قیمت جهانی نفت و قیمت کالاهای کشاورزی: شواهدی از بازار نوظهور

Oil prices are thought to have direct effect on agricultural prices followed by an indirect effect through the exchange rate. This paper examines the short- and long-run interdependence between world oil prices, lira–dollar exchange rate, and individual agricultural commodity prices (wheat, maize, cotton, soybeans, and sunflower) in Turkey. To this end, the Toda–Yamamoto causality approach and generalized impulse-response analysis for identification of the long- and short-run interrelationships are applied to the monthly data spanning from January 1994 to March 2010. The impulse–response analysis suggests the Turkish agricultural prices do not significantly react to oil price and exchange rate shocks in the short-run. The long-run causality analysis reveals that the changes in oil prices and appreciation/depreciation of the Turkish lira are not transmitted to agricultural commodity prices in Turkey. Hence, our results support neutrality of agricultural commodity markets in Turkey to both direct and indirect effects of oil price changes.

The upward trend in agricultural commodity prices starting in 2003 and continuing till 2008 has raised concern for countries that depend heavily on food imports. The agricultural prices considerably increased from the beginning of 2006 to mid-2008. Although during the summer of 2008 commodity prices started to fall, by the end of 2008 they reached their early 2007 levels. Average prices of agricultural commodities in the world are still above their historical levels (Cooke and Robles, 2009 and OECD (Organization for Economic Co-Operation and Development), 2009: 22). The agricultural commodity prices in Turkey follow closely these international dynamics (see Fig. 2 in Section 3). In order to apply appropriate policy options and to examine investment opportunities, both policy makers and global investors question the factors influencing the agricultural commodity markets. Piesse and Thirtle (2009) argue that since oil prices are expected to pick up following the current recession in the world economy, we need to improve the state of the world agriculture. They point out that agriculture has the potential to provide sufficient food and energy in the future if correct actions are taken now. Their suggestions of course are based on the presumption that world energy prices are the main factors causing a rise in world food prices. In the literature, several studies focus on a number of different factors (see Cooke and Robles (2009) for a review) to understand the drivers of surging agricultural commodity prices. Abbott et al., 2008 and Abbott et al., 2009 reduce the number of these factors to three key determinants: excess demand, the value of US dollar, and the energy-agriculture linkage. However, the rise in energy prices is considered to play the key role in explaining the recent dynamics of the agricultural commodity prices in the world. Energy and agricultural markets have become closely linked as production of biofuels surged since 2006. Ethanol and biodiesel are substitutes for gasoline and diesel, thereby the recent surge in agricultural commodity prices are attributed to increasing usage of crops in production of biofuels. Indeed, the link between energy and agriculture is not a new concern. An early study by Chenery (1975) notes the disruption in world trade due to rising oil and food prices. According to him substantial adjustments need to be made after the price rises due to changing productivities. He emphasizes the strain put on many developing countries as a result of the disequilibria experienced. The story does not seem to have changed too much since the 1970s. The joint upward trend in oil and food prices raises concern for many countries. Historically, agriculture has been an energy-intensive sector and therefore one can draw a direct linkage from oil prices to agricultural commodity prices. As discussed in Hanson et al. (1993) an increase in oil prices is followed by an increase in input costs which in turn causes agricultural prices to rise. Another link between oil prices to agricultural commodity prices is through the exchange rates. According to Harri et al. (2009), there is an indirect effect of oil prices on agricultural prices through the exchange rates. Oil trade is conducted mainly in US dollars; hence, changes in oil prices have direct impact on local currencies of all countries. The appreciation/depreciation of local currency, in return, influences the agricultural commodity imports/exports and the local prices of these commodities. Therefore, there are two links from oil prices to agricultural commodity prices: a direct link from oil prices to commodity prices and an indirect link through exchange rates. As will be shown in the literature review that follows, the interactions between oil prices, exchange rates, and agricultural commodity prices are not clear-cut, and the relationship is largely empirical. To the extent of our knowledge the direct and indirect effects of oil prices are not well studied for individual agricultural commodity prices in emerging markets or for developing countries. In this paper, we empirically assess the direct and indirect effects of world oil prices on agricultural commodity prices in an emerging market, Turkey, for the January 1994–March 2010 period. Following the suggestion of Baffes (2007), we examine how individual agricultural commodity prices, rather than an aggregate index for the agricultural sector prices, are affected by changes in oil price and the exchange rate. To that respect we employ a procedure developed by Toda and Yamamoto (1995) to test for long run Granger causality and generalized impulse responses by Koop et al., 1996 and Pesaran and Shin, 1998 to examine how each variable in the system responds to a generalized one standard deviation shock. The Toda and Yamamoto (1995) approach is superior to traditional approaches, because it avoids pre-test biases in the cointegration tests and cointegrating equation estimations. It is a rather flexible approach such that it can be applied to series with arbitrary degrees of integration. The generalized impulse responses are also superior to traditional impulse responses. The generalized approach results are not sensitive to the order by which variables enter the VAR system. Agricultural commodity prices are of great economic importance in Turkey, since the agricultural sector accounts for 8% of total production, 27% of total employment, and 9% of total exports as of 2008 (OECD, 2009: 200). Turkey also depends heavily on imported oil and the Central Bank of Republic of Turkey follows a flexible exchange rate regime. Although the conditions for both direct and indirect links between world oil prices, exchange rates, and agricultural commodity prices exist in Turkey, we find that the Turkish agricultural commodity prices are neutral to both effects. Furthermore, even though the correlations between agricultural prices and oil prices have increased after 2006, due to the recent upward trend in both price series, our re-examination for the sub-sample from January 2006 to March 2010 substantiates that neutrality. Although a lot has been learned about world commodity prices (Cashin et al., 2002), information on how local agricultural commodity prices respond to oil and exchange rate shocks is limited. Better insight on the dynamics of local agricultural prices will help policy makers establish sound monetary, energy, and agricultural policies. Producers will benefit from this information in crop selection and in determining the extent to which they are exposed to the exchange rate risk as well as to changes in world energy markets. Finally, investors will be able to identify portfolio diversification options in various local agricultural commodity markets seen as alternative investment areas. The rest of the paper is organized as follows. The next section is devoted to the literature on the oil prices-agricultural commodity prices nexus, data is described in Section 3 followed by the methodology and empirical findings in Section 4, and finally the concluding remarks are presented in Section 5.

This paper investigates the direct and indirect effects of world oil prices on Turkish agricultural commodity prices (wheat, maize, cotton, soybeans, and sunflower). Employing a procedure that does not necessitate a pre-test for cointegration, we find that oil prices affect individual agricultural prices in Turkey neither directly nor through the exchange rates. The generalized impulse response analyses, that are robust to the ordering of variables in the system, support the neutrality of agricultural commodity prices to world oil prices in the short run. When we follow the suggestion of Campiche et al. (2007) and just focus on the time period where correlations between oil and agricultural prices picked up, we still observe the neutrality of Turkish agricultural commodity prices. Our results are supportive of the findings of Yu et al. (2006) for vegetable oil and crude oil prices and Mutuc et al. (2010) for cotton and oil prices in the US. Our results are also in line with the conclusions of Zhang and Reed (2008) on Chinese agricultural prices; Zhang et al. (2010) on various world agricultural commodity prices. However, our findings contradict with those of Hanson et al., 1993, Kwon and Koo, 2009, Harri et al., 2009, Gohin and Chantret, 2010 and Baek and Koo, 2010 that focus on various agricultural prices in the US. The neutrality of all agricultural prices in Turkey to world oil price changes may be due to the relatively low energy intense production processes. Regardless of the reason, neutrality in agricultural prices has important implications for policy makers and investors. First of all, energy policies in Turkey are not expected to have significant impacts on the agricultural sector prices. This also implies that policies to stabilize agricultural prices do not need to account for developments in the world oil market. Second, in order to identify the causes of inflation in agricultural prices policy makers must focus on supply and demand dynamics within the agricultural sector, since agricultural prices are not responding to shocks in external factors such as world oil prices and the value of Turkish lira. Third, the insignificance of the indirect effect of oil price changes implies that changes in imports/exports of agricultural items due to appreciation/depreciation of the Turkish lira have no effect on agricultural prices. According to FAO's trade and production data, the ratio of export to total production as an indicator of the degree of trade openness is 7.57% for cotton, 0.23% for maize, 0.06% for soybeans, and 0.10% for wheat, which implies a low degree of openness in the Turkish agricultural sector. Policy makers can only influence the local prices of agricultural commodities through direct purchases/sales of these commodities in the international market, rather than exchange rate changes. Fourth, the local agricultural commodity markets can be viewed as alternative investment markets to world oil and currency markets for portfolio risk diversification purposes. This follows directly from the neutrality of the local agricultural commodity markets. Fifth, as pointed out by Zhang et al. (2010) the long run neutrality of agricultural commodity prices to oil and exchange rate changes is a cautionary phenomenon that should be accounted for by CGE and other numerical models. Although the Turkish agricultural sector seems to be protected from world energy shocks, the sector cannot benefit from any opportunities that may arise due to changes in the world energy markets and the exchange rate. Provided that the sector will benefit from openness, agricultural policy makers should try increasing the degree of openness in Turkish agricultural sector rather than protecting the sector from international factors. As the sector may become more energy intensive, the biofuel production may rise, and the sector may become more open through time, policy makers must continuously follow the sensitivity of agricultural prices to oil shocks and exchange rate movements for sound policy options. Investors must also continuously follow the developments in the local agricultural commodity markets. As the sensitivity of the local markets to oil and exchange rate shocks increase, investors must make necessary adjustments to maintain the same risk level for their portfolios. Our results suggest that agricultural prices in countries at different stages of development may show different degrees of responsiveness to changes in world energy prices. A new line of research would focus on the direct and indirect effects of oil on individual agricultural prices in emerging markets, developing countries, and underdeveloped countries. Another possible extension would be on impact of oil price changes on prices of various consumption food items. Additionally, studies to date have employed linear approaches to investigate the direct and/or indirect impacts of oil prices on agricultural commodity prices. Although we focused on the linear association between oil prices and agricultural prices, volatility of and structural breaks in prices may result in nonlinear linkages. This suggests future directions for researchers to re-examine this issue by focusing on nonlinear methods. Asymmetric affects of oil price shocks may also prove to be a fruitful area to explore.