# ABSTRACT:

Kazakhstan is located in the hinterland of Central Asia. Its virtuous geographical advantages and huge grain production potential make it one of the most important grain exporters in the world. The research on the problem of the grain trade in Kazakhstan is of great significance for food security. This study measured its international competitiveness using the International Market Share Index, the Revealed Comparative Advantage Index, Trade competitiveness index and calculated the international competitiveness and analyzed the influencing factors of grain export by constructing an extended gravity model and measured its export potential. Results showed that Kazakhstan has a low share of the international grain market; however, wheat, barley, and buckwheat have strong export advantages; the level of economic development and economic distance has significantly promoted the scale of grain exports. While geographical distance, the difference in GDP per capita, and the fact whether trading partner countries have joined the Eurasian Economic Union have caused obstacles to grain exports. Kazakhstan’s export potential to 6 countries including Russia, Kyrgyzstan and China shows an upward” trend, its export potential to 6 countries including Tajikistan and Ukraine showing a “stable” trend, and its export to 9 countries included Poland and Germany. The potential showed a “declining” trend.

Key words:
grain; potential for trade; international competitiveness; Kazakhstan

# RESUMO:

Palavras-chave:
grãos; potencial comercial; competitividade internacional; Cazaquistão

# INTRODUCTION:

Grain import and export trade is an important means to regulate the global food supply-demand imbalancing and maintaining national food security. The trade of grain and other agricultural products has become more and more important as a way to adjust resources between countries with abundant resources and countries with scarce resources (QIANG et al., 2013QIANG, W., et al. Agricultural trade and virtual land use: The case of China’s crop trade. Land Use Policy, v.33, p.141-150, 2013. Available from: <Available from: https://doi.org/10.1016/j.landusepol.2012.12.017 >. Accessed: Oct. 20, 2020. doi: 10.1016/j.landusepol.2012.12.017.
https://doi.org/10.1016/j.landusepol.201...
). International food prices have fluctuated sharply affected by factors such as novel coronavirus pneumonia (COVID-19), the tight relation of global food production and demand, macroeconomic, and geopolitical changes in recent years. The prices of major grains such as corn, soybeans, and wheat have risen and fallen sharply, which had a greater impact on ensuring national food security.

Kazakhstan, located in the hinterland of Central Asia, is the world’s largest food producer and exporter. According to FAO, Kazakhstan’s total grain output in 2018 was 20.04 million tons, ranking 28th in the world; total grain export was 8.53 million tons, ranking 13th in the world. The “2019 Global Food Security Index Report” shows that Kazakhstan’s food security level ranks 48th among 113 countries in the world, while in 2018 it was ranked as 58th (The Economist-Intelligence Unit, 2019). In recent years, Kazakhstan’s grain export trade has developed promptly and turn out to be an important potency in the international grain market. According to UN Comtrade Database, Kazakhstan’s total grain export volume reached 1.304 billion USD in 2018, with an increase of 57.59% over 2017. Among them, wheat was the most important grain export product, with an export value of 972 million USD, accounting for 74.54% of Kazakhstan’s total food export; the export value of barley was 294 million USD, accounting for 22.55%; the export value of rice was 26 million USD, accounting for 2.01%. The export volume of these three major categories of grain crop products accounted for 99.1%. In terms of time series (Figure 1), the total export value of wheat during 2008-2018 fluctuated dramatically, dropping from 1.459 billion USD in 2008 to 972 million USD in 2018, reaching the highest in 2012. Barley export was on a steady growth trend, increasing from 157 million USD in 2008 to 294 million USD in 2018. The export value of rice remained unchanged basically, less than 400 million USD throughout the time.

Figure 1
Main grain production, exports value, and volume of Kazakhstan between 2008 and 2018.

From the perspective of spatial pattern, the grain export market presented the regionally concentrated geographic patterns of “central Asia as the main area, radiating all around”. Meanwhile, Uzbekistan, Tajikistan, China, and Iran occupied the position of important trading partners. In 2018, the main destinations of Kazakhstan’s wheat exports were Uzbekistan, Tajikistan, China, Italy, Afghanistan, and Turkey, which accounted for 73.59% of Kazakhstan’s total wheat export volume together. The main export market for the barley was concentrated in Iran, and a small part was exported to Russia, Germany, the United Kingdom, and other western developed countries. In 2018, Kazakhstan exported 272.4 million USD of barley to Iran, accounting for 92.80%; 120 million USD of barley was exported to Uzbekistan, accounting for 4.08%; Asian countries accounted for 99.49% of Kazakhstan’s total barley export volume together (Figure 2).

Figure 2
The geographic pattern of Kazakhstan’s wheat exports in 2018.

Good geographical advantages and huge grain production potential make Kazakhstan one of the most important grain exporters in the world. It is considered by the Food and Agriculture Organization of the United Nations and the European Bank for Reconstruction and Development to be one of the only four countries in the world whose food production capacity is underutilized and can have a significant impact on meeting global food demand. However, an adjustment in agricultural policy as a result of changes in the market system has led to a significant decrease in the total amount of agricultural land used in Kazakhstan, with a 35% reduction in the area of arable land since independence in 1992 (SWINNEN & VAN HERCK, 2011SWINNEN, J., VAN HERCK, K. Food security and the transition region. Open Access Publications from Katholieke Universiteit Leuven. 2011. Available from: <Available from: https://www.ebrd.com/downloads/research/essay/foodsecurity.pdf >. Accessed: Oct. 20, 2020.
). The decrease in the area of arable land, instability of grain trade development, the extensive growth of grain trade, and the low contribution to the development of agricultural industry affect the optimization of grain trade structure and the development of grain trade potential in Kazakhstan (KHAN et al., 2017KHAN, Y., et al. Kazakhstan: Transport and Logistical Risks in Grain Export. SSRN Electr-onic Journal. 2017. Available from: <Available from: https://doi.org/10.2139/ssrn.2989522 >. Accessed: Oct. 20, 2020. doi: 10.2139/ssrn.2989522.
https://doi.org/10.2139/ssrn.2989522...
). At the same time, the high inland transportation cost also led to a relatively single market for grain exports, which weakened Kazakhstan’s competitiveness in international trade.

The research on Kazakhstan’s grain mainly includes three areas: grain production, grain trade, and food security. In the field of grain production, existing studies focused on the current situation, production potential, and influencing factors of food production in Kazakhstan. As the largest grain-producing country in Central Asia, Kazakhstan has abundant cultivated land resources (XIN et al., 2019XIN, W., et al. Temporal and Spatial Dynamics Analysis of Grassland Ecosystem Pressure in Kazakhstan. Journal of Resources and Ecology, v.10, p.667, 2019. Available from: <Available from: https://doi.org/10.5814/j.issn.1674-764x.2019.06.012 >. Accessed: Oct. 30, 2020. doi: 10.5814/j.issn.1 674-764x.2019.06.012.
https://doi.org/10.5814/j.issn.1674-764x...
). However, due to the limited level of agricultural modernization, crop growth is highly dependent on the weather, leading to large fluctuations in grain production (PAVLOVA et al., 2014PAVLOVA, V. N., et al. Modelling the effects of climate variability on spring wheat productivity in the steppe zone of Russia and Kazakhstan. Ecological Modelling, v.277, p.57-67, 2014. Available from: <Available from: https://doi.org/10.1016/j.ecolmodel.2014.01.014 >. Accessed: Oct. 30, 2020. doi:10.1016/j.ecolmodel.2014.01.014.
https://doi.org/10.1016/j.ecolmodel.2014...
). At the same time, factors such as agricultural policies, damaged infrastructure, insufficient storage capacity, and the use of fertilizers have also had a significant impact on Kazakhstan’s grain production (ZAVALIN et al., 2018ZAVALIN, A.A., et al. Fertilizer nitrogen use. by Spring Triticale and Spring Wheat on Dark-Chestnut Soil of the Dry Steppe Zone of Kazakhstan. Russian Agricultural Sciences, v.44, n.2, p.153-156, 2018. Available from: <Available from: https://doi.org/10.3103/S1068367418020209 >. Accessed: Oct. 30, 2020. doi: 10.3103/S1068367418020209.
https://doi.org/10.3103/S106836741802020...
). It is estimated that the maximum potential area of wheat harvest in Kazakhstan is 190,000 ha, the maximum yield potential is 1.6 tons per hectare and the total amount is 29 million tons (ZHU, 2014ZHU, Z.. Chin utilize the Wheat Market of Russia, Ukraine and Kazakhstan:Export Potential and Barriers. Issues in Agricultural Economy, v.35, n.4, p.42-50, 2014. Available from: <Available from: https://doi.org/10.13246/j.cnki.iae.2014.04.007 >. Accessed: Oct. 30, 2020. doi: 10.1324 6/j.cnki.iae.2014.04.007.
https://doi.org/10.13246/j.cnki.iae.2014...
). Kazakhstan will be able to further improve the natural disaster resistance for grain production and the efficient utilization of natural resources (CAO et al., 2011CAO, S., et al. Export competitiveness of Agri-Products Between China and Central Asian Countries: A Comparative Analysis. Canadian Social Science, v.7, n.5, p.48-54,2011. Available from: <Available from: https://doi.org/10.3968/j.css.1923669620110705.015 >. Accessed: Nov. 10, 2020. doi: 10.3968/j.css.1923669620110705.015.
https://doi.org/10.3968/j.css.1923669620...
).

In terms of grain trade, except for the reduction of grain crop production in 2010, it generally showed an increasing trend in other years. Based on meeting domestic food demand, Kazakhstan has considerable export potential (ZHANAKOVA et al., 2015ZHANAKOVA, M., et al. Modern State and Forecast of Food Production in Kazakhstan. Indian Journal of Science and Technology, v.8, n.S10, 2015.b. Available from: <Available from: https://doi.org/10.17485/ijst/2015/v8is(10)/85412 >. Accessed: Jan. 15, 2021. doi: 10.17485/ijst/2015/v8is(10)/85412.
https://doi.org/10.17485/ijst/2015/v8is(...
). In addition, factors such as food supply capacity, per capita income difference, labor force quantity, trade facilitation degree, foreign trade policy, and storage-logistics facility have also affected the expansion of grain trade in Kazakhstan to a certain extent (GRAFE et al., 2008GRAFE, C., et al. Beyond borders-Reconsidering regional trade in Central Asia. Journal of Comparative Economics, v.36, p.453-466, 2008. Available from: <Available from: https://doi.org/10.1016/j.j ce.2008.03.004 >. Accessed: Nov. 10, 2020. doi: 10.1016/j.jce.2008.03.004.
https://doi.org/10.1016/j.j ce.2008.03.0...
). China is Kazakhstan’s largest trading partner, and agricultural cooperation between the two countries also shows great development potential (RABALLAND & ANDRESY, 2007RABALLAND, G.; A, ANDRESY. Why should trade between Central Asia and China continue to expand?.Asia Europe Journal, v.5, n.2, p. 235-252, 2007. Available from: <Available from: http://doi.org/10.1007/s10308-007-0115-5 >. Accessed: Nov. 10, 2020. doi: 10.1007/s10308-007-011 5-5.
http://doi.org/10.1007/s10308-007-0115-5...
). However, the current China-Kazakhstan grain trade is small, and the trade varieties are particular due to factors such as the unstable agricultural production, insufficient grain storage, logistics capacity, low trade convenience, and the long-distance from the grain-producing area to China’s import area. The current total trade volume of major agricultural products between China and Kazakhstan is relatively small and the categories of agricultural products cross each other in which the two countries have comparative advantages in export. The complementarity between China’s exports and Kazakhstan’s imports is strong and on the rise as a whole, while the complementarity between Kazakhstan’s exports and China’s imports are low. The two countries have great potential for bilateral trade growth (LING WANG, 2015LING WANG. An Analysis of trade Structure,Comparative Advantage and Complementarity of Agricultural Products between China and the Main East Asian Countries.Asian Agricultu- -ral Research, v.7, n.5, p.14-20+24, 2015. Available from: <Available from: https://doi.org/ CNKI:SUN:AAG R.0.2015-05-004 >. Accessed: Dec. 10, 2020. doi: CNKI:SUN:AAGR.0.2015-05-004.
https://doi.org/ CNKI:SUN:AAG R.0.2015-0...
).

In terms of food security, most of the existing studies have expounded on the impact of Kazakhstan’s national food policy on the food security of the country and the world, from the perspectives of agricultural policy reform and temporary export restriction policies. At the current stage, Kazakhstan grain can be self-satisfied and can be exported for foreign exchange, so there is less pressure on food security. The main tasks of the Kazakh government are to improve the planting structure, increase production, maintain food price stability, and reduce inflation pressure for national food security (BAYDILDINA A et al., 2000BAYDILDINA, A, et al. Agricultural policy reforms and food security in Kazakhstan and Turkmenistan. Food Policy, v.25, n.6, p.733-747, 2000. Available from: <Available from: https://doi.org/doi. org/10.1016/S0306-9192(00)00035-X >. Accessed: Dec. 15, 2020. doi: 10.1016/S0306-9192(00) 00035-X.
https://doi.org/doi. org/10.1016/S0306-9...
).

The above research analyzed the problems of Kazakhstan’s grain production and trade from different perspectives, which provided a solid theoretical basis and practical reference for the study of this paper, but there are still some deficiencies. Based on the abovementioned research gaps, this article contributes to the field in two ways. Firstly, taking the temporal and spatial pattern of grain export trade in Kazakhstan as the starting point, through the calculation of the international market share, export advantage, and trade competitiveness of different grain crops, we can accurately grasp the current export situation and international competitiveness of all kinds of grain products in Kazakhstan, and provide basic data support for grain trade cooperation between Kazakhstan and other countries in the world. Secondly, by using the gravity model to calculate the potential and changing trend of Kazakhstan’s grain export trade, it can help us to find out the main reasons that affect the export potential and provide corresponding recommendations.

# MATERIALS AND METHODS:

International competitiveness model

The international competitiveness of grain exports not only reflects the production and export capacity of Kazakhstan’s grain but also an important indicator to measure its viability and trade status in the international market. Existing researches mainly use the International Market Share Index (IMS), the Revealed Comparative Advantage Index (RCA), and the Trade Competition Index (TC) to measure the international competitiveness of a certain product in a sustainable period of time.

International market share index (IMS)

IMS reflects changes in international competitiveness and the competitive position of a country’s certain industry or product. The calculation formula is:

${\mathit{IMS}}_{\mathit{ij}}=\frac{{X}_{\mathit{ij}}}{{X}_{\mathit{wj}}}\mathrm{}$

IMSij indicates the International Market Share Index, is the Xij total exports value of product j in country i, Xwj is the total export value of product j in the world. When IMSij>20%, the international competitiveness of product j in the country i is very strong; when 10%<IMSij<20%, it is relatively strong; when 5%<IMSij<10%, it is general; when IMSij<5%, it is weak.

RCA reflects the advantage degree of different products in a country’s foreign trade. The product with higher indices has better export advantages, its calculation formula is:

${\mathit{RCA}}_{\mathit{ij}}=\frac{{X}_{\mathit{ij}}/{X}_{i}}{{X}_{\mathit{wj}}/{X}_{w}}$

RCAij indicates the Revealed Comparative Advantage Index, Xij is the total exports value of product j in country i, Xi is the total export value of country i to the world, Xwj is the total export value of product j in the world, Xw is the world’s total export value. When RCAij>2.5, the export advantage of product j in-country is extremely strong; when 1.25<RCAij<2.5, the export advantage is very strong; when 0.8<RCAij<1.25, it is relatively strong; when RCAij<0.8, it is weak.

TC is a powerful tool to measure the international competitiveness of a certain product in a country. Its calculation formula is:

${\mathit{TC}}_{\mathit{ij}}=\frac{{X}_{\mathit{ij}}-{M}_{\mathit{ij}}}{{X}_{\mathit{ij}}+{M}_{\mathit{ij}}}$

TCij indicates the Trade Competition Index, Xij is the total export value of product j in country i, Mij is the total import value of product j in the country i. When TCij>0, it means that the production efficiency of product j in the country i is higher than the international level and has strong international competitiveness. When TCij=0, it means that the production efficiency of product j in the country i is equivalent to the international level, and its import and export are purely international exchanges of varieties. The value range of TC is from -1 to 1, among them, the higher TC value represents the stronger international competitiveness; otherwise, it indicates that a certain product or industry does not possess or lack international competitiveness.

Gravity model

The gravity model is one of the methods used to measure trade potential commonly, which takes influence factors of trade potential as explanatory variables, trade volume as an explained variable. This method uses regression analysis to determine the significant influence factors and the effective weight of each factor on the trade volume, then calculates the trade potential. The basic gravity model is:

${Y}_{\mathit{ij}}=\alpha \left({\mathit{GDP}}_{i}{\mathit{GDP}}_{j}\right)/{\mathit{Distance}}_{\mathit{ij}}$

Yij represents the total trade volume of exporting country i to importing country j, GDPi represents the economic aggregate of country i, GDPj represents the economic aggregate of country j, Distanceij represents the geographical distance between two countries, A is the regression parameter.

This paper uses the research of Aigner to construct a model of Kazakhstan’s grain export potential with random error (AIGNER et al., 1977AIGNER, D, et al. Formulation and estimation of stochastic frontier production function models. Journal of Econometrics, v.6, n.1, p.21-37, 1977. Available from: <Available from: https://doi.org/10. 1016/0304-4076(77)90052-5 >. Accessed: Dec. 15, 2020. doi: 10.1016/0304-4076(77)90052-5.
https://doi.org/10. 1016/0304-4076(77)90...
). This paper selects dummy variables such as economic development, population size, geographical and economic distances from Kazakhstan and its major grain trading partners to be explanatory variables, to study the influence of various factors on Kazakhstan’s grain export trade flow. In order to avoid the heteroscedasticity and non-normality of data residuals, the logarithmic gravity model is generally used in practical research. The estimated equation of the extended gravity model is:

$\mathrm{ln}{Y}_{\mathit{ijt}}={\beta }_{0}+{\beta }_{1}\mathrm{ln}\mathrm{ln}\left({\mathit{GDP}}_{\mathit{it}}{\mathit{GDP}}_{\mathit{jt}}\right)+{\beta }_{2}\mathrm{ln}\mathrm{ln}\left({\mathit{POP}}_{\mathit{it}}{\mathit{POP}}_{\mathit{jt}}\right)+{\beta }_{3}\mathrm{ln}\mathrm{ln}\left({\mathit{Gpcg}}_{\mathit{ijt}}\right)+{\beta }_{4}\mathrm{ln}\mathrm{ln}\left({\mathit{Dis}}_{i}\right)+{\beta }_{5}\mathrm{ln}\mathrm{ln}\left({\mathit{EF}}_{\mathit{jt}}\right)+{{\beta }_{6}\mathrm{ln}\mathrm{ln}\left({\mathit{ED}}_{\mathit{jt}}\right)+\beta }_{7}\mathrm{ln}\mathrm{ln}\left({\mathit{PD}}_{\mathit{jt}}\right)+{\beta }_{8}\mathrm{ln}\mathrm{ln}\left({\mathit{Lang}}_{j}\right)+{\beta }_{9}{\mathit{Bord}}_{j}+{\beta }_{10}{\mathit{Land}}_{j}+{\beta }_{11}{\mathit{WTO}}_{j}+{\beta }_{12}{\mathit{CIS}}_{j}+{\beta }_{13}{\mathit{EEU}}_{j}+{v}_{\mathit{ijt}}$

In the equation, i represents Kazakhstan, j represents Kazakhstan’s grain trading partner, t represents the year, Yijt represents Kazakhstan’s grain export trade value to importing country j, vijt represents the random disturbance term in the model. The detailed table is listed to further explain the meaning of each explanatory variable, expected influence direction, and related brief theory (Table 1).

Table 1
Meaning and theoretical description of each variable of the gravity model.

The applicability of the gravity model to study grain trade potential is mainly reflected in the following aspects: (1) Grain export is affected by factors such as the production capacity and product quality in the export country, the economic development and population size in the importing country, and the distance between the two countries, reflecting a gravity relationship; (2) To obtain the predicted value, Kazakhstan’s grain export potential needs to be calculated by gravity model; (3) The gravity model has a good explanatory ability in studying bilateral trade issues, and its data acquisition is simpler than other commonly used models.

Data source and sample description

This research selected the data of Kazakhstan and its 21 major grain trading partners from 2008 to 2018, with the list of countries in figure 3. The data was taken from the FAOSTAT, UN Comtrade Database, World Bank database, the Heritage Foundation, World Bank’s Global Governance indicators, CEPII database. The classification of grain and its products is selected according to categories 10, 11, and 19 in the International Convention for Harmonized Commodity Description and Coding System of 1992 edition (HS92) in table 2.

Figure 3
Trends of Kazakhstan’s grain export potential from 2008 to 2018.

Table 2
The specific classification of grain and its products.

# RESULTS:

Kazakhstan’s grain international competitiveness

From the overall trend, the international competitiveness of Kazakhstan’s grain had gradually increased from general, and finally stabilized within the range of relatively strong (10%< IMS <20%) (Figure 4). Wheat and barley were the most competitive grain crops, with their average annual IMS of 2.3% and 1.6% from 2008 to 2018 respectively; rye, corn, and rice had weak competitiveness in the international market, with their IMS below 0.2% throughout. The IMS of wheat, buckwheat, and barley fluctuated sharply. Among the common grain crops in Kazakhstan, the IMS of wheat, buckwheat, and barley fluctuated sharply during 2008-2018, while the IMS of the others maintained stable.

Figure 4
International market share of Kazakhstan’s main grain between 2008 and 2018 (%).

According to the TC of grain export (Figure 5), among the 8-grain crops, wheat, barley, oats, and millet had a strong international competitive advantage; while the TC of the other grain crops fluctuated constantly, and their trade competitiveness was unstable. The TC of wheat was above 0.96 all through; the barley’s TC decreased first and then increased, but it was always higher than 0; the oats and millet’s TC remained around 0.92. These indicated that the domestic productivity of wheat, barley, oats, and millet in Kazakhstan has always been higher than the international level, maintaining net export status, with a strong competitive advantage in international trade; however, the production efficiency of the other grain crops was unstable.

Figure 5
Dominant comparative advantage index of Kazakhstan’s main Grain between 2008 and 2018.

According to the TC of grain export (Figure 6), among the 8-grain crops, wheat, barley, oats, and millet had a strong international competitive advantage; while the TC of the other grain crops fluctuated constantly, and their trade competitiveness was unstable. The TC of wheat was above 0.96 all through; the barley’s TC decreased first and then increased, but it was always higher than 0; the oats and millet’s TC remained around 0.92. These indicated that the domestic productivity of wheat, barley, oats, and millet in Kazakhstan has always been higher than the international level, maintaining net export status, with a strong competitive advantage in international trade; however, the production efficiency of the other grain crops was unstable.

Figure 6
Trade competition index of Kazakhstan’s main Grain between 2008 and 2018.

Model regression results and analysis

In this study, Stata 16.0 software was used to perform statistical tests on the regression equations through mixed regression models, fixed-effects models, or random-effects models. In the test results of the mixed regression model and the fixed effects model, the P-value of the F-test was 0.000, indicating that the fixed effects model was significantly superior to the mixed regression model, but the F test was not effective without using robust standard error. Hence the test results needed to be further investigated by the LSDV method, where most individual virtual variables were significant (P<0.1). Therefore, the null hypothesis that “all individual dummy variables are 0” was rejected, and it was considered that the individual effects exist and mixed regression should not be used. Then, the Hausmann test was used to determine whether to establish a random-effects model or a fixed-effects model. The result showed that the p-value was 0.138, so the null hypothesis and the random effect model were accepted. Considering that the research contained time-invariant variables, and the possibility of heteroscedasticity, serial correlation, and cross-sectional correlation, the random effects were selected for estimation, which was suitable for analyzing the panel regression with time-invariant variable (Table 3).

Table 3
Comparison of panel data regression estimation results based on the gravity model.

It could be seen from the regression results of the random-effects model, that the significance level of F value was 0.000, and R2 was equal to 0.595, indicating that the model had good fitting and reliability. Among the selected variables, the distance of GDP per capita and whether to join the Eurasian Economic Union passed the significance test at 1% statistical level; the geographic distance in bilateral trade passed the statistical significance test at 5% level; the actual economic level and economic distance passed at 10%; the other factors even did not pass at 10%.

Export potential measurement

First, the theoretical value of Kazakhstan’s total grain exports to various countries was calculated based on the regression parameters of the random effect model. Then calculate the ratio of actual value to theoretical value to obtain the potential of the bilateral grain trade.

${\mathit{TP}}_{t}={T}_{t}/{T}_{t}^{*}$

TPt is the trade potential index of Kazakhstan during t period, Tt and Tt* is the actual trade volume and the theoretical trade volume respectively. Referring to Liu and Jiang (.QINGFENG & SHUZHU, 2002QINGFENG, L, SHUZHU, J. China’s bilateral trade arrangements from the model of trade gravity. Zhejiang Social Sciences, n.6, p.16-19, 2002. Available from: <Available from: https://doi.org/10.14167/j.zjss.2002.06.004 >. Accessed: Jan. 15, 2021. doi: 10.14167/j.zjss.2002.06.004.
https://doi.org/10.14167/j.zjss.2002.06....
) classification standard for trade potential, the trade potential is divided into three types: “potential for reinvention” (TPt≥1.2); “developable potential” (0.8<TPt<1.2); “huge potential” (TPt≤0.8).

Figure 7
The spatial distribution pattern of Kazakhstan’s grain export potential in 2018.

Model test

Unit root test

Panel data has the dual characteristics of time series data and section data. It is necessary to test the stability of the panel data before building the panel data regression model in order to avoid a false regression and ensure the validity of the estimated result (DEJONG et al., 1992DEJONG, D N, et al. The power problems of unit root test in time series with autoregressive errors. Journal of Econometrics, v.53, n.1-3,p. 323-343, 1992. Available from: <Available from: https://doi.or g/10.1016/0304-4076(92)90090-E >. Accessed: Jan. 15, 2021. doi: 10.1016/0304-4076(92)900 90-E.
https://doi.or g/10.1016/0304-4076(92)90...
). According to the results of Breintung, IPS, and ADF tests, all the variables reject the null hypothesis that “statistical variables have unit roots” at the level of 10%. Therefore, these sequences have no unit roots, they are first-order single integer sequences (Table 4).

Table 4
Unit root test.

Co-integration test

Since all the variables are first-order integral sequence, there may be some long-term stable co-integration relationships among these indicators. Panel co-integration test methods are mainly divided into the following two categories: one is the panel co-integration test method based on the unit root test of co-integration regression test residuals, such as the Pedroni and Kao test; and other is the Johansen trace test method. In this study, the Kao test is used to determine whether there are co-integration relationships among variables. Kao and Chiang proposed a method to test panel co-integration based on DF and ADF tests, using the residuals of static panel regression to construct statistics (KAO, 2000KAO, M. C. On the Estimation and inference of a Cointegrated Regression in Panel Data[J]. Chihwa Kao, v.15, n.1, p.109--141, 2000. Available from:<Available from:https://doi.org/10.1016/ S0731-90 53(00)15007-8 >. Accessed: Jan. 15, 2021. doi:10.1016/S0731-9053(00)15007-8.
https://doi.org/10.1016/ S0731-90 53(00)...
). If the null hypothesis that “there are no co-integration relationships among variable sequences” is rejected at the significance level of 1%, it indicates that there are long-term stable equilibrium relationships among variables (Table 5).

Table 5
Panel Cointegration Test.

Robustness test

In order to test the endogenous problems caused by the correlation between omission variables and explanatory variables in the process of model design test, this paper introduced the one-period lagging GDP as an instrumental variable to carry out 2SLS robustness analysis on the model (KUTTNER, 2011KUTTNER, K. Monetary policy and Asset Price Volatility: Should We Refill the Bernanke-Gertler Prescription?. Department of Economics Working Papers, 2011.Available from: <Available from: https://web.williams.edu/Economics/wp/KuttnerMonetaryPolicyAndAssetPriceVolatility.pdf >. Accessed: Jan. 15, 2021.
https://web.williams.edu/Economics/wp/Ku...
). This instrumental variable was highly correlated with the actual economic level of the current period, but cannot affect the economic level of the previous period. In the first stage, the dependent variable was the actual economic level, and the independent variables were instrumental variables and other control variables. The p-value of the regression result is 0.000, indicating that the instrumental variables were positively correlated with the actual economic level significantly. The second stage regression results showed that the regression parameter values changed after controlling the endogenous problems. For example, the regression coefficient of the WTO accession changed, and the regression parameters of the gap in GDP per capita and the real economic level decreased in significance among trading parties. The gap in GDP per capita passed the 5% test level, and the actual economic level is no longer significant on both sides of the international trade with Kazakhstan. The relationship between the other explanatory variables and the explained variables was consistent with the results of the significance test, indicating that the argument obtained by the random-effects model remained valid, under the 2SLS regression for controlling endogenous problems (Table 6).

Table 6
Robustness test.

# DISCUSSION:

This paper analyzed the spatial and temporal pattern of Kazakhstan’s grain export trade and the level of international competitiveness, and uses the extended gravity model to measure its grain export potential. The main conclusions included the following:

# ACKNOWLEDGEMENTS

This research was funded by the National Natural Science Foundation of China, grant number 71673222 and 72064009, Humanities and Social Science Fund of Ministry of Education of China, grant number 15XJA790005, China Scholarship Council Funded Project, grant number 202106300001.

• CR-2021-0199

# Publication Dates

• Publication in this collection
24 Sept 2021
• Date of issue
2022