MEASURING TRADE POTENTIAL FOR BOTSWANA S EXPORTS IN THE SOUTHERN AFRICAN DEVELOPMENT COMMUNITY (SADC)

Measuring Asian-African Trade Journal Potential of Economics for Botswana s and Econometrics, Exports in the Vol. Southern 14, No. African... 2, 2014: 227-235 227 MEASURING TRADE POTENTIAL FOR BOTSWANA S EXPORTS IN THE SOUTHERN AFRICAN DEVELOPMENT COMMUNITY (SADC) Imogen Mogotsi *, Narain Sinha ** and Gotlhe Medupe *** ABSTRACT This study investigates the presence of untapped trade potential for Botswana in the region, using the results of the gravity model estimation. The results depict that there is still untapped potential for Botswana s exports in most SADC countries except for the DRC and Zambia, where the potential has been exceeded. A conclusion is made therefore, that Botswana should endeavour to diversify its export markets beyond SACU INTRODUCTION Botswana is a small landlocked economy with a narrow product and export base; it has a high level of commodity and geographic concentration of its exports. For example, Botswana s exports are highly concentrated on primary products, especially diamonds and copper/nickel (diamonds contributing 82.6 per cent to total exports in 2013 (Bank of Botswana, 2013); copper/nickel 7.22 per cent) and to a lesser extent meat and meat products (MTI, 2009; WTO, 2011). Geographic concentration is evidenced by the fact that the direction of exports is mainly towards Europe, especially the United Kingdom (64.5 per cent), the Southern African Customs Union (SACU) (20.1 per cent); Zimbabwe (4.5 per cent); USA (1.0 per cent); rest of Europe (6 per cent); and other African countries (1.5 per cent),while the bulk of imports (76.1 per cent) were from South Africa (MTI, 2009; WTO, 2011). To meet its target of export diversification it need not only concentrate on manufacturing of new commodities but also identify different markets for its produce.one such market is the Southern African Development Community(SADC). Given that Botswana has signed a free trade agreement with the region, it now has access to 80 per cent of a market of around 277 million consumers. 1 The problem is that although Botswana s levels of trade internationally are relatively high, its levels of regional trade have remained low. From the statistics above it shows that of the regional blocs it mainly trades with SACU countries and Zimbabwe, and within SACU it is South Africa which accounts for the bulk of the trade. As for the rest of SADC the trade is relatively minimal.it is therefore important to establish the export destinations with potential for trade within SADC. Considering only three sectors, * Associate Professor, Department of Economics, University of Botswana, Private Bag UB 705, Gaborone, Botswana, E-mail: mogotsii@mopipi.ub.bw ** Associate Professor, Department of Economics, University of Botswana *** Lecturer, Limkokwing University, Botswana

228 Imogen Mogotsi, Narain Sinha and Gotlhe Medupe Makochekanwa (2010) established that there is trade potential to be harnessed within the region for Botswana. As stated earlier, Makochekanwa (2010) considered only three sectors for Botswana, This paper mainly aims to fill the gap and to establish whether there is trade potential for Botswana s total exports within the regionso that it is possible to know if full potential has been wholly tappedfor the enhancement of trade. In other words, projecting trade potentials will be a good indication of whether there are still prospects for the economy in the path to economic growth and development through an increase in intra-regional trade volumes (Ozdeser and Ertac, 2010)The purpose of this paper is to investigate markets with untapped trade potential for Botswana s exports in the SADC region. This involves measuring the trade potential, using the gravity model approach. Moreover, export policy is a valuable instrument for destination or country-specific export promotions; therefore the study will help in coming up with recommendations for effective and efficient trade policies. Few studies have been carried out for Botswana, and only one looked at potential trade, and even then, for a few selected sectors.this paper, therefore, examines the bilateral trade potential of Botswana and individual SADC member states.. Whether trade potential within Southern Africa exists for Botswana, is therefore an empirical question. 1. TRADE POLICY FOR BOTSWANA Botswana has implemented a number of agreements with various trade blocs to attain its mandate of economic development and growth. The most important is its membership of the oldest customs union, the Southern African Customs Union (SACU). In the region, the other important trade agreement is with Zimbabwe. The agreement was initially signed in 1956 and later amended in 1987 (MTI, 2009). This led to an increase in its level of trade with the country. The agreement provides for duty and quota free trade for goods produced or grown in the two countries. Botswana being a member of the Southern African Development (SADC), this has facilitated its access to that market. Recently SADC has focused on trade liberalization as a means to stimulate trade between member states; as a result in 2000 the SADC Protocol on Trade came into effect. However, only 11 countries were signatories to the trade protocol when it was launched in 1996, while Angola, 2 Congo, Madagascar, 3 and Seychelles wereyet to sign. Botswana signed the Protocol on Free Trade;this has enhanced trade liberalization as Botswana now potentially has access to a market with an estimated 277 million consumers,which gives Botswana a potential to diversify its exports. For example, it could move away from the status quo of trading mainly with South Africa and Zimbabwe in SADC as shown in Figure 1 below. As shown in the figure, there is heavy geographic concentration as more than 90 per cent of its goods are destined to 5 region/countries. Since the inception of the SADC trade protocol, SADC has experienced an increase in its intra-regional trade. The member countries are increasing their trade with each other as evidenced by the increase in the share of exports from SADC countries sold within the bloc. Particularly, Zimbabwe, Namibia, Botswana, South Africa, Tanzania and Zambia have seen their share of exports increase within SADC (Negasi, 2009). Comparatively though, the general

Measuring Trade Potential for Botswana s Exports in the Southern African... 229 Figure 1: Botswana s top 5 export destinations for 2010 Source: WTO trade profiles increase in exports within SADC is negligible given that most of the exports are destined to markets outside Africa.As shown by Figure 1, the main export destinations for Botswana are outside the region, mostly the European Union.. This implies that Botswana is missing out on the opportunities to take advantage of the integration initiatives that SADC has put in place. Hence the objective of this paper, to examine the extent to which Botswana can still export to the SADC countries. 2. LITERATURE REVIEW Trade potential for trading partners is estimated using the Gravity model. The potential volume of trade between countries is defined as the volume of trade that would prevail if trade were explained by the same factors determining trade between the countries as determined by the gravity model (Nilsson, 2000). Estimating the trade potential between bilateral trade partners is crucial as it gives direction as to which countries/regions the trade strategies should be directed. This ensures that trade potential is fully exploited to enhance economic growth. Trade potential can be estimated using two approaches, the out-of-sample and in-sampleapproach. For the out-of-sample approach, parameters of highly integrated countries are estimated by a gravity model, and then the same coefficients are applied to project natural trade relations between these benchmark countries and countries starting to integrate. The difference between the actual and the estimated trade flows are interpreted as the trade potential. For the in-sampleapproach the transformation of countries starting to integrate are included in the regression of the gravity equation and then the residual of the estimated equation is interpreted as the difference between the actual and the potential flows (Ozdeser and Ertac, 2010).

230 Imogen Mogotsi, Narain Sinha and Gotlhe Medupe Many studies have been carried out to investigate the trade potentials in different trading blocs. One such study, by Cassim (2000) found that non SACU members within the SADC region have unrealized trade potential, higher than that of SACU member states. Moreover, they found out that there is increasing scope for non-sacu member states to increase their exports to South Africa given the higher trade potentials. Similar results were observed bykisu (2010) while evaluating the trade potential in SADC. However, a study by Eita and Jordan (2007) on South Africa found that its actual exports were all above potential exports, that is, South Africa had exceeded most of its trade potential with the other SADC countries. On the other hand, for the other SADC countries the potential still existed, but they found it to be relatively smaller, especially because the countries seemed to be trading in homogenous goods. Other studies have looked at trade potential of Southern African countries and the rest of the world. Considering three sectors namely diamond, textiles and meat and meat products sectors,makochekanwa (2010) established that Botswana had untapped trade potential in these sectors From their results, Israel and Switzerland were the main destinations for which there was untapped trade potential for Botswana in the diamond sector. Trade potential for the textile industry was mainly untapped in Canada, Denmark, Ghana and Mozambique. Last but not least, Italy, Mauritius, Namibia and Norway were the main destinations with unrealized trade potential in the meat and meat products industry. 3. MODEL SPECIFICATION The gravity model estimations are often used to deduce the bilateral trade potential for a specific group of countries. The theoretical estimation of the gravity model, from which simulation of the trade potential is undertaken, is presented by equation 1 below (see Mogotsi et al., 2014): 8 1 2 3 4 5 6 7 9 10 ij ij i j i j ij ij j i ij X Y Y N N D EX CONS SADC OPN OPN u (1) Where: X ij,t is the total value of exports from country i (exporting country) to country j (importing country) at time t; Y i,t and Y j,t are the real GDP for the exporting and the importing country at time t respectively;the population of both the exporting and importing countries are represented by N i,t and N j,t respectively; OPN j,t and OPN i,t are the respective trade openness variables for the importing and exporting countries at time t; EX ij,t is the real exchange rate, expressed as the US dollar per Pula at time t; CONS ij,t is the consulate dummy representing whether the exporting country (Botswana) has an embassy or consulate in the importing country at time t; SADC ij,t represents whether a country has signed the FTA or not at time t. Finally, D ij is the distance between the exporting and importing countries economic trade centres? Estimating Trade Potential Although the gravity model is not dynamic, one is able to derive dynamic-like results (Kisu, 2010). The in-sample approach to trade potential from country i to group k is defined as

Measuring Trade Potential for Botswana s Exports in the Southern African... 231 TP ik j s k j sk X ij X (3) ij Where: S k is the set of countries belonging to group k (in this case k is SADC) as importers. is an estimate of the trade volumes calculated from the regression (Equation 1) and Xij are the actual trade levels used in estimating the gravity model. If the ratio TP ik is higher than one, it means the country has not exported to its full potential, and if the ratio is smaller than one it implies that the country has exceeded its expected level of exports (Proenca et al. 2008). The method consists of estimating the bilateral trade equation, based on the gravity model in the first phase and then the equation is used for simulation in the second phase. As stated above, to estimate bilateral trade potential for a specific group of countriesconsists of estimating the bilateral trade equation, based on the gravity model in the first phase; then it employs the equation for simulation in the second phase (Makochekwana, 2010). Therefore, this section depends on the model estimates from the bilateral trade estimation based on the gravity model. Similarly, the absolute difference between the potential and actual level of trade; that is, the value of (P-A) can also been used to classify countries with potential for expansion of trade. A positive value indicates future possibilities of trade expansion, while a negative value indicates that the country s exports have already exceeded their trade potential. 5. TRADE POTENTIALS In addition to its usefulness in indicating the determinants of bilateral trade, the gravity model is a useful tool for the measurement of trade potential. Thus the estimates obtained (Mogotsi et al., 2014) from running Equation 1, given in Table 1 below are used to calculate the trade potentials. It is interesting to observe that GDP of the countries, openness and population of the importing country have no role to play in the bilateral trade in the SADC region. Rather trade policies influencing openness of the exporting country is significantly influencing the bilateral trade. Trade potentials are necessary to point policy makers in which direction to focus their country specific trade strategies. This is because, evaluating the potentials help establish which: SADC trading partners it still has untapped potential with; Trading partners Botswana has reached its trading potentials with; And which trading partners it has gone beyond the trade potential. j sk j sk X ij

232 Imogen Mogotsi, Narain Sinha and Gotlhe Medupe Table 1 Determinants of Bilateral Trade for Botswana and SADC Countries Categories of Variables CONSTANT Model 1 Model II DEPENDENT VARIABLE: DEPENDENT TOTAL EXPORTS VARIABLE: INDIVUAL EFFECTS FROM MODEL 1-38.2318 (0.0000)*** Policy variable/ REAL EXCHANGE 2.324559 trade resistance RATE (0.0947)* Trade enhancing REAL GDP -5.000963 supply factors (exporting country) (0.1713) OPENNESS TO TRADE (exporting 13.3827 country) (0.0152)** Trade enhancing REAL GDP 0.476278 demand factors (Importing countries) (0.7437) POPULATION 2.884872 (importing countries) (0.6246) POPULATION 21.82571 (exporting country) (0.0582)** OPENNESS TO TRADE (importing 0.914111 country) (0.2591) Trade resistance DISTANCE -0.349185 (0.0648)* Trade preference CONSULATE 3.235523 factors (0.0000)*** SADC FTA 2.956213 (0.0000)*** Model diagnostics R-squared 0.793242 0.899480 Adjusted R-squared 0.767558 0.894220 F-statistic 31.16940 Prob(F-statistic) 0.000000 Durbin Watson stat 2.087096 The figures in parenthesis are p-values: ***significant at 1%, **significant at 5%, *significant at 10% The results actually point to markets that can be tapped into so as to diversify exports and export destinations. Therefore, this analysis is very crucial given that Botswana has been trying to diversify its export and export destinations. Thereby, it actually gives an indication as to where policy makers should focus their attention within the SADC region when it comes to coming up with initiatives that could increase the level of exports. To get the trade potential,equation 1 as it appears aboveis simulated to get the within export potential: As stated above, the trade potentials are evaluated as the ratio of the predicted trade potential, (as predicted by the model: P) and actual trade values (A); that is, trade potential/ actual trade (P/A). If P/A is greater than 1 that means there is potential to expand trade with the respective country. That is, if the ratio exceeds 1, it points to a market that the country can export further to in the future. Otherwise if it is less than 1, the country has exceeded its trade

Measuring Trade Potential for Botswana s Exports in the Southern African... 233 potential. If the latter scenario presides then the country should refocus its strategy on exporting more to countries where there is potential to be tapped into. Similarly, the trade potential can be shown by the difference between the predicted and the actual trade values (P-A). If the difference is positive it shows that there is trade potential to be tapped into in the particular country or region. A negative figure on the other hand means that the country has exceeded the trade potential and therefore, there is little room for expanding the exports in that particular market. Trade Potential of Botswana in SADC Table 2 presents the estimations of trade potential for Botswana s exports within the SADC region 4, for the period 1998-2010. Table 2 SADC Countries with Untapped Trade Potential YEAR 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Angola 0.65 18.78 39.12 0.39 2.37 0.55 1.35 0.08 0.05 0.6 0.3 0.8 4.8 Lesotho 1.39 1.7 0.46 0.84 0.26 0.58 1.67 2.81 1.26 2.87 0.49 0.73 1.22 Malawi 1.22 0.3 0.47 0.51 0.43 0.75 0.83 3.21 1.43 3.68 1.53 0.95 1.74 Mauritius 3.73 0.95 0.74 0.55 0.57 1.95 0.95 0.12 0.27 1.16 0.33 3.3 15.57 Mozambique 0.71 0.15 0.46 1 18.07 0.35 0.69 6.04 0.8 3.97 0.83 0.21 1.35 Namibia 0.49 1.03 0.78 2.01 7.46 0.68 1.2 1.17 0.48 1.3 0.28 0.59 1.71 South Africa 0.81 1.18 1.14 0.68 1.3 0.58 0.98 1.33 0.9 1.33 0.75 0.91 1.69 Tanzania 1.4 0.64 0.26 0.12 3.23 0.66 0.94 1.65 0.89 4.32 1.04 1.94 1.4 Swaziland 0.67 0.46 0.78 0.8 36.25 0.43 1.57 1.67 0.72 0.6 0.75 0.35 1.1 Zimbabwe 2.86 2.85 1.03 0.88 1.33 0.71 0.93 0.87 0.34 0.46 0.85 0.8 1.65 As indicated by Table 2, the analysis on Botswana s trade potential shows that for the latest year in the analysis (2010), Botswana had untapped potential with most of the SADC countries: Angola, Lesotho, Malawi, Mauritius, Mozambique, Namibia, South Africa, Zimbabwe, Tanzania, and Swaziland; although with Swaziland the potential seemed to be low, at 1.1. Moreover, when we look at the trend over the period 1998-2010, for most of these countries, the trade potential seemed to grow over the years. Botswana must therefore, export more to these countries so as to exhaust the potential and thereby reap the benefits of its membership in the large regional trade bloc of SADC, and meet its mandate of export diversification. Particularly, from the results it can be shown that Botswana should focus more on exporting to Mauritius, also to Angola as these are the countries with the highest levels of potential respectively. Historically, Botswana has been known to trade more with the SACU countries within SADC, and from the results it shows that the trend could still continue as it still has some potential for intra-sacu trade.of the SACU countries it still has the most potential with Namibia and South Africa. Although Botswana already trades the most with South Africa among all the countries investigated, the untapped potential depicts that South Africa is still relevant for Botswana s trade. In other words South Africa has been and still is a crucial player in Botswana s trade industry, and it looks like it will still play an important role in the future. Interestingly,

234 Imogen Mogotsi, Narain Sinha and Gotlhe Medupe Botswana has the least potential with Lesotho and Swaziland, which are also members of SACU; and the potential is almost exhausted, at 1.2 and 1.1 respectively. Table 3 Countries without Trade Potentials Year 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 DRC 0.23 0.61 2.78 0.59 0.73 1.31 7.49 3.22 0.59 2.07 0.61 0.43 0.60 Zambia 1.71 1.95 0.72 0.59 1.08 2.29 3.00 1.40 0.63 2.95 0.34 0.22 0.48 Conversely, Botswana seems to have exceeded its trade potential with Zambia and DRC. This means there is no longer room to increase its level of exports to these 2 countries which are also members of SADC;from the results in Table 3 above it shows that the ratio is less than 1. This means that the level of potential exports is less than the level of the actual exports. The implication is that Botswana should reduce its efforts to try and promote its level of exports to the Zambian and the DRC markets and concentrate on such markets as the SACU region as well as Mauritius, Angola and others where there is actually some potential to be tapped into. A comparison of these results with those of other studies that also looked into the trade potential of the SADC countries,one such study is a cross sectional study in 1996 by Chauvin and Gaulier (2002). Based on the results Chauvin and Gaulier established that Angola, DRC, Lesotho, Mauritius, Mozambique, Namibia, Seychelles, Swaziland and Tanzania all still had potential to import goods from Botswana. The results with regards to untapped potential are similar to the ones obtained in the current study, a decade and a half later, except in the case of DRC and Zimbabwe. For the current study DRC has exhausted the potential, while in their study they found that it still had potential to be tapped into. In the case of Zimbabwe, they found out that Botswana had exceeded its trade potential with Zimbabwe, while we find that Botswana still has potential with Zimbabwe which is relatively higher than for four of the other SADC countries. Chauvin and Gaulier(2002) established that South Africa and Malawi also had exceeded the potential levels of trade with Botswana, which is contrary to the findings of the present study. Given that the study was carried out in 1996, one may postulate that the new potential can be attributed to the signing of the SADC FTA. The results for the potential shows that for those countries which still have untapped potential, some of them the potential increased after the signing of the FTA although with subsequent fluctuations. Therefore, the role of the FTA in increasing trade potential between member states needs to be looked into thoroughly as Botswana endeavours to proceed with the other SADC countries in further liberalizing trade. 6. SUMMARY, CONCLUSIONS AND POLICY RECOMMENDATIONS With the objective whether there is any trade potential to be tapped into in the SADC region by Botswana,the trade potential analysis was carried out by simulating potential trade from a gravity model estimated, and taking the ratio of the potential trade to the actual trade levels.

Measuring Trade Potential for Botswana s Exports in the Southern African... 235 The main conclusion is that Botswana still has untapped potential with a number of SADC countries, all except with the DRC and Zambia. This means that Botswana could therefore, concentrate on diversifying its exports beyond the SACU countries. Trade liberalization, in the form of SADC countries being signatories of the SADC Free Trade Area seems to play a significant role in creating potential for countries to trade with one another. Therefore, for the SADC countries to trade more with one another, they need to move up the ladder in terms of economic integration, by becoming signatories to the SADC FTA. Notes 1. Botswana has access to 80% of the SADC market as not all member states have signed the FTA. Statistics accessed from SADC website at: http://www.sadc.int/investment/about-sadc/socio-economic-indicators/ population/ 2. Angola ratified the trade protocol, although it has yet to put forward its tariff offer. 3. Madagascar joined SADC in 2005 bringing the membership to 15, but its tariff offer was approved at the SADC Heads of State Summit in September 2006. 4. Two countries namely Madagascar and Seychelles have been dropped from our estimation because the former signed the protocol in 2006 and the latter is yet to sign. Bank of Botswana Annual Report (2013). Selected References Cassim, R.(2000), The Determinants of Intra-Industry Trade in Southern Africa with Special Reference to South Africa and the Rest of the Region. Paper Presented for the Second Research-in-Progress Workshop. Development Policy Unit Division. Chauvin,,S., and Gaulier, G. (2002), Prospects for Increasing Trade Among SADC Countries. Centre d EtudesProspectivesetd Informations Internationals (CEPII). Eita, J.H., and Jordan, A.C. (2007), South Africa Exports of Metals and Articles of Base Metal: A Gravity Model Approach. Journal of Studies in Economics and Econometrics, 31(3), pp. 81-95. Kisu, S.(2010), An Empirical Evaluation of Trade Potential in Southern African Development Community.Munich Personal RePEc Archive. MPRA Paper No. 15894 posted 07. June 2010 / 14:04 from: Online at http://mpra.ub.uni-muenchen.de/15894/. Makochekanwa, A. (2010), An Econometric Analysis of Botswana Sectoral Export Trade Flows. University of Pretoria (Faculty of Economics and Management Science). Medupe, G. (2012), Determinants of Botswana Bilateral Trade with Southern African Development Community (SADC) Countries: A Gravity Model Approach. University of Botswana. Mogotsi I., Sinha N and Medupe G. (2014), Bilateral Trade of Botswana with the Southern African Development Community (SADC) Countries: An Application of the Gravity Model, in International Journal of Economic Issues, Vol. 7, No. 1, pp 1-19. Ogunkola, E.O. (1998), An Empirical Evaluation of Trade Potential in the Economic Community of West African States. AERC: Research Paper 84. Ozdeser, H., and Ertac, D. (2010), Turkey Trade Potential with EURO Zone Countries: A Gravity Study. European Journal of Scientific Research, 43 (1), pp. 15-23. Proenca, I., Fontura, M., P., and Martinez-Galala, E. (2008), Trade in Enlarged European Union, a New Approach on Trade Potential. Portuguese Economic Journal, 7, pp. 205-224.