Impact of the Structural Adjustment Program on the Agricultural

Document Sample
Impact of the Structural Adjustment Program on the Agricultural Powered By Docstoc

Impact of the Structural Adjustment Program on the Agricultural
                 Sector and Economy of Nigeria


            Nii O. Tackie and Odiase S. Abhulimen

                 104 Morrison-Mayberry Hall
                     Tuskegee University
                     Tuskegee, AL 36088
                   Phone: (334) 727-8813
                     Fax: (334) 727-8812

                          July 2001


       Nigeria is the most populous nation in Africa with about 103 million people. The

land area is almost 360,859 square miles and has potential for development with her

enormous natural and human resources (Burren, 1998). However, despite these blessings

Nigeria is still characterized by inequality in income distribution, poor health and

education standards, high unemployment rate, high debt, and relatively low agricultural

productivity. These shortcomings can be attributed to mismanagement and a gross

misconception of the ideals of economic growth and development by key political

leaders. Economic growth is a measure of the increase in economic indicators such as

per capita income and gross domestic product (GDP). Economic development, on the

other hand, is perceived as "a multidimensional process involving major changes in social

structures, popular attitudes, and national institutions, as well as the acceleration of

economic growth, the reduction of inequality, and the eradication of poverty" (Todaro,

1997, p.16). Economic development, therefore, necessarily entails economic growth.

       The Nigerian post independence economy can be viewed in three distinct phases,

namely, first phase from 1960 to 1973; second phase, 1974 to 1982; and third phase,

1983 to present (Oshikoya, 1990). During the first phase, the economy was largely

sustained by agriculture. Substantial expansion in infrastructure, public utilities, and the

construction sectors was supported by the agricultural sector. Economic growth in the

second phase was largely propelled by increased oil exports. This induced huge public

investments and over importation of foreign-made goods. Increases in oil prices in

1973/74 and 1979/80 further precipitated huge transfer of wealth to the country.

Policymakers perceived this windfall to be the opening to extensive development plans.

The government embarked on expansion of urban-based construction, transportation,

communication networks, as well as an ambitious construction of a new national capital

in Abuja. Increases in public sector investment were also accompanied by expansion of

general government consumption. Aggregate expenditure thus exceeded domestic output

by a large margin. In fact, in the second phase, agriculture took a back seat to the oil


       As a result of the massive squandering of resources and mismanagement in the

second phase, the third phase witnessed serious economic deterioration, external debt

crisis, financial fragility, and rising inflation. Buren (1998) attributed the decline in

economic growth to falling and unstable world oil prices after 1981. Thus, the

government became increasingly over-extended financially, with insufficient revenue

from petroleum to pay the rising cost of imports, finance major development projects,

and service external debt payments.

       To deal with the deteriorating economic conditions, the government with the

assistance of the World Bank designed a structural adjustment program (SAP) which was

to be implemented beginning June 1986. The SAP aimed at facilitating economic

growth as a means of jump-starting the economy towards sustainable economic growth

and development. The objectives of the program included the following:

(1) reconstructing and diversifying the productive base of the economy, by reducing the

dependence on oil and imports, (2) laying a basis for sustaining non-inflationary growth,

(3) making substantial progress towards fiscal and balance of payment viability,

(4) improving efficiency of the private sector’s contribution to economic growth, through

liberalized trade and privatization of public sector enterprises, (5) devaluing the naira,

and (6) reducing government deficits (Buren, 1998; Oshikoya ,1990). Also included in

the SAP, were policies geared towards reversing the trend of Nigeria being a heavily

indebted food-importing economy to it becoming a strong domestic food-producing and

exporting economy. Thus, the following policy measures were implemented: (1) removal

of all government subsidies on essential consumer goods such as petroleum products and

food, (2) active export promotion of all items including food staples, other agricultural

products, and raw materials, and (3) general import restrictive measures for all items

including food, medicine, and raw materials (Harrison, 1993).

       Generally, structural adjustment entail policies, designed by world financial

institutions such as the World Bank and the International Monetary Fund (IMF), aimed at

improving the socioeconomic conditions of implementing nations. Adoption and

implementation of such policies (e.g., currency devaluation, trade liberalization,

privatization, and removal of subsidies) in the 1980s and 1990s was perceived as a means

of reversing the pervasive social and economic problems of developing nations. There

has been considerable debate about the effects of such SAP measures. A host of

researchers (e.g., Reed, 1996; Nwosu, 1992; Olomola, 1994) believe that structural

adjustment policies are essential prerequisites for economic recovery, adjustment to, and

development in the new global market place. On the contrary, many other economists

and social scientists such as Igbedioh and Aderiye (1994), Awoyomi (1989), and

Momoh, (1995) argue that SAP measures have led to recessions and poor standards of

living in developing countries. There is limited empirical work on the impact of the SAP

on the Nigerian agricultural sector and economy as a whole. This study is intended to

help fill the gap in the literature. The goal of this study is to analyze the effects of SAP

on the agricultural sector and economy of Nigeria. Specifically, it assesses the influence

of structural adjustment on selected indicators – agricultural production, net agricultural

exports, contribution of agriculture to GDP, and real GDP growth rate -- and the relative

impact of these indicators on the agricultural sector. The study is largely limited to the

agricultural sector because of its relative importance in the Nigerian economy. On the

average agriculture has accounted for 32 percent of GDP and more than 70 percent of

employment. As such, deceleration in the growth of the agricultural sector would

influence the rate of growth of the whole economy if it is not accompanied by greater

increases in other sectors of the economy.

               Specific SAP Policy Measures in the Agricultural Sector

       SAP policy measures in agriculture included the following items:

(1) the removal of all government subsidies on food and other agricultural products,

(2) promotion of the production and export of nontraditional agricultural products,

(3) import restrictive measures on food and other locally produced agriculturally based

raw materials, (4) the establishment of the Directorate of Food, Road, and Rural

Infrastructure as a major instrument for fostering rural and agricultural development, and

(5) increase of the budgetary allocation to the system of agricultural development projects

as a major instrument for agricultural development (Harrison, 1993; Kajisa et al. 1997;

Nwosu 1992).

       The overall objective of implementing structural adjustment in the agricultural

sector was to increase agricultural production and export of agricultural products.

Because of the relative importance of agriculture to the economy, this was supposed to

contribute to improvement in the growth of the economy.


Measuring Impact of Programs

       Several methods have been used to measure the impact of programs on whole

economies or sectors of economies. First, there is the before-after approach that

compares the values of variables in the period before a program is implemented to those

in the period after implementation. The major shortcoming of this approach is it assumes

that all program outcomes are the result of program variables. Although the before-after

approach has some degree of bias as an estimation procedure, it nevertheless, has

inherent objectivity (Moshin and Knight, 1985). The shortcoming of the before-after

approach can be greatly reduced if further statistical tests are conducted on the issue in

question. Studies, which have used the before-after approach, include those by

Reichmann and Stillson (1978), Connors (1979), Kelly (1982), Zulu and Nsouli (1985),

Pastor (1987), and Gylfason (1987).

       Second, there is a modified form of the before-after approach called reference

(control) group approach or with-without approach. This method assumes that the

outcome of subjecting program and control group countries to non-program determinants

would be similar for both groups had not the program countries received the program.

Any differences between the two groups, therefore, are attributed to the program

determinants. The bias here, compared to the previous approach, is lower, yet other

errors may be present because program countries may differ from control group countries

in terms of characteristics (Moshin and Knight, 1985; Goldstein and Montriel, 1986).

Examples of Studies that have used this approach are Donovan (1981, 1982), Loxley

(1984), Pastor (1987), and Gylfason (1987).

       Third, there is the actual-versus-target approach. This approach compares actual

program performance for key macroeconomic factors to targets for these factors set by

the host country and multilateral agency. The success of a program can be gauged by the

extent to which program targets are achieved, but knowing this requires access to

confidential information on country. This information is less likely to be released by the

multilateral agency or host country. In addition, program variables may be affected by

other nonprogram variables which may cause targets to be underachieved or

overachieved (Khan, 1990). Reichmann (1978), Beveridge and Kelly (1980), and Zulu

and Nsouli (1985) used this approach in their studies.

        Fourth is the counterfactual approach. This approach compares the actual

behavior of key macroeconomic variables in the program country with the outcomes that

would have been observed in the absence of the program (Moshin and Knight , 1985).

The downside of this approach is it is very subjective.

       The fifth and final approach to be alluded to here is the comparison-of-

simulations approach. This method uses simulations of economic models to determine

the hypothetical performance of Fund-type policies or policy packages and alternative

policy packages (Khan, 1990). The drawbacks to this method are two fold: one, the real

world effects of program performance may be different from simulated results, and two,

program performance may be different when supported and implemented by a

multilateral agency because of credibility attached to agency. Of course, when program

is implemented outside multilateral agency authority another set of outcomes may be

obtained. As Khan (1990, p. 209) puts it, “such credibility effects are automatically

captured by the outcome-based approaches, but not necessarily by the model-based

approach like the comparison-of-simulations method.” Khan and Knight

(1985, 1981) used this approach.

Choosing the Approach

       Since a mere descriptive analysis is narrow in its focus and may be biased, and

also for reasons indicated above, an explanatory analysis delving into the relationships

among variables is provided. This explanatory analysis relies on a model.

       The model reflects the effects of SAP on the agricultural sector, and thus, the

general economy. Figure 1 highlights the model under investigation and shows the

expected relationships among the variables under consideration; the signs are indicative

of the combined influence of the SAP variable on a target variable. Starting from the

left-hand side, SAP is presumed to have an improvement in overall agricultural

production. The overall improvement in agricultural production is expected to increase

net agricultural exports (agricultural exports minus agricultural imports), this in turn is

supposed to improve the contribution of agriculture to GDP, translating ultimately to a

positive growth in real GDP growth rate, ceteris paribus.

       The study, therefore, did not follow the mathematical model used by other studies

cited previously in this study (e.g., Zulu and Nsouli, 1985; Pastor, 1987; Khan, 1990).

Rather, it adopts the methodology of path analysis, used by Rajaonarivony (1996) in

analyzing effects of IMF programs on Madagascar’s economy. In path analysis,

predictors change depending on the variables being analyzed at a particular time (Kim

and Kohout, 1975).

                                **** Put Figure 1 Here ****

Data Collection

       Data for the variables are from the Food and Agriculture Organization (FAO) data

files. The data are as follows: agricultural production, measured as production index with

1989-91 as base year; agricultural exports, measured in dollar value; agricultural imports,

measured in dollar value; contribution of agriculture to GDP, measured in percent; real

GDP growth rate measured in percent; and SAP measured as a dummy. The data covered

the period from 1970 to 1997.

Data Analysis

       The set of relationships shown in Figure 1 is called a multi-stage path model or

path analysis. That is, a dependent variable at a particular stage in the model becomes an

independent variable for a subsequent stage. For instance, agricultural production is

dependent on SAP, while SAP and agricultural production are assumed to influence net

agricultural exports. Similarly, SAP, agricultural production, and net agricultural exports

are supposed to influence contribution of agriculture to GDP. An interconnected series of

multiple regressions is commonly used to evaluate the model.

       The beta coefficient was used to evaluate the model. The beta coefficient

measures the relative impact or importance of an independent variable on the dependent

variable (e.g., when there are two beta values of say .02 and .34, the .34 value has more

impact than the .02 value). When there is one independent variable, the beta coefficient

is the bivariate correlation coefficient r. The larger the beta coefficient, the stronger a

variable’s relationship to the dependent variable.

       The SAP variable is expected to have both direct and indirect effects on target

variables. As stated before, SAP should increase agricultural production; thus there

should be a strong direct impact. Similarly, agricultural production is presumed to

influence net agricultural exports directly. The SAP variable also has an indirect effect

on net agricultural exports. It has an indirect effect on net agricultural exports through its

influence on agricultural production, in addition to whatever direct impact it has on net

agricultural exports.

       The major advantage of path analysis is that it allows estimation of both direct

and indirect effects of a variable on another in the causal model. The direct influence is

the beta for each independent variable in a particular multiple regression, or the bivariate

r if there exists only one independent variable. Indirect effects can only occur if a two or

more-stage linkages exist between a dependent variable and an independent variable.

The indirect effect is then estimated by the product of the betas along the indirect path.

The sum of the direct and indirect effects yields the combined effect at any level. For

example, in the causal model, Figure 1, the SAP variable is hypothesized to influence net

agricultural exports both directly and indirectly. Indirectly, SAP influences net

agricultural exports through agricultural production as an intervening variable. The direct

influence of SAP on net agricultural exports is its beta in multiple regression with

agricultural production. That is, SAP and agricultural production are independent

variables and net agricultural export is the dependent variable. The indirect effect of the

SAP variable is calculated by multiplying the bivariate r of SAP and agricultural

production with the beta of agriculture production. Similarly, all direct and indirect

influences are calculated. See Appendix A for mathematical details.

       In general, the following combined effects should result: implementation of SAP

will cause an increase in agricultural production (positive effect); a rise in net agricultural

exports (positive effect); an increase in contribution of agriculture to GDP (positive

effect); and an increase in real GDP growth rate (positive effect). In short, the total

causal effect of implementation of SAP on real GDP growth rate should be positive.


       Table 1 shows the results of the bivariate analysis between the SAP variable and

the other variables. The bivariate results reflect the direct effect of SAP if there were

only one independent variable. The program’s effects on agricultural production, net

agricultural exports, and real GDP growth rate are positive as expected. The program’s

effect on contribution of agriculture to GDP is negative, contrary to expectation. The

direct effect of SAP on agricultural production was .85.

                                **** Put Table 1 Here ****

       Tables 2 through 5 show the results of the multi-stage analysis. The multi-stage

analysis shows the path progression of the effects of SAP in the agricultural sector (see

Appendix A for mathematical details). Table 2 shows the results of the multiple

regression analysis on net agricultural exports. The beta for SAP is .45 and for

agricultural production -.37. The direct effect of SAP on net agricultural exports is .45

and the indirect effect -.31. The combined effect of SAP on net agricultural exports is .14

(Appendix A). As expected the influence of SAP on net agricultural exports is positive.

                               **** Put Table 2 Here ****

       The results of the multiple regression analysis on contribution of agriculture to

GDP are in Table 3. The betas are .25, -.47, and .48 for SAP, agricultural production,

and net agricultural exports, respectively. The direct effect of SAP on contribution of

agriculture to GDP is .25 and the indirect effect -.19. The combined effect of SAP on

contribution of agriculture to GDP is .06 (Appendix A), a positive expected sign.

                               **** Put Table 3 Here ****

       Table 4 shows the multiple regression analysis on real GDP growth rate.

Agricultural production, net agricultural exports, and contribution of agriculture to GDP

had a positive influence on real GDP growth rate. The respective betas are .36, .60, and

.12. However, the SAP variable had a negative direct impact of -.36 (Appendix A) on the

real GDP growth rate. This unexpected sign may be explained by the inflation factor.

That is, with inflation accounted for in real GDP the “true” direct effect of SAP reflects a

negative or decreased value.

                                   **** Put Table 4 Here ****

       It was hypothesized that the SAP will improve agricultural production which will

positively influence net agricultural exports, which will in turn positively influence

contribution of agriculture to GDP, and this will ultimately improve real GDP growth rate

in total. The results indicate that the hypotheses are not contradicted.

        The total causal impact of SAP in the agricultural sector on real GDP growth rate

is in Table 5. The SAP had an overall expected positive causal impact of .69 on real

GDP growth rate. This comprises a direct and an indirect effect. The direct effect is -.36.

The indirect effect passes through three intervening variables, agricultural production

(.85), net agricultural exports (.14) and contribution of agriculture to GDP (.06). This

indirect effect sums up to 1.05. It is the sum of the direct and indirect effect at each stage

that yields the total causal impact. The indirect effect offset the negative value of the

direct effect. It is likely that some aspects of SAP, incentives, such as favorable producer

prices may have influenced the positive effect. The statistical analysis shows that

overall, SAP in the agricultural sector resulted in an improvement in real GDP growth


                                    **** Put Table 5 Here ****

                                    Summary and Conclusion

        The focus of the study is to analyze the effects of SAP on the agricultural sector

and economy of Nigeria. The data for study are from the files of the FAO covering the

period 1970 to 1997. Multiple regression and path analysis are used to analyze the data.

        The results reveal that the hypotheses, in terms of the path analysis, are not

contradicted. SAP had a positive impact on agricultural production, which in turn, had a

positive impact on net agricultural exports, which in turn, had a positive impact on

contribution of agriculture to GDP, which ultimately led to a positive impact on real GDP

growth rate.

        This indicates that overall SAP is beneficial to the Nigerian agricultural sector and

economy. Therefore, the Nigerian authorities should keep the SAP policies in the

agricultural sector in place. Since agriculture is very important to the Nigerian economy,

an improvement or growth in this sector ultimately influences growth of the overall


        The contribution of this study to the body of literature on this topic

notwithstanding, it has some limitations. First, all improvements in the agricultural

sector and the Nigerian economy cannot be attributed to SAP. The results presented here

can be viewed as suggestive. Second, one cannot exhaust in any one analysis the many

variables that might account for performance of the agricultural sector and economy of

Nigeria. Third, no agreement has been reached as yet as to the most appropriate way of

evaluating impact of programs or policies. Indeed, none of the approaches is totally



1. Note that levels of significance are not stated in the findings section. The reason is
that in multiple regression analysis, the assumption normally is significance tests are
performed on data set that is a representative random sample of a population. This study
does not have a random sample from a population. The data, then, are the population.
When data include all observations on a population, coefficient estimates are best viewed
as population parameters. Hence, tests of statistical significance become inappropriate.
The most appropriate method for evaluating the significance of coefficients is by
evaluating the relative impacts of beta coefficients (Ringquist, 1994; McClosky, 1985;
Henkel, 1976).

2. For interested readers, coefficient for AGP r in Table 1 was significant at .01 level
(2-tailed); coefficient for NAGEX in Table 3 was significant at .05 level (2-tailed);
and coefficient for NAGEX in Table 4 was significant at .05 level (2-tailed).


Awoyomi, B., (1989), “Structural adjustment, risk aversion and livestock production in
     Nigeria,” Rural Development in Nigeria, 3, 82-87.

Buren, R., (1998), Africa south of the sahara, 27th (Ed.), (Europa Publication Ltd., New

Beveridge, W. A. and M. R. Kelly, (1980), “Fiscal content of finance program
       supported by stand-by arrangement in the upper credit tranches, 1969-78,” IMF
       Staff Papers 27, 205-249.

Connors, Thomas A., (1979), “The apparent effects of recent IMF stabilization
      programs,” International Financial Discussion Papers, No. 135 (U. S. Board of
      Governors of the Federal Reserve System, International Finance Division,
      Washington, D. C.).

Donovan, Donald J., (1981), ‘Real responses associated with exchange rate action in
      selected upper credit tranche stabilization programs,” IMF Staff Papers 28,

Donovan, Donald J., (1982), “Macroeconomic performance and adjustment under fund-
      supported programs: the experience of the seventies,” IMF Staff Papers 29,

Goldstein, M. and P. Montriel, (1986), “Evaluating fund stabilization programs
       with multi-country data: some methodological pitfalls,” IMF Staff Papers 33

Gylfason, T., (1987), “Credit policy and economic activity in developing countries with
       IMF stabilization programs,” Studies in International Finance, No. 60 (Princeton
       University, Princeton).

Harrison, Fidel E., (1993), “Structural re-adjustment in Nigeria,” American Journal of
       Economics and Sociology, 52, 193-207.

Henkel, R., (1976), Test of significance (Sage, Beverley Hills).

Igbedioh, S. O. and J. B. I. Aderiye, (1994), “Increase in food prices and food
       consumption pattern in some university students in Makurdi, Nigeria,” Ecology of
       Food and Nutrition, 31 , 219-226.

KaJisa, K., M. Mardia and D. Boughton, (1997), “Transformation versus stagnation in
       the oil palm industry: a comparison between Malaysia and Nigeria,” Staff Paper
       No. 97-5 (East Lansing: Michigan State University).

Kelly, Margaret R., (1982), “Fiscal adjustment and fund-supported programs,” IMF
       Staff Papers 29, 561-602.

Khan, M., (1990), “The macroeconomic effects of fund supported adjustment
      programs,” IMF Staff Papers, 37, 195-231.

Khan, M., and D. M. Knight, (1981), “Stabilization program on developing
      countries: a formal framework,” IMF Staff Papers 28, 1-53.

Khan, M., and D. M. Knight, (1985), “Fund-supported adjusted programs and economic
      growth,” Occasional Paper 41, (IMF, Washington, D.C.).

Kim, J. and F. J. Kohout, (1975), Special topics in general linear model, in:
       N. H. Nie, C. H. Hull, J. G. Jenkins, K. Steinbrenner, and D. H. Bent (Eds.),
       SPSS: Statistical Package for Social Sciences (McGraw-Hill, New York) 368-

Loxley, J., (1984), The IMF and the Poorest Countries (North-South Institute, Ottawa,

McClosky, D., (1985), The rhetoric of economics (University of Wisconsin Press,

Momoh, S., (1995), “Food production policies and nutritional status in Nigeria,”
     Agricultura Tropica et subtropica, 28, 15-22.

Nwosu, A. C., (1992), Structural adjustment and the Nigerian agriculture: an
      initial assessment, Staff Report No. AGES 9224 (USDA-ERS, Washington,
       D. C.)

Olomola, S. A., (1995), “Financing agricultural development in Nigeria: issues and
      policies,” presented at a seminar sponsored by the Dept. of Agric. Sc., Tuskegee
      University, Tuskegee, Alabama.

Olomola, S. A., (1994), “Changes in rural and agricultural credit policies under
      structural adjustment in Nigeria,” Journal of International Agriculture, 33, 23-34.

Oshikoya, T. W., (1990), The Nigerian economy: a macro-econometric and input-
      output model (Preager, New York).

Pastor, M., (1987), “The effects of IMF programs in the third world: debate and
        evidence from Latin America,” World Development 15, 249-262.

Rajaonarivony, N., (1996), The examination of the impact of IMF-supported
      program on the economic performance of low-income countries: the case of
      Madagascar Ph.D. Dissertation. Auburn University.

Reed, D., (1996), “Structural adjustment, the environment, and sustainable development,”
       The Courier ACP-EU No. 159, 82-83.

Reichman, T. M., (1987), “The fund conditional assistance and the problem of
      adjustment, 1973-75,” Finance and Development 15, 38-41.

Reichman, T. M. and R. Stillson, (1978), “Experience with programs of balance of
      payments adjustments: stand-by arrangements in the higher tranches, 1963-72,”
      IMF Staff Papers 25, 293-307.

Ringquist, E.J., (1994), “Policy influence and policy responsiveness in state control,”
      Policy Studies Journal 22, 25-43

Todaro, M. P., (1997), Economic development 6th (Ed.), (Addison-Wesley Publishing
      Company, New York)

Zulu, J. B. and S. M. Nsouli, (1985), “Adjustment programs in Africa: the recent
        experience,” Occasional Paper 34, (IMF, Washington, D.C.).

                                         Appendix A

                                    Stages of Relationships

1. Influence of SAP on agricultural production

               AGP = f (SAP)

       Direct effect: bivariate r
                          .85                                                     .85

2. Influence of SAP on net agricultural exports

           NAGEX = f (SAP, AGP)

           Direct effect: beta of SAP
           Indirect effect: bivariate r of SAP-AGP x beta of AGP ) .45 + (-.31)   = .14
                                    .85                      -.37

3. Influence of SAP on contribution of agriculture to GDP

            CAGGDP= f (SA, AGP, NAGEX)

       Direct effect: beta of SAP

       Indirect effect: bivaraite r of SAP-AGP x beta of AGP x beta of NAGEX

                           .85                      -.47              .48    .25 + (-.19)

                                                                              = .06

4. Influence of SAP on real GDP growth rate

               RGDPGR = f (SAP, AGP, NAGEX, CAGGDP)

       Direct effect: beta of SAP

       Indirect effect: bivariate r of SAP-AGP x beta of AGP x beta of NAGEX
                                   .85                   .36            .60
                           x beta of CAGGD

5. Total Causal Impact of SAP on RGDPGR equal to:

      Direct causal: beta of SAP in 4, -.36 plus
      Indirect causal: through AGP in 1, .85, through NAGEX in 2, .14, and
                        through CAGGDP in 3, .06 (i.e., -.36 + 1.05 = .69

           Figure 1. Diagram of the Research Model

                         Agricultural Production
   +                                                                  +

                                                            Net Agricultural

Adjustment Program

                                                           Contribution of
                                                         Agricultural to GDP

                               Real GDP
                     +                               +

                                     Table 1

              Bivariate Correlation of SAP Dummy with Other Variables

Variables                                              r

Agricultural Production                               .85

Net Agricultural Exports                               .14

Contribution of Agriculture to GDP                    -.08

Real GDP Growth Rate                                   .02

                                       Table 2

        Results of the Multiple Regression Analysis on Net Agricultural Exports

Independent Variables                                   beta

SAP (Dummy)                                             .45

Agricultural Production                                 - .37

                                       Table 3

   Results of the Multiple Regression Analysis on Contribution of Agriculture to GDP

Independent Variables                                   beta

SAP (Dummy)                                             .25

Agricultural Production                                 - .47

Net Agricultural Exports                                 .48

                                     Table 4

        Results of the Multiple Regression Analysis on Real GDP Growth Rate

Independent Variables                                  beta

SAP (Dummy)                                          - .36

Agricultural Production                               .36

Net Agricultural Exports                              .60

Contribution of Agriculture to GDP                   .12

                                         Table 5

      Causal Impact of SAP in the Agricultural Sector on Real GDP Growth Rate

Relationships                                      Partial             Total

Direct Causal                                                            -.36

Indirect Causal                                                          1.05

       Through Agricultural Production                 .85

       Through Net Agricultural Exports                 .14

       Through Contribution of Agriculture to GDP       .06

Total Causal                                                             .69