Impacts of Corruption to Agricultural Export Potential of the Gambia Competitive Neighbours

The purpose of this study was to investigate the mutual interaction between corruption and agricultural export variations with using corruption perception index, exchange rate and gross domestic product affect. Study, centers the Gambia country, but her neighbour and competitive countries in order to make comparisons. Study constructs a multinominal logit model to analyze the determinants of agricultural export variations and depends on panel data belongs to six Sub-Saharan African countries. Study shows that, the corruption perception index had a probality on agricultural export. Value of corruption production index in the previous year was likely to contribute to the current agricultural export of Sub-Saharan Africa countries. Increases at Gross Domestic Product provide benefits to agricultural export in parallel with corruption perception index. Policy makers, trade sector, business and civil society movements and governmental approaches have significance on preventing corruption climate atmosphere. To the best of authors knowledge, this study constitutes the first attempt to build a theoretical framework to explore how the interactions between determinants of agricultural export variations and corruption inflluence.


INTRODUCTION
The countries located in Sub-Saharan Africa have pursued other developing regions of the world in the growth for the last 50 years. The region is one of the poorest regions in the world and remained poorer as the rest of the world develop faster. A dramatical describtion of Sub-Saharan Africa's historical situation in a report of International Monetary Fund (2000)as follows: "...the economic and social situation in sub-Saharan Africa remains fragile and vulnerable to domestic and external shocks, and the region has a long way to go to make up for the ground lost over the past two decades. Despite some upturn in economic growth rates, poverty is still widespread and in many parts of the continent extremely acute. Investment remains subdued, limiting efforts to diversify economic structures and boost growth. Furthermore, a number of countries have only recently emerged from civil wars that have severely set back their development efforts while, sadly, new armed conflicts have erupted in other parts of the continent. These conflicts and other adverse factors, notably poor weather conditions and a deterioration in terms of trade, have led to some loss in economic momentum in the region over the past two years".
The phrases that was put forwarded in the above paragraph are including delicasy, poverty, weather conditions and weak economic growth. If we take historical economic aspects as a guide for what is possible in the future growth condition, we can infer that global inequality will remain high for a long time. In order to understand clearly, it should be known of how long would it take for those with per capita incomes at the poorest 10%, having 480 USD/year, boundary to achieve the per capita incomes of the richest 10% boundary, having 14500 USD/year. The richest 10% boundary level is generally the level of GDP per capita above which the poverty headcount gets close to 0% for most countries. Unfortunately, the answer is not at all desirable. Even under a very optimistic scenario, it will take several decades for the less developed countries such as sub-Saharan African countries to reach the income level of the global top 10%. If those countries grow 2%, the answer is 172.1 years while they grow 6% it is 58.5 years. However, we have not observed growth rates of more than 6% that will enable reaching the level of intended per capita income levels for a long time in the region (Anonymous, 2019). When anthropological factors such as domestic conflicts and corruption added to all of these, a more pessimistic picture emerges. Then there is no other way to return to the beginning of agricultural development. This fundamental field of a usual economy, unfortunately, is not able to adequately support the subsequent sectors and of course larger economic growth path in sub-Saharan African economies. Even though agriculture has substantially contributed to the GDP and export value in African countries, the continents share in the global markets, and the annual growth rates have diminished and stagnated over the years. Global agricultural exports by region showed that Europe (40.8%) sustained the highest share in the world, followed by Asia (22.4%) and North America (15.7%) and South and Central America (12%). Yet, Africa, which has heavily depended on agriculture for food, economic growth, and development, merely accounted for 3.6 of total global food exports in 2014 (Verter, 2017).
All of these can be explained by low trade liberalization. Indeed, there is a known phenomenon that trade liberalization and integration to foreign markets play an important role in economic development. Although sub-Saharan Africa countries have generally explored a liberalization route over the past two decades, their economic performance has been disappointing as stated above. Indeed, trade liberalization has been a key part of the policy prescriptions often made by international financial institutions to support economic development. Yet, it is almost impossible to achieve intended conclusions with more than 80 per cent of the world's population lives in over 100 developing countries that produce less than 20 per cent of the world's products and services (Desta and Hirsch, 2012).
As can be seen, domestic and external factors put the Africa economies on the bottleneck. Resolving external factors is not a matter of an individual country alone. However, each country may progress in domestic factors such as domestic conflicts and corruption in order to reach a particular level of development. Current study considers the corruption effects of this domestic factors on agricultural exports in the selected sub-Saharan countries, competitive ones, by featuring the Gambia.
The corruption is stated as the abuse of entrusted power for private gain, and the existence of the corruption depends of three elements, which are benefits, abuse of power and private-public sector. The corruption is mostly linked with organizational structure of a state in terms of economic and managerial. It is seen as basic impediments to the governmental efficiency. Indeed, private sector or governments can be conceived to a symptom that something has gone wrong in the management of the state. Advices on sound policies, well-designed incentives and efforts may not reach the desired results when the corruption starts and grows. Finally, economic inefficiency hit the subsistence people living in an actually poor country. Corruption draws attention many policy-makers and scientific researchers across the world, that has been explained by an index since 1995. The Corruption Perception Index (CPI) is an index published annually by Transparency International which ranks countries by their perceived levels of public sector corruption as determined by expert assessments and opinion surveys.
There are more than twenty reasons of the corruption, and its reasons and effects are investigated by several authors in the literature. Freedom in press, theoretically, provides anticorruption norms and expose the corrupt behavior (Treisman, 2000;Bhattacharyya and Hodler, 2015). Freedom of the press is more of an issue affecting the general economy, such as economic growth, investment and foreign direct investment, FDI, (Zakharov, 2019). Therefore, in specific studies, such as trade-corruption relationship, statistical significance may be lower.
Excessive bureaucracy, insufficient efforts on trade simplification and weak political management are elements of corruption. Existence of those elements may give a kind of monopoly power to the officials (Kaufman and Wei, 1999). An African study stated some limits to the FDI in post-colonial Africa (Mlambo et al., 2019). Liberal economic policies giving chance how to produce, sell and use your own production lead to face lower-level corruption in a usual country. Indeed, investigators found an evidence with a negative association between more liberal economic policies and corruption levels (Paldam, 2002;Saha et al., 2009;Musila, 2019). Economic growth is one another topic on the corruption researches. Some authors suggest that corruption slows down growth because investors hesitate to embark new economic actions (Pellegrini and Gerlagh 2008;Cieslik and Goczek, 2018). Some authors also suggest that growth hampers the corruption due to strong government institutions and their efficiency (Pose and Zhang, 2019).
Dincer, (2008) studied on ethnicity and corruption relationship and observed that larger ethnic groups in a country may increase the corruption due to group favoritism. The resources and opportunities are favorably allocated to group members (Seim and Robinson, 2019). Women employment contributes business world in terms of productivity, patience, saving and acumen behaviors. It is expected that women will be less prone to corruption behavior (Glover et al., 1997). Globalization and size of the country usually shows correlation. A larger country integrates more into the world to meet its needs. Foreign capital inflow may bring to a certain extent structural activity and higher norms (Sandholtz and Koetzle, 2000;Cai et al., 2018). Baklouti and Baujelbene (2016) support this phenomenon indeed. They examined country size-corruption relationships and found that government size can lead to a decrease in corruption if the democracy level is sufficiently high and, in contrast, can lead to an increase in corruption if it is too low. Decentralized government structure draws attention to corruption researchers. It is briefly explained by dispersed decision-making process that prevents government-originated distortions. This may protect many economic agents exposed to welfare losses (Blackburn and Gonzalo, 2010;Neudorfer and Neudorfer, 2015;Efobi, 2015). The connection of governmental regime types and corruption has received research interest. It is a widespread phenomenon that there has been negative relationship between two variables. Regular voting raises the democratic conditions and increases corruption control (Persson and Tabellini, 2000). In addition, if fair elections are experienced in a country, the officials who are capable of performing governmental services and live away from corruption are likely to remain in their critical positions (Nur-teging and Czap, 2012). Historical factors such as colonization, religion, legal systems, and political stability are also investigations by several authors. Treisman (2000) states that British Colonies had lower-level corruption compared to Spanish, French and Portuguese ones. Uberti (2018) finds evidence that the corruption observed in Eastern Europe, Central Asia and the Balkans did not arise from socialist or Ottoman legacies. Excessive and uncontrolled competition makes the market intensify, and competent firms may engage in corrupt or unethical activities (Dimant and Tosato, 2017). This is associated with an increase in the level of corruption. Political competition also facilitates the businesses of certain groups and increases the tendency of corruption. In fact, public and private sector employees who finds their governmental organizations less ethical and undisturbed are prone to corruption (Gorsira et al., 2018). Corruption researchers find evidences that community structure such as poverty may trigger the corruption. Weak legal system cannot prevent corruption and they are desensitized to be more prone to accept and give bribe all socio-economic life (Apergis et al., 2010). Yet, it is stated that such tendencies may arise as a result of the injustice of income and the interaction of poverty (Peiffer and Rose 2016). Urbanization have communities lose the social control of family and religion and therefore it provides the necessary conditions for corruption. (Holbook and Meier, 1992). However, there are also opposite approaches. It has been suggested that stigmatization could be observed in urban society where urbanization and corrupt applications can be detected more easily (Goel and Nelson, 2010).
As can be seen in the literature review, although there are many reasons for corruption, there is no consensus above the reasons yet. Even if there is a general evidence that corruption reasons depend on related region, these reasons may affect each other by influencing one another. In other words, one reason for corruption may be the result and cause of another corruption reason. The Gambia, which was the subject of this study, is an agricultural country under the influence of domestic factors such as corruption. However, income from agriculture is low. Moreover, the effective distribution of these revenues is one of the problems that the Gambia has to solve as many domestic problems.
Unlike the current literature, we assume a statistical probability that all these corruption's causes will gather and affect the agricultural trade system of the subsequent year. We employed the CPI, exchange rate, very useful to explain the export, and Gross Domestic Products (GDP). Variations at agricultural export is explained by a probability involving multiple choices, and it is stated with multinomial logit model (MNL). Current study using a MNL model to analyze the determinants of agricultural export variations are different from former econometric techniques that prefer Ordinary Least Squared (OLS) regression.

SOCIO-ECONOMIC BACKGROUND OF THE GAMBIA
The Gambia is located west sub-Saharan Africa and surrounded by Senegal. She has a narrow economic base, relying heavily on agriculture which provides employment for about 75% of the labour force and accounting for 70% of the country's foreign exchange earnings (Otaiku 2018). Performance of the sector has fluctuated, contributing on average 23 per cent of Gross Domestic Product (GDP) between the years of 2004 and 2016 and an average growth rate of 4 per cent per year. Despite its potential, agricultural production evidences low and unpredictable yields and high susceptibility to droughts and erratic climate patterns. The sector is basically rain-fed, with only 3% of the arable land estimated under irrigation. The prevalence of drought as a climatic phenomenon was very severe in 2011 to the extent that the gains in the sector were almost completely wiped-off in a single year's drought situation. However, Gambian agriculture is mainly characterized by farmers who grow food mainly for family consumption as a result leaving little for commercial purposes. This inadequate volume of production is mainly due to the retarded progress in farming methods and scattered pieces of land holdings. Most of the farm land is cultivated by small scale farmers with traditional agricultural practices.
Smallholder farmers are caught in a vicious cycle of risks, limited use of inputs, low productivity and low income. The sector is predominantly subsistence, rain fed with very little irrigation or use of improved seeds and fertilizers. In regions where population growth is rapid and rural population density is high, the size of the average household's farming system has been rapidly declining. Thus, exposing The Gambia to be considered as a country where food insecurity has become endemic owing to repeated incidence of crop failure, incidence of animal disease outbreak, rising food prices and the lack of adequate support mechanisms to victims (Gibba, 2017).
In the 2016 fiscal year, the agricultural, industrial and service sectors contributed 19 percent, 13 percent and 62 percent to real GDP of the country, respectively (GBOS, 2016). The share of the agricultural sector to the whole economy is decreasing from time to time, whereas the growth in service sectors increasing at high rate while the industrial remains constant.
Despite the agricultural sector's decreasing contribution to the overall GDP of the country, it is still the country's biggest employment provider and contributes the largest share of raw materials for local industries The economy of the country is mainly dependent on the agricultural sector thus the export sector is also highly dependent on agricultural productivity. Since the export sector is characterized by dependence on primary commodities, the country faces different problems such as: the low volume of exportable products, the limited degree of diversification of exports, frequent economic crises and artificial trade barriers by trading partners among others.
In order to increase the contribution of the agriculture sector various measures have been taken by the government. However, agricultural exports have been largely concentrated on three agricultural products -Groundnuts, Cashew nuts and Fisheries -which are incessantly subjected to fluctuations in quantity, price and low competitiveness in the international market. Countries competing with Gambia in agricultural exports are Senegal, Ghana, Nigeria, Tanzania and Guinea.

MATERIAL and METHOD
If the corruption is closely linked to the organizational structure of a country, we should also be suspicious of the fact that it may affect export facilitates. Traditionally, the determinants of a country's export potential are evident, and those are can be listed as exchange rate, change in economic growth, historical experience of the country, openness of the trade and geography of the country. In addition to all of these, corruption can be an important factor that is justified as in the previous sections. In fact, the organization of economic activities is also implicitly located in the CPI. In agricultural related export, natural conditions may become determiner of a country's Agri-export performance. This foresight varies according to geographical location in the countries where similar crops are cultivated. Yet, climatic conditions are resemble for many countries located at Sub-Saharan Africa Region. For this reason, exchange rate, CPI and economic growth which are the closest variables to the possibility of being an important determinant of agricultural exports were taken into consideration in this study.
As we may understand from the existing literature, different authors find different results on the same reason that explains the corruption. In fact, vast majority of abovementioned 'corruption reason set' having more than 20 reasons is linked with economic growth. As known, agricultural growth and export are main drivers of economic growth in the countries where there is no alternative other than agriculture. Obviously, one or more reasons feature that may raise KSÜ Tarım ve Doğa Derg 24 (4): 886-894, 2021 KSU J. Agric Nat 24 (4): 886-894, 2021

Araştırma Makalesi
Research Article or decrease the impacts of one another reason. In other words, the variations on the agricultural exports, depending tightly on weather conditions, are explained by the impacts of several probabilities. We predict that the variations at CPI, exchange rate and GDP affect subsequent year's agricultural export change. At this point, we contribute to existing corruption literature. CPI score calculation flows four basic steps: selection of source data, standardize data sources, aggregating the rescaled data and then reporting a measure for uncertainty. The CPI draws upon 13 data sources which capture the assessment of experts and business executives on a number of corrupt behaviours in the public sector, including: bribery, diversion of public funds, nepotism in the civil service and state capture. Each of this data soruces using to calculate the CPI. Then this data sources is standardized to allow for the aggregation into the CPI score. The standardisation converts all the data points to a scale of 0-100 where a 0 represents the highest level of perceived corruption, and 100 the lowest level of perceived corruption. After this, each country's CPI score is calculating as a simple average of all the available rescaled scores for that country. At the end, the CPI score is reporting alongside a standard error and 90 per cent confidence interval which reflects the variance in the value of the source data that comprises the CPI score(Transparency International, 2019).
Current study centers the Gambia country, but her neighbor and competitive countries, abovementioned, are also examined in order to make comparisons.
As distinct from general logistic regression modelling framework, several other modelling frameworks can be used to the model relationship between categorically dependent variable and a number of independent variables (Motsoari et al., 2015). In this study, variations at agricultural export is explained by a probability involving multiple choices, and it is stated with MNL. Current study using a MNL model to analyze the determinants of agricultural export variations are different from former econometric techniques that prefer OLS regression.
Current study depends on panel data belongs to six sub-Saharan African countries, and the variations at agricultural exports figures is a probability having three alternatives. It may be increased, decreased or not changed, and these are explained by J=3 alternative codes which are 1 if rising is observed, 0 if decline is observed and 2 if there is no change in the ith year. Those numerical values have no meaning, and they are assigned arbitrarily. Explanatory variables are also categorical variables that are CPI, exchange rate and economic growth rate in the i-1th year. We employed one-year-previous value because it may affect current export value. As in the dependent variable export value change, explanatory variables' alternatives are coded as 1 if rising is observed, 0 if decline is observed and 2 if there is no change in the ith year. Let i A be a random variable symbolizing the export variation value observed by any investigated country in related year. We assume that each export variation faces a set of discrete variables of the CPI, GDP rate and exchange rate. The MNL model for agricultural export variation defines the following relationship between the probability of observing option i A and the set of explanatory variables X as (Hill et al., 2011): where βj is a vector of coefficients on each of the independent variables X. If it is assumed that a single explanatory factor, xi, then, in the multinomial logit specification, the probabilities of observing alternatives j = 1, 2, 3 at ith year are: The parameters 12 and 22 are specific to the second alternative while 13 and 23 are specific to the third alternative. The parameters specific to the first alternative are set to zero to solve an identification problem and to make the probabilities sum to one (Hill et al., 2011). Estimation of current model is by maximum likelihood.

RESULT AND DISCUSSION
The model was solved by SPSS statistical package that is widespread at the social sciences area. Table 1 gives the estimated and having statistical significance coefficients. Country specific models have k-1 dependent outcome where k is the number of levels of the outcome variable at the MNL estimation. Thus, software considers the option 2 as reference group. The code 2, as mentioned before, indicates the option that there is no any change at the export variation. In other words, the parameter estimates are relative to the reference group.
The Gambia model has statistical significance on "decreased export condition" relative to "unchanged export condition". According to intercept, the intercept is the multinomial logit estimate for the decreased export condition relative to unchanged export condition when the predictor variables in the model are evaluated zero. In this case, the probability will be unchanged side due to negative sign. The cpi parameter coded 1, indicates that when the cpi value increases a unit, log-odds of the decreased export KSU J. Agric Nat 24 (4): 886-894, 2021 Araştırma Makalesi Research Article condition relative to unchanged export condition would be expected to be 18.89 unit while holding all other variables in the model constant. However, same interpretation is made with the increased export condition relative to unchanged export condition in terms of the cpi. At this statistic, the cpi parameter coded 1, indicates that when the cpi value increases a unit, log-odds of the increased export condition relative to unchanged export condition would be expected to be 20.13 unit while holding all other variables in the model constant. In this case, probability has greater than previous one and, we can infer from this result that the cpi has a probability on agricultural export positively. The GDP parameter coded 1, indicates that when the GDP increases according to previous year, log-odds of the decreased export condition relative to unchanged export condition would be expected to be 17.68 while holding all other variables in the model constant. In other words, the increases at GDP provides benefits to agricultural export in the Gambia because the coefficient of GDP coded zero (39.04) is greater than that of GDP coded one.
The Senegal Model has statistical significance on the GDP coded 0. The parameter value indicates that declines at the GDP have probability on decreasing export condition relative to unchanged export condition. Then, the Gambian agricultural policy makers should consider the Senegal's GDP values while making competition because it can affect subsequent year's agricultural export rate.
The Ghana Model has statistical significance on the cpi both of coded 0 and 1. It means that the cpi of Ghana have similar probability on the increased export condition relative to unchanged export condition. The negative sign indicates that unchanged export conditions have probability at the Ghana. Likewise, if the cpi decreases at the Ghana, this effect has probability on unchanged agricultural export conditions.
The Guinea and Tanzania models have only statistical significance on the GDP coded 1 on the probability of decreased export condition relative to unchanged export condition. In the Guinea, it is explained that unchanged agricultural export conditions occurs when the increases are observed at the GDP. Tanzania model, contrary to Guinea model, the GDP coded 1 is explained that probability of decreased agricultural export conditions occurs when the increases are observed at the GDP. In others word, GDP have positive impacts at Guinea agricultural export while Tanzania experiences inverse probability condition.
The Nigeria model have significance on the cpi values both of probability of "decreased export condition" relative to "unchanged export condition" and "increased export condition relative to unchanged export condition". They have symmetric results that the cpi values have opposite signs. Yet, unit changes point out the zero coded cpi parameter value on the probability of decreased export condition because the coefficient of this probability have greater value than that of coded 1.
According to the results of this study which is supported by a small number of authors in the literature review, it is possible that the value of cpi in the previous year is likely to contribute, unfortunately, to the current agricultural export of sub-Saharan Africa countries as explained in (Anderson & Marcouiller, 2002). In fact, some regions and countries' corruption record is not clean-cut when the historical background is examined. In the Nigeria's oil export created opportunities in the 1980s. Other cases are come from Eastern Europe and former the Union of Soviet Socialist Republic. It is because foreign investors search stronger bargaining power relative to domestic investors (Torrez, 2002). The motivation of foreign investor grab is observed in transitional economies. Although the Gambia is typical lessdeveloped country, especially, security issues and unsupervised conditions may have occurred in the export path.
Pervasive and cumbersome governmental conditions prevail in corruption-efficient countries. "Speed money" phenomenon is often observed in such countries. It is a kind of bribes that works kickback like legal for piece rate paying to government employees. The speed money may play significance roles to bypass bureaucratic delays.
However, this situation, which initially seems innocent, is not suitable for both domestic and international law. There are also views about the loss of efficiency in the economy (Whalley, 2003). In addition, if the fact that the people who are willing to do business without any effort and the people who are willing to do business is admitted same by related community, the sustainability of her economic life will put in jeopardy. This kind of unfair resource transfer becomes a hidden tax loaded on transaction cost of a country and may reduce its international trade share in the medium and long term.
Such a diverse effect of corruption between domestic and foreign stakeholders justifies strong corruption reducing policies in relatively more open economies.

CONCLUSION
Despite the important contribution to national GDP, the Gambia's agricultural export sector continues to perform far below its optimum potential. Its performances are constantly undermined by series of challenges that make it difficult to compete with other exporting nations hence losing great economic benefits. Today, several challenges trapped inside to the corruption phenomenon remain the Gambia in a very difficult position in the international trade. Some of these challenges include: domestic security, marketing difficulties due to inadequate transport facilities, perennial technical problems in the groundnut processing factories and facilities, low producer prices, frequent credit buying due to inadequate crop financing capacity.
Reducing the challenges arising from corruption is possible and it needs the establishment of an interinstitutional collaboration that would help support the agricultural trade sector. This will bring together both government and private institutions, NGOs as well as civil society organizations to formulate policies that will help boost the exportation of agricultural products. Research and development institutions will also have a big role to play in terms of conducting market researches and help in selecting crops that best adapt to climatic conditions such as drought and pest resistant varieties.
Policy makers, trade sector, business and civil society movements could be the key strategic institutions through which agricultural product export can be improved, governmental approaches have significance on preventing corruption climate atmosphere.
Gambia has little chance of competing in agricultural products trade with globalization and other fastgrowing economies. However, with the improvements in the organizational structure, it can bring foreign trade to effective levels. Processed products industry should be developed on an industrial project basis. The share of basic agricultural products in the value chain is quite low. The transition to partial organic farming can provide certain advantages. It can provide privilege in international agricultural products trade.
Besides the effect of the corruption on the economy, cleaning it is also an important cost. It means a serious monetary investment and time. However, initial costs may return more favorably in the future.

Statement of Conflict of Interest
Authors have declared no conflict of interest.

Contribution of the Authors as Summary
Authors declares the contribution of the authors is equal.