Liberalised
Honduras Minor import substitution policies Mild interventions (not ImpSub) No change Little change - Major Liberalisation
Nicaragua Shift towards import substitution Import Substitution interventions Strong intervention Moderate liberalisation Radical & liberalisation
Comments CACM officially begun in 1963 CACM at its peak in first half of decade (without Honduras); insurrection in Nicaragua 1977-79 Collapse of the CACM; war in El Salvador & Nicaragua War continues in El Salvador & Nicaragua, ceases in both countries by end of decade Government changes in Nicaragua (1990)

13


Table 2: Growth Rates of Constant Price Agricultural Value Added,
Central America, 1960-1994

Costa Rica El Salvador Guatemala Honduras Nicaragua
1960-64 3.5 6.4 5.0 5.3 12.5
1965-69 7.7 1.8 3.8 6.2 6.1
1970-74 3.6 4.8 6.8 .7 6.6
1975-79 2.6 3.9 3.4 3.9 0.1
1980-84 2.8 -3.2 -.1 1.9 -1.3
1985-89 3.1 -.5 2.2 3.9 -3.6
1990-94 3.6 4.0 3.1 3.9 0.0

Source: World Bank 1995, except for Guatemala, which used CEPAL (1996b) for
1960-1986.

Table 3: Trend Rate of Growth of Constant Price Exports
& ‘Liberalisation Shift’, Central America, 1980-1995

Trend Shift Dummy R2 & F-stat
Costa Rica 6.6 @ .01 neg non sig 0.739 17.0
El Salvador -1.7 ns pos non sig 0.035 0.2
Guatemala 3.5 @ .10 neg- @ .05 0.267 2.19
Honduras 1.9 @ .10 neg non sig 0.196 1.4
Nicaragua -7.1 @ .01 pos non sig 0.681 12.8

Note: The ‘shift dummy’ assumes a value of unity during 1985-1994 for Costa
Rica and Guatemala; during 1990-1994 for the other countries. The ‘trend’
coefficient has been multiplied by 100 to yield a percent rate of growth.

Source: World Bank 1995.

14


Table 4: Net Agricultural Exports by Period, 1965-1995
A. Millions of current US dollars (annual averages)

Countries: Periods: Costa Rica El Salvador Guatemala Honduras Nicaragua Central America Excluding Costa Rica
1965-69 98 103 143 101 103 548 450
1970-74 179 172 232 117 158 858 679
1975-79 464 493 658 285 401 2300 1837
1980-84 519 401 708 441 225 2293 1774
1985-89 641 288 648 535 108 2220 1580
1990-95 853 67 563 408 " .45 7 ’ 1936 . ' ' "1083 7
B. Annual Averages as Percentage of Sectoral Value Added
Countries: Periods: Costa Rica El Salvador Guatemala Honduras Nicaragua Central America Excluding Costa Rica
Annual Averages: 1965-69 59.0 43.5 na 48.4 65.4 na na
1970-74 68.6 51.9 34.7 50.3 62.2 49.4 45.9
1975-79 75.9 64.8 46.2 66.7 88.4 63.2 60.5
1980-84 73.9 47.1 32.2 76.8 39.3 46.5 42.0
1985-89 76.2 40.8 32.3 75.0 18.1 46.2 39.9
1990-95 77.5 10.8 22.1 70.0 9.1 J 39.7 . 28.6,..^

Sources: FAO 1996 (exports and imports) and World Bank 1995 (GDP). See Annex.

15


Table 5:

Regression Analysis of the Annual Change in Net Agricultural Exports
As Proportion of Agricultural Value Added ([X - M]/AgricGDP]), 1970-1994
[Dependent variable: Absolute first difference]

Coefficient Std Error T-stat Significance
Constant .0749 .0280 2.678 .009
Explanatory variables (first relative difference)
1. Change in Relative prices agric/manuf (In) (RPAM): .1567 .0756 2.074 .040
2. Change in Real
exchange rate (In) (RER) .4501 .0903 4.987 .000
3. Growth rate of constant
price GDP (squared) (RGDPSQ) -2.5714 1.1777 -2.183 .031
Dummy Variables 4. Liberalisation -.0626 .0241 -2.599 .011
5. Conflict -.1444 .0353 -4.092 .000
6. El Salvador .0232 .0364 .636 .526 (nsgn)
7. Guatemala -.0210 .0336 -.626 .533 (nsgn)
8. Honduras -.0439 .0345 -1.273 .206 (nsgn)
9. Nicaragua .0028 .0373 .074 .941
(nsgn)
Adj R-squared .2340 df=l 10
& F-stat 5.038 sgn of F:
.000
Correlation among independent variables (signif of F):
1. RER&RPAM .047 sgn
2. RER&RGDPSQ .354 nsgn
3. RPAM&RGDPSQ .134 nsgn
Durbin-Watson 2.2984 serial corr
statistic hypothesis
denied

Note: For sources and explanation of variables, see Annex.

16


Table 6: Changes in Relative Prices, Central America, 1960-1994

A. Real Exchange Rate (average for 1970-1994 = 100)

Country Years Costa Rica El Salvador Guatemala Honduras Nicaragua
1970-74 165 196 177 161 174
1975-79 97 118 105 111 110
1980-84 94 75 69 73 66
1985-89 78 63 77 64 70
1990-94 66 48 73 91 81

B. Ratio of Agricultural to Manufacturing Price Deflator
(average for 1970-1994= 100)

Country Years Costa Rica El Salvador Guatemala Honduras Nicaragua
1970-74 85 111 106 100 85
1975-79 111 148 120 89 111
1980-84 113 107 94 71 113
1985-89 99 77 96 110 99
1990-94 88 46 81 136 88

Note: For El Salvador, Honduras and Nicaragua, the last year is 1993. For Guatemala, 1970-1986 is
the external terms of trade. Sec Annex.

Source: World Bank 1995.


1980-1995

Table 7: Regression-calculated International Price Trends
for the Major Central American Agricultural Commodities,

Crop Trend Statistic (percent p. a.) R-square & F-statistic
Exportables 1. Coffee -3.8 .327
(.02) 6.8
2. Sugar -2.1 .183
(.10) 3.3
3. Cotton 0.8 .032
(ns) 0.5
4. Bananas 2.2 .818
(.01) 62.8
5. Livestock .000 .001
(ns) 0.1
Importables 1. Maize -2.1 .420
(.01) 8.7
2. Beans - 1.1 .033
(ns) 0.4
3. Rice -2.9 .568
(.01) 15.8
4. Wheal - 1.4 .246
(.10) • 3.9
USA WPI 1.8 .909
(.01) 130.3

Note: The prices are the average paid for Central America as a
whole.

Source: CEPAL 1996a.

18


Table 8: Net Exports of Staples, 1970-1994
( as percentage of national production)

A. Maize

Countries: Periods: Costa Rica El Salvador Guatemala Honduras Nicaragua Central America Excluding Costa Rica
1970-79 -18.8 -2.0 -4.4 -1.3 -4.5 -5.7 -3.6
1980-84 -70.5 -10.8 -2.9 -5.2 -36.4 -10.5 -8.0

1985-89 -88.9 ' -8.9 â– 24: ; -4.3 -14.2 -8.9 -5.3
1990-95 “ -568.9 “ -20.5 ■'SlSW Mao

B. Rice

Countries: Periods: Costa Rica El Salvador Guatemala Honduras Nicaragua Central America Excluding Costa Rica
1970-79 7.1 1.5 -12.2 -15.3 6.9 -2.3 -4.8

1980-84 15.9 -13.9 -7.4 -3.8 -21.8 -1.0 -13.5

1985-89 -9.1 -20.3 -14.1 -4.4 -53.1 -20.4 -27.6

1990-95 .• ;-274fe : -43.7 / ",-84.0”'

C. Beans

Countries: Nicaragua 8.0 Central America -3.2 Excluding Costa Rica 6.1
Periods: 1970-79 Costa Rica El Salvador Guatemala -3.1 Honduras 10.9
-37.4 -7.3
1980-84 -97.3 -0.1 -0.9 3.4 -22.2 -9.4 -4.4
1985-89 r 7.6 -1 -5.3 : -1.9 1 -3.4 -25.4 -6.6 -8.0
1990-95 ; -38.3 -17.1 -3.3 „ -5.4 :

Sources: Weeks 1990 & CEPAL 1996a.

Table 9:

Proportion of Agricultural Export Earnings Contributed
by Coffee, Cotton, Sugar, Bananas & Meat, 1980s & 1990s

Countries: Periods: Costa Rica El Salvador Guatemala Honduras Nicaragua
1980-84 89 92 75 85 88
1985-89 84 95 75 89 92
1990-94 78 81 68 85 80
percentage pt change (lib - prelib) -11 -14 -9 -4 -12

Source: CEPAL 1996b.

19


Relative Prices REER

Figure 1: Real Exchange Rates in Central America, 1970-1994

Years

Figure 2: Ratio of Agricultural to Manufacturing Deflator,
Costa Rica, El Salvador, Hondruas & Nicaragua, 1970-1994

20


Annex:

Date Sources:

The major sources are FAGSTAT, a data base on ‘floppy’ disk from the FAO,
and World Development Indicators of the World Bank, on CD-ROM. The latter has
been replaced by a new CD-ROM, more technically advanced, but with a shorter time
series. There are no relative price data for 1965-1969 in the case of Guatemala. Thus,
the regression is over the years 1970-1994.

1. Agricultural exports and imports: FAOSTAT (1996), in current US dollars.

2. Agricultural value added (GDP): in current US dollars, World
Development Indicators, World Bank. Slightly different timme series are found in
SIECA (1981, 1988, & 1990) and CEPAL (1996b).

3. Constant price GDP: in 1978 prices, World Development Indicators,
World Bank (‘Stars’ on CD Rom).

4. Relative prices: nominal exchange rate, GDP deflator, agricultural sector
deflator, manufacturing sector deflator, World Development Indicators, World Bank
(‘Stars’ on CD Rom).

Regression variables:

1. Change in net agricultural exports, as proportion of sectoral value added,
measured as the absolute change in the fraction,

AXnt =[(Xt - M^/GDPagricJ - [(Xt.j - Mt_i)/GDPagrict_i]

2. Change in the real effective exchange rate

Since for all the countries the US wholesale price index would serve as the
measure of international prices, it is not included in the calculation. The variable is
calculated as the natural logrithim of the ratio of the exchange rate (deflated by the
GDP price index) in years t and t-1.:

AREER = IntRn/PgdpJ/fRnt.j/Pgdpt.fJ

3. Change in the relative prices of agricultural and manufacturing prices.
This variable is a proxy for relative price changes within tradable goods.

ARelPr = lntPagric/PmanufJ/tPagric^j/Pmanuf^i]

For Guatemala, sector price deflators begin in 1987. For 1970-1986, the
external net barter terms of trade were used (FAOSTAT).

21


4. Change in constant price GDP. This variable is:

AGDPt = [(GDPt/GDPt.j) - l]2 , with the sign of the variable prior to squaring

retained. That is, [-.05]2 becomes, -.0025.

5. Conflict'. This variable assumes a value of unity for El Salvador and
Nicaragua in the following years:

El Salvador: 1981-1989
Nicaragua: 1977-1979,1983-1989.

6. Liberalisation'. This variable assumes the value of unity for the following

years:

Costa Rica
El Salvador
Guatemala
Honduras
Nicaragua

1982-1994

1989- 1994
1985-1994

1990- 1994
1989-1994

For an explanation of the choice of these time periods, see Weeks (1995, pp.

70-73).

22










Full Text

PAGE 1

! "!#

PAGE 3

!"!#$$#$%"#"&! #"%%"$"&"##'!"#%#&%&"#(()$)#&*&!(#(!"!#+#",#"&!"!"#&()" "!!)"#+ * ())$#")("%+!#"$!#")"$$+#-'!"#%#&#"'!"#%#&%&&)#")"$&"#($"( ##()).("/$0123-24546728",-2454659%#&###):-;4#"%<;;<&()""$0&)&")?*%#%!)%##")%#@'*)! 0(#<;A
PAGE 4

A;DE&#"#"$$ ()&#?"&&%&.()"D%&#'')/#-AA<))"#%"F&%"?@@ C<01#+#%&#''#+-A6<8(@@"!#%"$*)%-8"%/"((#&"<)*-A<#(&#(")$=(#%#"$#!&##""&%<$()*-A7<"#"$0#"!"8"&")#%-")#&")**<*+2AB<&")&(#"$# "#").(#<"#$%&#''0#262
PAGE 5

?/F/'0E/E/8?00/?00#'*)!%#)&H"&&)#")"$&"#($"( ##()).("/$0123( (#BB'"#-##.(#$#(##"(#D!%

PAGE 7

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

PAGE 8

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

PAGE 9

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

PAGE 10

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

PAGE 11

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

PAGE 12

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

PAGE 13

###%!"&#)")"#+#$(&"$(%%*"")&" #*")((###""&&(#$$"I)")$D9,:$&($"*+##&(#+)")*"$"$&"" ("+%"="%$&)&#$(!#$&%&"&#*#+!&"))#" &()#(")&#"&&%%$"##+"I&)&#D*"")#$(&$+##*")((#&)&#G"&#$"9" "+,:8"))+##%$"#"+B2GBB;+"!)$( $(%%*"")&(#9#0#"&"%##$:()#%$)) ,&"!#$")A%$)"&&(#"(#".("####")*""##,!#9#$ $%"$=(#$G.("54;:+"$#*")))"#! &"#"#"!")#)#"G#("$#9 &"G#"##&:,!)""#*"")&%####&"))!$&#$ "$" &"#"))&""#)#"*!&#!")###)"#*!&&#"$' #"&&(#$+#)")"#*"")" "#* "$#"##&")) &"#"#)#"#!&#!")#&&#$&"##"#"&)")"#"""&"#$#%#"",!&#" !# "#*#)"#*#,!#+) #))#"")# "#*#"&"#$##&)&#*"")#&(#$(%%*"")"#"##&")) &"#()#.(#%!#"#"$&#"#&*#")$%+&!#"0#"&"""!"#&()"(&&)")"#"% #0#")%&"&(#+&"(##"#$#$ ()"#%"#!###&(#"$#%* (####&"#"#0#"&"!(($)")"## "#&%%#%#+##"##&")()#*") &"#$&#"*(#" &()#("),!#"&&(##&#)"#*!&"$"#")&%+78##+9*9!## +"$(,!&#$+"#()#+#"#.(#+$$#!)&##"$)")"##"*#!$&#$(#&%#" &()#(")�#")%&"C#$,"!!"&()$#(##%*%#)"#*!&(%$)#"##.(#-"&%!""#*

PAGE 14

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

PAGE 15

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

PAGE 16

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

PAGE 17

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

PAGE 18

&-F()%%"+>&#BB5:,)8,;+09/$-##(#/"#%&"#($:0%&%&"!""%&"/"#"90/:BB6"<1#099)0+#=>?9?,&-0/:BB6<: $9?,&-0/:8$"$ &()#(") ""#98:BB6.0@,0,9%-8:9I)!!D$:/" +BB5
PAGE 19

")-0""&#"#)& %$+B62GBBA0(#$B62GB6BB2GBBB72GB7;B7AGB7BBB2GBBA0#"&"##"$%!#(##(#%!#(##(##*#?$"#)")"#&'()*%+)")*"$##"$%!#(##(#%!#(##(##*## #*#?$"#)")"#,("#%")"##"$%!#(##(#%!#(##(##*#?$"#)")"#-"1$("?%!#(##(#!)&?)$#*#9#%!(:&" /##)&" ,./&"" ("##"$%!#(##(#%!#(##(##*## #*#?$"#)")"#01$-))00?&")) (B6400?"##!"#")$&"$9#(#1$(":J(&#&"" ("BGB0))"!#00?J")")*"$H&"" (""&#()")*"$H&"" ("+&"#&(#$$&"$*%#&" &"" ("9BB2:4

PAGE 20

")5-#"#0#"#& &()#(")>")($$$+0#")%&"+B62GBB;0#"&")")*"$("#%")"1$("&"" ("B62G6;4A6;A2A4'2B6AG6B#'#747 '6B2G;46'% '%66BAGB564B4;4B2B72G7;57G45GBG4B7AG7B'GA'4BG46BB2GB;' '3''*3'3(&-)$F"BBA+,&!#("#%")"+&($0/9BB6:B62GB76")4-$"##0#"#&,!#HI/")"##D+0#")%&"+B72GBBA$#'(%%5H8G#"#0#"&"66V2 24B2)")*"$G! 224A25("#%")"4AV2 GV2A2565B1$("BV2 2B6;&"" ("GV2! 26757#-I#$(%%D"(%"*")((#$( B7AGBB;0#"&""$("#%")"J$( BB2GBB;##&(#I#$D&&#"%()#!)$22#)$"!&#"# #(&-)$F"BBA;

PAGE 21

");-# &()#("),!#$+B6AGBBA?))&(#E$))"9"(")"*" :0(#-$-0#"&")")*"$("#%")"1$("&"" ("0#")%&",&)($ 0#"&"B6AG6BB724;4224A;7;A2B2G;B5545A77A76BBAGB;6;*6A757A333%#B72G7;AB;2#3%;;55A55B4;B7AG7B 577 %22275552A72BB2GBA%2 #2 3%4'2D* '5543%F(")*" "&#" &#")>")($$$0(#-$-0#"&")")*"$("#%")"1$("&"" ("0#")%&",&)($ 0#"&"(")*" -B6AG6BAB2;4A";7;6A;""B2G;676AB4;A24655;B;;ABBAGBAB '% '66%%' ' 3'2B72G7;4B;455# '%4B4;6A;52B7AG7B# ';27'A27;654BBBB2GBA##'23'%'#3'3*'*'#'%' 6''&(&-8BB69,!#"$%!#:"$)$F"BBA9':,A

PAGE 22

")A")#(")0" # &()#("),!#!# &()#(")>")($$$9M3G?NK &'N:+B2GBB;M'!$#*"")-)(##$&N0&##$G#"# &"&0#"#2;B2572567'33*,!)""#*"")9#)"#*$&:0" )"#*!&" &K%"(9:9?:-A62A652;'3350" "),&" "#9:9:;A22B24;B7'3334#"#&#"#!&'9.("$:9'W:G5A;G574'3'(%%>"");/")"#G265625;G5ABB'3A0)&#G;;;24A4G;2B5'3336)")*"$2545246;646A569 :("#%")"G2522446G656A449 :71$("G2;4B24;AG545269 :B&"" ("22572442;B;9 :$=G.("$54;2$L)2H8G#"#A247 8-'3330)"#"% $!$#*"")9 8:-H?'3#5H'W4A; 4?H'W4; '(G"#55B7;1#"##&7#-8(&"$,!)""#*"")+,6

PAGE 23

")6-0" )"#*&+0#")%&"+B62GBB;"),&" "#9"*" B2GBB;L22:0(#U"0#"&")")*"$("#%")"1$("&"" ("B2G; 2* ## #BAGBB72A2B72G7;B;A6B466B7AG7B#%64##6;2BB2GB; ;7#*%F"# &()#(")#?"("&#( &')"#9"*" B2GBB;L22:0(#U"0#"&")")*"$("#%")"1$("&"" ("B2G;7A26227ABAGB%37BB72G7;2B;B7AG7BBBB62BBBB2GB;77;6% %%#-8)")*"$+1$(""$&"" ("+#)"#"BB48("#%")"+B2GB76#,#")#%#"$&,(&-)$F"BBA

PAGE 24

B72GBBA")G&")&()"#$#"#")&$#?"=0#")%&" &()#(")0%%$#+0!$#"##&9!&#!":G.("H8G#"##&,!#")0G4745925:675( "G57492:4440##272459:2A;F"""'77('3+657A/*#&222229:2%!#")?"@G5;5292:75F"G2449:2;4&G5BA6792:A7;")G;5;692:X4BE7B2B92:424#-!&"#"*" !"$0#")%&""")(&-0/BB6"7

PAGE 25

")7-#,!##"!)+B2GBB;9"!&#" "#")!$(&#:?"@0(#-$-0#"&")")*"$("#%")"1$("&"" ("0#")%&",&)($ 0#"&"B2GBG77G52G;;G4G;AGAG46B72G7;G2AG27G5BGA5G46;G2AG72B7AG7BG77B
PAGE 26

0<18 (-"),&" "#0#")%&"+B2GBB;8 (5-"# &()#(")#?"("&#( ')"#+0#"&"+)")*"$+1$("H&"" ("+B2GBB;52

PAGE 27

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

PAGE 28

;9"%!*"")-'#LM9'#K'#=:G)N5+## #*"")!#.(" #"$"#+MG2AN5&%+G225AA#A!*"")"(%"*")((#)")*"$"$&"" ("#)) "-)")*"$-B7GB7B&"" ("-BGBB+B74GB7B6#A!*"")"(%#*")((##)) "-0#"&")")*"$("#%")"1$("&"" ("B75GBB;B7BGBB;B7AGBB;BB2GBB;B7BGBB;8",!)""##&&##%!$+9BBA+!!2G4:55