The stability of the global wheat trade in the post-soviet space: A trade duration approach

Jaghdani, Tinoush Jamali; Glauben, Thomas; Götz, Linde Johanna; Svanidze, Miranda; Prehn, Sören Article The stability of the global wheat trade in the post-soviet space: A trade duration approach German Journal of Agricultural Economics (GJAE) Provided in Cooperation with: Gesellschaft für Wirtschaftsund Sozialwissenschaften des Landbaues e.V. (GEWISOLA) Suggested Citation: Jaghdani, Tinoush Jamali; Glauben, Thomas; Götz, Linde Johanna; Svanidze, Miranda; Prehn, Sören (2024) : The stability of the global wheat trade in the post-soviet space: A trade duration approach, German Journal of Agricultural Economics (GJAE), ISSN 2191-4028, TIB Open Publishing, Hannover, Vol. 73, Iss. 3, pp. 1-26, https://doi.org/10.52825/gjae.v73i3.1353 This Version is available at: https://hdl.handle.net/10419/313413 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ German Journal of Agricultural Economics Vol. 73 (2024), No.3, 1-26 Original Research Article https://doi.org/10.52825/gjae.v73i3.1353 © Authors. This work is licensed under a Creative Commons Attribution 4.0 International License Submitted: 14 Aug. 2024 | Accepted: 04 Dec. 2024 | Published: 14 Jan. 2025 The Stability of the Global Wheat Trade in the Post-Soviet Space: A Trade Duration Approach Tinoush Jamali Jaghdani , Thomas Glauben , Linde Götz , Miranda Svanidze and Sören Prehn Leibniz Institute of Agricultural Development in Transition Economies (IAMO), Halle (Saale), Germany Correspondence: Tinoush Jamali Jaghdani, [email protected] Abstract: The collapse of the Soviet Union and the Iron Wall as well as the emergence of Kazakhstan, Russia, Ukraine and Romania as major actors on international grain markets since 2000 had increased the hope for a more stable international grain market. However, various short-run trade policy interventions of post-Soviet grain exporting countries during the 2000s and 2010s have caused temporary disruptions in global grain supply chains. Moreover, the growing number of protectionist state interventions and sanctions since the beginning of the 2010s, as well as growing geopolitical tensions, may also fragment global trade and thus threaten the stability of grain trade relations. Against this backdrop and given the current public and political debates about suitable “de-risking strategies” to stabilise international trade, this article aims to explore the stability of the global wheat trade in terms of the duration dependence of trade relations between the major grain exporters and their destinations from 2001 to 2021. We test whether there are differences between the relatively “new” post-Soviet exporters that have emerged and the “old” ones. Furthermore, we examine the correlation between the number of trade partners and the trade duration. We employ a discrete-time hazard model to annual trade data to estimate the baseline hazard and survival rate for eleven major wheat exporting countries. The results indicate that, by having overall duration dependence, no different pattern in trade stability can be identified between “new” and “old” actors, and initial indications suggest that having more trading partners favours the survival of trade relations. Keywords: Global Wheat Supply Chain, Trade Duration, Baseline Hazard, Logit, KaplanMeier Estimator, Survival Rate, Post-Soviet 1 Introduction For the vast majority of countries, food security cannot be guaranteed by domestic production and the importing of different food and agricultural products is essential (Brenton et al., 2022). Recent studies show that approximately two-thirds of the world’s population is dependent on imported foods (Kinnunen et al., 2020). The dependency on the grain trade has historical roots and is not a new phenomenon. For instance, historical records show that, as the Roman population increased, it became more dependent on wheat imports from other Mediterranean countries. Josephus recorded in his Bellum Judaicum that, in the late 70s A.D., North Africa (excluding Egypt) fed Rome for two-thirds of the year, while Egypt, in addition to providing financial resources for Rome, sent sufficient amounts of grain to feed the city for the rest of the year (Rickman, 1980). The shape of this trade has changed in modern times due to the discovery of the New World, grain cultivation in new regions and a drop in transport costs, which caused wheat prices to fall in the late 19th and early 20th centuries (O’Rourke, 1997). Tsarist 1 Jaghdani et al. | Ger J Agr Econ 73 (2024), No. 3 Russia was a major producer and exporter of grain before the Bolshevik Revolution in 1917 (Goodwin, Grennes, 1998). As its successor, however, the Soviet Union could never become the major wheat exporter that Tsarist Russia was, and it instead became a wheat importer. Grain imports from the Soviet Union were rather affected by geopolitics and the political economy, especially after World War II (Atkin, 1995). However, the post-World War II grain trade regime changed after the collapse of the Soviet Union due to structural changes to the agricultural systems in the new states in the post-Soviet space. With emerging wheat exporters such as Kazakhstan, Russia, Ukraine and Romania (collectively referred to as the ECA1 wheat exporters in this study), the international wheat trade network was expected to become more stable and resilient (González-Esteban, 2018). However, the 2007-08 food crisis, followed by increased food price volatility, the COVID-19 pandemic and increasing geopolitical tensions since around 2020, such as the US-China trade conflict and the Russian invasion of Ukraine, have brought into question the stability and resilience of the international food network (Glauben, Duric, 2024), especially for grains (Glauben et al., 2022; Yugay et al., 2024). Our analysis is motivated by two diverging issues: on the one side exporting and importing countries are exposed to the vulnerabilities of the international wheat trade network (GutiérrezMoya et al., 2021) and on the other side international trade relations also offer a safety net in the presence of climate or policy-related supply bottlenecks (Glauben, Svanidze, 2023). To investigate the stability of trade relations, we measure the temporal trade duration of the trade relations of the top eleven wheat exporters with their import partners from 2001 to 2021 and focus on two issues. Firstly, we look at whether the newly emerged ECA wheat exporting countries are showing remarkably different trade duration patterns compared to the historically established wheat exporting nations, such as the USA, Canada, Australia, Germany, France, the UK and Argentina. By selecting these top eleven major wheat exporters, the dynamic development in the international wheat market can be represented. Secondly, we explore whether there are differences in trade duration between exporters with a larger number of trading partners (i.e. more diversification) and those with a smaller number of trading partners (i.e. less diversification). The latter issue is particularly due to intensive recent debates in the political and public sphere about suitable “de-risking strategies” to secure supplies and thus the stability of trade relations (Glauben, Duric, 2024). On the one hand, it is argued that greater diversification, i.e. a large number of importing countries, reduces the risk of supply gaps via trade disruptions. At the same time and in contrast, it is argued that more stable international trade relations can be better achieved with a smaller number of selected and particularly reliable (“friendly”) trading partners and even self-sufficiency. We apply a trade duration approach (Besedeš, Prusa, 2006a, 2006b; Hess, Persson, 2012) by estimating baseline hazard and survival rates for the wheat trade. Duration approaches complement the well-known gravity models (e.g. Anderson, 2011; Anderson, Yotov, 2012), as gravity approaches focus on the trading volumes but are not suitable for measuring the duration of trading relationships. Relatively long-term trade survival, i.e. high stability in the sense of comparatively few interruptions, between exporters and importers could be an additional indicator of well-established and robust wheat markets. Trade duration approaches are employed to examine and identify the stability of trade relationships (Córcoles et al., 2015; Obashi, 2010; Wang et al., 2019). When it comes to the substantial costs and risks of (re-)entering into and (re-)establishing new trade relationships, maintaining existing trade relationships might be crucial for the profitability of export and import partners. Additionally, rather longer-term trade relationships between partners could be a sign of higher vertical coordination and lower transaction costs (Ketokivi, Mahoney, 2020). Stable trade relations might allow agricultural and food producers in exporting countries to generate more revenues at lower risk and might ensure the availability of food products in importing countries (Engemann et al., 2022), particularly in the context of recurring market interventions and increasing geopolitical tensions. With a specific focus on the grain sector, we can observe the precise market conditions that lead (or can lead) 1 Europe and Central Asia (ECA) is a categorization by the World Bank that refers to a list of countries in Southeast and Eastern Europe, the Caucasus and Central Asia: https://www.worldbank.org/en/region/eca 2 Jaghdani et al. | Ger J Agr Econ 73 (2024), No. 3 to unstable trade relations. Market conditions can be natural fluctuations of supply or demand over the years and at certain times. Fundamental market factors (such as weather events) rather than certain trade policies can be the cause of price formation and, as a result, can lead to changes in supply or demand. In certain years, when natural conditions are relatively good, importing countries can have high levels of self-sufficiency and may suspend imports, at least temporarily, as they are no longer dependent on them. Some countries have wheat self-sufficiency policies that may be in place for some years when there is higher rainfall (see Jaghdani et al. (2024a) for an example of the volatile wheat trade relationship between Russia and Iran). From the perspective of the exporting country, this can be considered an unstable trade relationship. From the perspective of the importing country, it is not necessarily the case. However, stable or long-term trading relationships are not a “gold standard” indication of the functionality of markets. For example, frequent changes between trading partners, i.e. rather unstable relations, could be a sign of a high level of adaptability and flexibility and could thus add to the functionality of markets. This article explores several specifications of the trade duration model by distinguishing between the single-spell and the multiple-spells data structures in a discrete-time hazard framework that is estimated in a logistic regression. Furthermore, as a robustness test, we compare the estimated survival rate of the discrete-time hazard model with the traditional Kaplan–Meier estimator, which assumes continuous-time structures (Kaplan, Meier, 1958). This article is organised as follows: Section 2 presents background information on developments in the international wheat market, and Section 3 provides a literature review on food trade duration. In Section 4, the methodology and data are presented, while the results are presented in Section S before conclusions are drawn in Section 6. 2 Emerging ECA Wheat Exporters in the International Wheat Market 2.1 Structural Changes Kazakhstan, Russia and Ukraine (KRU) were the three main wheat and coarse grain producing republics in the Soviet Union. However, on aggregate, the Soviet Union was a grain importer rather than an exporter (Fellmann et al., 2014). Upon the collapse of the Soviet Union, there was a huge drop in the size of agricultural production in the new-born republics and most of them turned into importers of many agri-food products. Later on, a series of economic reforms were started in their agricultural sectors, with these reforms having serious consequences, especially in KRU. During this transitional period, the institutions that had been left from the collectivisation of agriculture in the Soviet Union were transformed. Soviet agriculture was based on state-owned farms (Sovkhozy) and collective farms (Kolkhozy), which were demolished in the transitional period, and smallholder farmers with small to medium sized farms, large private farms or agricultural corporates emerged during this time. These agricultural corporates, often called agro-holdings, as well as large private farms, were able to increase productivity and efficiency, attract financial resources for investments, and integrate into international markets and global food supply chains. Downsizing the livestock sector was another change in this period. Through these structural changes observed since 1992, KRU became three major exporters of grains, oilseeds and feedstuff to international markets (Gafarova et al., 2015; Liefert, Liefert, 2015). Romania is another country which was a member of the former Socialist Bloc that has joined the top wheat exporters in recent years. It turned first into a wheat importer from 1990 to 1992, but, following a reform policy (FAO, 1993), Romania has since observed a total factor productivity increase in its agricultural sector (Tebaldi, Gobjila, 2018). Figure 1 shows that Russia has transitioned from a country importing more than 13.5 million tonnes of wheat in 1992 to an exporting country with exports amounting to more than 40 million tonnes of wheat in 2018. Similarly, Ukraine, Kazakhstan and Romania also became major exporters. 3 Jaghdani et al. | Ger J Agr Econ 73 (2024), No. 3 Figure 1. Net wheat exports from Russia, Ukraine, Kazakhstan and Romania (1991 to 2021) Data source: UN Comtrade and Trade Map The importance of agricultural exports is not similar but of noticeable economic importance for all four countries. For the 2017-2020 period, exports of agri-food products accounted for 42% of Ukraine’s, 10% of Romania’s, 6% of Russia’s and 6% of Kazakhstan’s total exports in value terms approximately. Specifically, for the same period, exports of wheat accounted for 7% of Ukraine’s, 2% of Romania’s, 2% of Russia’s and 2% of Kazakhstan’s total exports in value terms approximately2. The agricultural exports of these countries mainly consist of grains, oilseeds and oilseed oil. Figure 2 shows that Russia and Ukraine are among the top five and Kazakhstan and Romania are in the top eleven wheat exporters in the world. Since 2000, Russia was able to slightly pass the USA and become the largest wheat exporter in the world. In 2019, Russia, Ukraine, Romania and Kazakhstan accounted for 18%, 11%, 3.5% and 3% of world wheat exports, respectively. 2 Data sources are UN Comtrade for wheat exports and WTO Statistics for agri-food and total exports (https://stats.wto.org/). -20 -10 0 10 20 30 40 50 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Net export (million tonnes per year) Year Russia Ukraine Kazakhstan Romania 4 Jaghdani et al. | Ger J Agr Econ 73 (2024), No. 3 Figure 2. The share of wheat exports among the top 11 exporters for global wheat exports, 2001-2020 (HS Code: 1001) Source of Data: UN Comtrade and Trade Map By considering the average exports of wheat during the 2017-2020 period, Egypt, Indonesia, Bangladesh, the Philippines, Morocco, Tunisia, and Turkey were the main importers of Russian wheat while Egypt, Turkey, Bangladesh, Sudan, Nigeria, Yemen and Vietnam were the main importers of Ukrainian wheat. In the same period, politically unstable countries such as Lebanon, Libya and Ethiopia were on the list of wheat importers from both Russia and Ukraine. The processing and export of Russian wheat as flour has been observed among different countries in the Middle East and in particular Turkey (Heigermoser et al., 2022). For Kazakhstan, as it is a landlocked country, its main wheat export destinations are its neighbours, and, for the period of 2017-2020, Uzbekistan, Tajikistan, Afghanistan and China were the main importers of Kazakhstan’s wheat (Gafarova et al., 2015). Egypt, Jordan, South Korea, Sudan and Israel are major importers of Romanian wheat3. Currently Kazakhstan, Russia, Ukraine and Romania are the biggest net wheat exporters of the former Socialist Bloc and have therefore been selected as new actors for this article. These countries faced many challenges in producing high-quality wheat in the post-Soviet space (Carson, Edwards, 2009). Although the quality of wheat can be determined by a range of factors, protein content is one of the main indicators. Durum wheat with a protein content of approximately more than 13% is suitable for pasta. Hard wheat with a protein content of approximately more than 13.5% is suitable for highprotein flour bread. With a protein content of 11.5% to 13.5%, it is used for flour bread. Soft wheat with a protein content of 8.5% to 9.5% is suitable for cakes, biscuits, puddings and pastry. Additionally, soft wheat with a protein content of 9.5% to 11.5% is a thickener and suitable for groceries. Mixed wheat types with a protein content of 9.5% to 11.5% are used in white Japanese and yellow Chinese noodles. Finally, mixed wheat types with a protein content 3 Data source of wheat export destination: UN Comtrade (https://comtradeplus.un.org/) and Trade map (https://www.trademap.org). 5 Jaghdani et al. | Ger J Agr Econ 73 (2024), No. 3 of 10.5% to 12.5% are used for baking flatbread, which is common in the Middle East and North Africa (MENA) region (Wrigley, 2009). There is not much information available on the type of wheat exported by ECA wheat exporting countries. The little data available shows that the bulk of wheat exported from Russia, Ukraine and Romania has a protein content of between 11% and 12%, which makes it ideal for flatbread production in the MENA region (AEGIC, 2016a, 2016b; Marinciu et al., 2022). In addition to its lower price, this is a major reason why these countries export to the MENA region. In contrast, Kazakhstan produces higher-quality wheat, which is suitable for wheat flour (Altaibaeva et al., 2016). In 2021, Kazakhstan’s durum wheat exports amounted to 3.8%. However, unlike Russia, Ukraine and Romania, Kazakhstan is a landlocked country with limited access to the international market (Svanidze et al., 2019). Therefore, wheat exports have remained relatively stable since 2001 due to transport costs. However, by expanding its flour-milling industry and producing higher-quality wheat for bread, Kazakhstan has increased its wheat flour exports to neighbouring countries (Altaibaeva et al., 2016). UN Comtrade wheat flour export data shows that, since 2005, Kazakhstan and Turkey have been the leading wheat flour exporters. Kazakhstan’s maximum wheat flour export was 2.39 million tonnes in 2016. Similar to wheat, the flour is also mainly exported to Afghanistan, Uzbekistan and Tajikistan. However, because the focus of our article is on wheat, we avoid further analysis of wheat flour exports, as it is a different commodity. 2.2 Trade Policies Since the emerging ECA exporters joined the global wheat export market, they have followed very unstable trade policies. Russia has repeatedly implemented export-related restriction policies (ERPs) for wheat (Svanidze et al., 2022), consisting of an export ban in 2010-11, and an export tax in 2007-08 as well as in 2015 (Figure 3). Since 2020 export quotas have been imposed temporarily, and since 2021 a flexible export tax system has been in place (Götz, Svanidze, 2023). Figure 3. Russia’s historical monthly wheat export development considering the trade policy volatilities and domestic and international prices in Russian ruble (RUB) (2006-2022) Source: Götz, Svanidze (2023) 2020 export quota + ban 2015 export tax 2007/08 export tax 2010/11 export ban free trade since 2021 quota + floating tax 6 Jaghdani et al. | Ger J Agr Econ 73 (2024), No. 3 Ukraine has also implemented grain ERPs over the last two decades. For instance, from July 2007 to October 2007, Ukraine introduced a restrictive total grain export quota of 12,000 tonnes (3,000 tonnes each for wheat, barley, rye and corn), which essentially acted as an export ban. On another occasion, from October 2010 to July 2011, Ukraine implemented a grain export quota of 6.2 million tonnes for total grain exports. Further, from July 2011 to December 2011, export taxes were introduced. These were set at 9% for wheat, 12% for maize and 14% for barley (Fellmann et al., 2014). Additionally, Götz et al. (2013) found that rumours of export restrictions caused wheat price volatility in Ukraine’s domestic market from 2005 to 2012. Kazakhstan also implemented grain ERPs, including an export ban from April 2008 to September 2008 (Fellmann et al., 2014). Further examples of grain ERPs can be found for Ukraine, Kazakhstan and a few for Romania4. Among the old actors, Argentina has had the highest number of ERPs since 2001 (AMIS, 2024). Australia also had a few registered export taxes and licensing requirements in place from 2003 to 2014. No further ERPs on wheat were found for wheat exports from the USA, Canada, the UK and the EU (AMIS, 2024). 3 Literature Review The concepts of hazard or survival were mainly developed within the natural sciences, such as in biology and medicine (e.g. Cox, 1972; Kaplan, Meier, 1958), or, for instance, in labour economics and poverty research (e.g. Glauben et al., 2012). The “death” or loss of an object (or failure of an event) of a study was the main characteristic of time-to-event data in the natural sciences, for example, which “technically” (i.e. in terms of the model specifications) corresponds with an interruption in trade relations in trade duration models (Hess, Persson, 2012). Almost all trade duration studies lack a rigorous theoretical framework in the sense of consistent and clear behavioural assumptions (Engemann et al., 2022; Hess, Persson, 2011). The seminal works by Besedeš, Prusa (2006b, 2006a) build on reduced-form models by analysing trade duration for many commodities and countries. Trade duration models became theoretically founded with the introduction of the heterogeneous firm trade model by Melitz (2003). Consequently, extensive and intensive margins of trade were studied by considering a firm’s heterogeneity (Besedeš, Prusa, 2011). On the other hand, trade duration models also underwent significant methodological advancements over the last decade. In particular, the discretetime hazard model was introduced by Hess, Persson (2012), whereas, prior to that, the survival of trade relations was explored using the non-parametric Kaplan-Meier estimator (Kaplan, Meier, 1958) and COX proportional hazards model (Cox, 1972). Most recent studies on the international agri-food trade duration specifically focus on perishable food items and the effects of trade policies and trade barriers. For instance, Peterson et al. (2018) find that sanitary and phytosanitary measures decrease trade duration, particularly in the first years of policy implementation for highly perishable fruit and vegetable products in the US. Likewise, Luo, Bano (2020) investigate New Zealand’s dairy exports and trace back the decreasing likelihood of continuous trade relationships to technical barriers to trade implemented by importing countries. A particular and novel strand of the trade duration literature focuses on the analysis of the seafood trade, for which perishability is an even more critical issue (e.g., Jaghdani et al., 2024b). With the improvements in the availability of firm-level trade transaction data, the number of trade duration studies conducted at the firm level also increased, but they are still rare. For instance, Gullstrand, Persson (2015) find the trade duration of food items for Swedish firms to be higher in core markets (the firms’ primary export destinations) than in peripheral markets (the firms’ minor export markets). In another study, Jaghdani et al. (2024) apply a heterogeneous firm trade model (Melitz, 2003) to firm-to-firm trade transaction data on the salmon trade, finding that the salmon trade duration is relatively short-lived (around 2.5 years on average). 4 The records of different export restrictions can be followed via the Global Trade Alert (GTA) platform: https://www.globaltradealert.org/. 7 Jaghdani et al. | Ger J Agr Econ 73 (2024), No. 3 Although the EU is a major importing market for Norwegian salmon, they found that the stability of trade between different partners is not huge for major and minor importing markets or different continents. To the best of our knowledge, Imamverdiyev et al. (2015) and Jaghdani et al. (2020) are the only studies that analyse the duration of wheat exports. Imamverdiyev et al. (2015) exclusively focused on Kazakhstan. They concluded that the trade duration of Kazakhstan’s wheat exports can be explained by high trade costs, the use of local production factors, and a lack of price competitiveness and experience. By focusing on France, Jaghdani et al. (2020) observed more stable trade relations between France as the exporter and other EU importing countries compared to non-EU importing countries. In contrast to the available literature, this article is the first comparative study to focus on the analysis of trade duration for the largest eleven wheat exporting countries. Moreover, this study also makes a methodological contribution to food trade duration studies by estimating the survival function of each wheat exporter through a discrete-time hazard model (Hess, Persson, 2012). 4 Methodology and Data In this Section, we first present the discrete-time hazard model building on Hess, Persson (2012), Peterson et al. (2018) and Engemann et al. (2022). However, the empirical approach in this study is limited to estimating the baseline hazard model (Tutz, Schmid, 2016) and the corresponding survival function. Second, as a robustness test, we compare the results of the novel discrete-time estimator to the standard classical non-parametric Kaplan-Meier survival estimator, which is presented in the subsequent Section. Last, we discuss the estimation strategy and data structure. 4.1 Baseline Hazard Model We start by defining a spell of trade. A spell of trade (i, i=1,…,n) is defined as the realisation of the trade relationship for a period with uninterrupted trade of a product between an exporter country X and an importer country M (Besedeš et al., 2023). The duration of each spell (Ti) is then defined as the number of consecutive years with non-zero trade between partners. Let Ti be a non-negative, continuous random variable that measures the survival time of the ith trade relationship without interruption (or ith spell). Next, we define a set of discrete-time intervals [t1, t2,.., tk, … tmax] when k=1 means tk = t1 = 0 and 1 ≤ k < max. The probability that the ith trade relationship (or ith spell) ends in tkth time interval is conditional on the uninterrupted trade relationship (or spell) surviving up to the beginning of that time interval and on a set of explanatory variables in the model. When trade failure happens in a particular discrete time interval [tk, tk+1), then the duration of each spell or Ti = tk+1 (Tutz, Schmid, 2016, chapter 3). This means that the duration of the trade spell is the length of time the individual trade relation existed, usually recorded as the last period in which the trade is observed (Singer, Willett, 1993). As suggested by Hess, Persson (2012), in the discrete-time hazard model, trade duration is estimated as a conditional probability that the trade of a product will terminate in a particular discrete time interval [tk, tk+1) defined above (1 ≤ k < max). This conditional probability of trade interruption, which is also defined as a discrete-time hazard rate (ℎ𝑖𝑖𝑖𝑖), can be formulated as follows: ℎ𝑖𝑖𝑖𝑖 =ℎ(𝑡𝑡|𝒙𝒙)=𝑃𝑃(𝑇𝑇𝑖𝑖<𝑡𝑡𝑖𝑖+1|𝑇𝑇𝑖𝑖≥𝑡𝑡𝑖𝑖,𝑥𝑥𝑖𝑖𝑖𝑖) (1) =𝐹𝐹�𝛾𝛾𝑖𝑖+ 𝒙𝒙𝑖𝑖𝑖𝑖 ´𝜷𝜷� where x'ik is a vector of time-varying covariates, P() is conditional probability and β is a vector of coefficient parameters that should be estimated. Furthermore, 𝛾𝛾𝑖𝑖 is the baseline hazard as 8 Jaghdani et al. | Ger J Agr Econ 73 (2024), No. 3 Figure 5. Survival rates for trade spells between each exporter and all their partners (multiple-spells data structure) Source: study findings Figure 6. The survival rate for trade spells between each exporter and all its partners (single-spell data structure) Source: study findings 15 Jaghdani et al. | Ger J Agr Econ 73 (2024), No. 3 Based on spells with a duration of up to 18 years, the survival rates amount to 13%, 11% and 10% for Romania, Ukraine and Germany, meaning they are “medium survival rate” exporters. As Romania is a rather new actor in the world market, the dataset is limited to 18 years. However, its survival rate of up to 18 years is similar to that of Ukraine and Germany. Ukraine and Romania have similar (and even slightly higher) survival rates compared to an established actor such as Germany. The UK, with a 7% survival rate, and Kazakhstan, with an 8% survival rate, have similar patterns considering the 18 years of the longest trade spell in the multiplespells setting. UK wheat exports are declining and mainly consist of fodder wheat (Fradgley et al., 2023), which competes with other feedstuff. Kazakhstan has limited export possibilities due to its geographical landlocked conditions and thus limited access to the world market (Svanidze et al., 2019). If Kazakhstan restricted its wheat exports to only a few neighbouring countries, it could have a higher survival rate, as higher volumes could be traded with fewer partners and each partner would become more important. Nevertheless, as Kazakhstan exports to countries that are further away, more unstable, short-lived trade spells appeared that reduced its survival rate. With 2.5%, the lowest survival rate after 18 years of trade is found for Argentina. This rather low rate could result from unstable export trade policies, unstable exchange rate policies, periodically high domestic inflation or regular implementation of ERPs. Table 3. Survival rates of the two data structure (multiple-spells and single-spell) estimation frameworks for comparison at the spell length of 11 and 18 years using the discrete-time hazard model Survival rate retrieved from logit estimation for spells of length 11 Survival rate retrieved from logit estimation for spells of length 18 Countries Multiplespells data Singlespell data Diff* Multiplespells data Singlespell data Diff* Old actors USA 0.231 0.140 0.091 0.198 0.084 0.114 Canada 0.254 0.135 0.119 0.227 0.106 0.121 France 0.206 0.099 0.107 0.16 0.058 0.102 Australia 0.252 0.152 0.100 0.222 0.127 0.095 Argentina 0.064 0.014 0.050 0.025 0.004 0.021 Germany 0.118 0.040 0.078 0.103 0.024 0.079 UK 0.090 0.024 0.066 0.07 0.009 0.061 New actors Russia 0.228 0.105 0.123 0.19 0.076 0.114 Ukraine 0.130 0.036 0.094 0.11 0.024 0.086 Romania 0.134 0.038 0.096 0.134 0.031 0.103 Kazakhstan 0.097 0.028 0.069 0.08 0.014 0.066 *Diff measures the difference between survival rates of the multiple-spells and single-spell data structures Source: study findings The results of the estimated survival rates through the discrete-time hazard model can be cautiously summarised as follows: Initially, we cannot see different types of clusters among the “new” and “old” actors. Subsequently, it seems that the largest exporters (namely the USA, Russia, Canada and France), both in terms of the amount of wheat traded and the number of trading partners, have the highest survival rates. Furthermore, high wheat quality could encourage long-lasting bilateral trade relations (e.g., Australia and Canada). Moreover, limited market access, such as in Kazakhstan as a landlocked country, seems to work against longlasting trade. In the context of the logit estimator, survival rates with a single-spell structure are substantially underestimated. This issue is further tested and discussed in Section 5.3. To visualize the relationship between survival rates and the number of trading partners, we have plotted the number of importing trading partners of each exporter (as an indicator of diversified partners) versus the average length of spells of each exporter (see Section 5.1) and the survival rate of the 18-year spell (see Figure 5 and Table 3). The results of this depiction, which should be viewed very conservatively due to the low number of observations, show that 16 Jaghdani et al. | Ger J Agr Econ 73 (2024), No. 3 there is a pattern between the average length of spells of each exporter and the number of partners (see Figure 7, panel A), with a correlation coefficient of 63% at 5% significance. We can also recognise a rising correlation (with a correlation coefficient of 56% at 10% significance) between survival rates and the number of partners (see Figure 7, panel B). The result shows that, on average, exporters with more trading partners also have more stable trade relations. This may result in long-term bilateral relations with many individual trading partners, as exporters can switch between individual partners and therefore quickly find a “new” buyer in the overall world market. Hence, stability in the overall pool of destination markets can be achieved either by (a) flexible adaptation or (b) continuity of individual relationships. Figure 7. The relationship between the number of trading partners, the average length of the spells and the survival rate of the 18-year spells Source: study findings 5.3 Robustness Test of the Survival Rates with the Kaplan-Meier Estimation and Study Limitations Figure 8 and Figure 9 show the estimated survival rates using a non-parametric Kaplan-Meier estimator for the multiple-spells and single-spell structures, respectively. In the multiple-spells structure, survival rates as well as their patterns are nearly identical across the Kaplan-Meier estimation method and the discrete-time hazard model with the logistic regression approach (compare Figure 5 and Figure 8). In contrast, the survival rates of the Kaplan-Meier estimator are higher in every instance with the single-spell structure (compare Figure 6 and Figure 9). Additionally, the pattern of survival rates in Figure 9 is different compared to all other estimators (Figure 5, Figure 6 and Figure 8). For ease of interpretation, Table 4 compares the survival rates derived from the Kaplan-Meier estimator for spells of 11 and 18 years between the multiple-spells and single-spell data structures. For instance, in the multiple-spells structure, Russia has a survival rate of 22.8% at the spell length of 11 years, which decreases to 19% at the spell length of 18 years, whereas the comparable changes are 62.1% and 52.2% in the singlespell structure, respectively. 17 Jaghdani et al. | Ger J Agr Econ 73 (2024), No. 3 Figure 8. Kaplan-Meier survival estimator in the multiple-spells data structure Source: study findings Figure 9. Kaplan-Meier survival estimator in the single-spell data structure Source: study findings 18 Jaghdani et al. | Ger J Agr Econ 73 (2024), No. 3 Table 4. Survival rates of the two data structure (multiple-spells and single-spell) estimation frameworks for comparison at the spell length of 11 and 18 years using the Kaplan-Meier (KM) estimator Survival rate retrieved from KM estimation for spells of length 11 Survival rate retrieved from KM estimation for spells of length 18 Countries Multiplespells data Singlespell data Diff* Multiplespells data Singlespell data Diff* Old actors USA 0.231 0.514 -0.283 0.198 0.403 -0.205 Canada 0.253 0.556 -0.303 0.227 0.535 -0.308 France 0.206 0.443 -0.237 0.16 0.372 -0.212 Australia 0.252 0.51 -0.258 0.222 0.477 -0.255 Argentina 0.0637 0.266 -0.2023 0.0255 0.171 -0.1455 Germany 0.118 0.392 -0.274 0.103 0.302 -0.199 UK 0.0903 0.24 -0.1497 0.0703 0.18 -0.1097 New actors Russia 0.228 0.621 -0.393 0.19 0.522 -0.332 Ukraine 0.13 0.483 -0.353 0.11 0.395 -0.285 Romania 0.13 0.474 -0.344 0.13 0.4 -0.27 Kazakhstan 0.0973 0.348 -0.2507 0.0803 0.285 -0.2047 *Diff is multiple-spells data survival rate minus single-spell data survival rate Source: study findings In a nutshell, the following can be stated: firstly, with a multiple-spells structure, the survival rates for the Kaplan-Meier estimator are almost of the same order of magnitude as for the logit estimator. Secondly, the survival probabilities are relatively overestimated for the Kaplan-Meier estimator considering the single-spell in comparison to the other estimators. Thirdly, the previously mentioned patterns between the various exporting countries also apply to the KaplanMeier estimator with multiple-spells. The findings of this article must be considered in light of some limitations. As mentioned in the introduction, stable or long-term trading relationships cannot necessarily be considered a “gold standard”. Frequent changes between trading partners, i.e., unstable relations, could be a sign of a high level of adaptability and flexibility. Furthermore, as the focus of this article is on the estimation of baseline hazard and survival rates, i.e. the stability (duration) of trade relations (between nations) regardless of the extent of trade relations, considering the extent of trade relations would go beyond the aim of this study and address another (additional) question. In this regard, the quantities of trade per destination are not considered as a driver. Therefore, the results for a country with very stable trade relations with its main trading partners may be strongly influenced by some fluctuations in trade regimes with some very minor trading partners which are just above the 50-tonne threshold applied. Although this is an issue for all exporters in our dataset and all of them can be affected by that at different magnitudes, it can be analysed in the full hazard model and the quantity of trade could be one of the different covariates that are common in trade duration studies. The level of domestic wheat production of the importer could be another covariate. These are topics for future research. 6 Conclusion This article compared trade duration differences between eleven major wheat exporters for the 2001-2021 period by using aggregated six-digit HS code wheat trade data. We addressed the research question of whether there are different survival rate clusters between newly emerging ECA wheat exporting countries (KRU and Romania) and “old” actors (the USA, Canada, Australia, Germany, France, Argentina, and the UK) and if there is a correlation between trade stability and trade partner diversification. Trade duration data was used to estimate the baseline hazard ratio of the wheat trade between each major exporter and all their importers. The 19 Jaghdani et al. | Ger J Agr Econ 73 (2024), No. 3 estimated baseline hazard through logistic regression was employed to estimate the discrete time-to-event survival function for each exporter. Furthermore, the classical non-parametric Kaplan-Meier estimator was also employed to test the robustness of the estimated survival rates. The results show that, primarily, we can observe a kind of duration dependence in continuing trade relations that differs between countries, and longer trade relations in the past can increase the probability of trade perpetuation. Secondly, there is no clear distinction between “new” and “old” actors in terms of the stability of trade relations. The trade survival pattern is diverse between the two groups, suggesting that factors such as the types of wheat traded, the size of the trade, the geographical position of the exporters, and unstable macroeconomic conditions might play a larger role than the length of the presence of an exporting country on the world market. Our estimation results indicate that the discrete-time hazard model within the logit estimation approach provides comparable but remarkably lower survival rates in the single-spell data structure compared to the multiple-spells data structure. Moreover, in the multiple-spells data structure, the discrete-time hazard model and the continuous-time KaplanMeier estimator generate the same numerical outputs. However, we observe that the survival rates are significantly inflated within the Kaplan-Meier estimator in the single-spell data structure compared to the multiple-spells data structure or even to the discrete-time hazard model in the single-spell data structure. The overestimation of the survival rates of the single-spell data by the Kaplan-Meier estimator and underestimation by the logit model are due to the ignoring of disruptions in the Kaplan-Meier estimator and the counting for disruptions in the logit model. This is not the case for the multiple-spells structure. Therefore, we can conclude that a multiple-spells data structure is a more robust choice across different estimators and should be preferred in further regression analyses with discrete-time hazard models or the COX proportional hazards model for the wheat market (or any other commodities markets), as it can provide more reliable baseline hazards. Moreover, we observe a positive correlation between exporters’ trade stability and the diversification of trade partners, which is worthy of further research in the future with further observations and/or with firm-level data. With future research and a more mature, strengthened database, different covariates that are common in gravity types of studies could be added to each regression to find influential factors on hazard ratios. Additionally, comparative variables could be added to test the role of the intensive margin in the wheat trade. Furthermore, wheat production in importing and exporting countries will be a potential variable that can be considered. Acknowledgment We gratefully acknowledge financial support from the Research Council of Norway through the project SEAVID19 Ensuring economic sustainability in the seafood industry during crisis, grant No. 326647. We are also thankful to the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) for their financial support through the project BlackSeaGrain: Russia’s growing wheat export business: What drives export growth, trade stability and pricing strategies?, project No. 514853816. This research was conducted during the presence of the five authors at the Leibniz Institute of Agricultural Development in Transition Economies (IAMO) in Germany. Open Access publication of this manuscript is enabled by GJAE. Data Availability Statement The trade data used in our empirical analysis are provided at https://doi.org/10.15456/gjae.2025002.0938236463. 20 Jaghdani et al. | Ger J Agr Econ 73 (2024), No. 3 References AEGIC (Australian Export Grains Innovation Centre) (2016a): Russia’s wheat industry: Implications for Australia. Report. Australian Export Grains Innovation Centre (AEGIC). https://www.aegic.org.au/wp-content/uploads/2021/03/AEGIC-Russia-factsheet.pdf. AEGIC (2016b): Ukraine: An emerging challenge for Australian wheat exports. Australian Export Grains Innovation Centre (AEGIC). https://aegic.org.au/wpcontent/uploads/2021/03/Ukraine-Supply-Chain-Full-Report.pdf. Altaibaeva, Z., Bauer-Kenzhebolatova, M., Zhaltyrova, O. (2016): Development of the Flourmilling Industry in the Republic of Kazakhstan in Modern Times. International Journal of Economics and Financial Issues 6 (2): 1-8. https://dergipark.org.tr/en/pub/ijefi/issue/31980/352595. AMIS (2024): AMIS Policy Database. Agricultural Market Information System (AMIS), FAO. https://web-archive.oecd.org/agriculture/amis-policy-database/index.htm. Anderson, J.E. (2011): The Gravity Model. Annual Review of Economics 3: 133-160. https://doi.org/10.1146/ANNUREV-ECONOMICS-111809-125114. Anderson, J.E., Yotov, Y. (2012): Gold Standard Gravity. National Bureau of Economic Research. https://doi.org/10.3386/w17835. Atkin, M. (1995): The international grain trade. Woodhead Publishing Ltd., Cambridge. Besedeš, T., Moreno-Cruz, J., Nitsch, V. (2023): Trade integration and the fragility of trade relationships: Theory and empirics. http://pseweb.eu/ydepot/seance/515341_besedestrade agreements and duration.pdf. Besedeš, T., Prusa, T.J. (2006a): Ins, outs, and the duration of trade. Canadian Journal of Economics/Revue Canadienne d’économique 39 (1): 266-295. https://doi.org/10.1111/j.0008-4085.2006.00347.x. Besedeš, T., Prusa, T.J. (2006b): Product differentiation and duration of US import trade. Journal of International Economics 70 (2): 339-358. https://doi.org/10.1016/j.jinteco.2005.12.005. Besedeš, T., Prusa, T.J. (2011): The role of extensive and intensive margins and export growth. Journal of Development Economics 96 (2): 371-379. https://doi.org/10.1016/j.jdeveco.2010.08.013. Brenton, P., Chemutai, V., Pangestu, M. (2022): Trade and food security in a climate change‐ impacted world. Agricultural Economics 53 (4): 580-591. https://doi.org/10.1111/agec.12727. Carson, G.R., Edwards, N.M. (2009): Criteria of Wheat and Flour Quality. In: Khan, K., Shewry, P.R. (Eds.): WHEAT: Chemistry and Technology. 4th ed.: 97-118. AACC International, Inc., St. Paul, MN, USA. https://doi.org/10.1094/9781891127557.004. Córcoles, D., Díaz-Mora, C., Gandoy, R. (2015): Export Survival in Global Production Chains. The World Economy 38 (10): 1526-1554. https://doi.org/10.1111/TWEC.12249. Cox, D.R. (1972): Regression Models and Life-Tables. Journal of the Royal Statistical Society: Series B (Methodological) 34 (2): 187-202. https://doi.org/10.1111/J.25176161.1972.TB00899.X. Engemann, H., Jafari, Y., Heckelei, T. (2022): Institutional quality and the duration of agri‐food trade flows. Journal of Agricultural Economics. https://doi.org/10.1111/1477-9552.12491. FAO (1993): The State of Food and Agriculture 1993 (1st ed.). The Food and Agriculture Organization (FAO), Rome. Fellmann, T., Hélaine, S., Nekhay, O. (2014): Harvest failures, temporary export restrictions and global food security: the example of limited grain exports from Russia, Ukraine and Kazakhstan. Food Security 6 (5): 727-742. https://doi.org/10.1007/s12571-014-0372-2. Fradgley, N.S., Gardner, K.A., Kerton, M., Swarbreck, S.M., Bentley, A.R. (2023): Balancing quality with quantity: A case study of UK bread wheat. Plants, People, Planet. https://doi.org/10.1002/PPP3.10462. 21 Jaghdani et al. | Ger J Agr Econ 73 (2024), No. 3 Gafarova, G., Perekhozhuk, O., Glauben, T. (2015): Price discrimination and pricing to market behavior of Black Sea region wheat exporters. Journal of Agricultural and Applied Economics 47 (3): 287-316. https://doi.org/10.1017/aae.2015.16. Glauben, T., Duric, I. (2024): BRICS: World Heavyweight in Agricultural Trade. Intereconomics 59 (3): 160-166. https://doi.org/10.2478/IE-2024-0033. Glauben, T., Herzfeld, T., Rozelle, S., Wang, X. (2012): Persistent Poverty in rural China: Where, Why, and How to Escape? World Development 40 (4): 784-795. https://doi.org/10.1016/j.worlddev.2011.09.023. Glauben, T., Svanidze, M. (2023): Globaler Agrarhandel: robustes Sicherheitsnetz zur Reduktion von Hungerrisiken in Krisenzeiten. Wirtschaftsdienst 103 (7): 491-497. https://doi.org/10.2478/wd-2023-0139. Glauben, T., Svanidze, M., Götz, L., Prehn, S., Jaghdani, T.J., Đurić, I., Kuhn, L. (2022): The War in Ukraine, Agricultural Trade and Risks to Global Food Security. Intereconomics 57 (3): 157-163. https://doi.org/10.1007/s10272-022-1052-7. González-Esteban, Á.L. (2018): Patterns of world wheat trade, 1945-2010: The long hangover from the second food regime. Journal of Agrarian Change 18 (1): 87-111. https://doi.org/10.1111/joac.12219. Goodwin, B.K., Grennes, T.J. (1998): Tsarist Russia and the World Wheat Market. Explorations in Economic History 35 (4): 405-430. https://doi.org/10.1006/EXEH.1998.0706. Götz, L., Goychuk, K., Glauben, T., Meyers, W.H. (2013): Export Restrictions and Market Uncertainty: Evidence from the Analysis of Price Volatility in the Ukrainian Wheat Market. Issue 150308. Washington, DC. 2013 AAEA Annual Meeting, August 4-6, 2013, Washington, D.C. 10.22004/ag.econ.150308. Götz, L., Svanidze, M. (2023): Getreidehandel und Exportbeschränkungen während des Ukrainekriegs - Wirtschaftsdienst. Wirtschaftsdienst 103 (13): 37-41. https://doi.org/10.2478/wd-2023-0065. Gullstrand, J., Persson, M. (2015): How to combine high sunk costs of exporting and low export survival. Review of World Economics 151 (1): 23-51. https://doi.org/10.1007/s10290-0140204-7. Gutiérrez-Moya, E., Adenso-Díaz, B., Lozano, S. (2021): Analysis and vulnerability of the international wheat trade network. Food Security 13 (1): 113-128. https://doi.org/10.1007/s12571-020-01117-9. Heigermoser, M., Jaghdani, T.J., Götz, L. (2022): Russia’s Agri-Food Trade with the Middle East and North Africa. In: Wegren, S.K., Nilssen, F. (Eds.): Russia’s Role in the Contemporary International Agri-Food Trade System. 1st ed., Vol. 1: 253-277: Palgrave Macmillan, Cham. https://doi.org/10.1007/978-3-030-77451-6_10. Hess, W., Persson, M. (2011): Exploring the duration of EU imports. Review of World Economics 147 (4): 665-692. https://doi.org/10.1007/s10290-011-0106-x. Hess, W., Persson, M. (2012): The duration of trade revisited: Continuous-time versus discrete-time hazards11. Empirical Economics 43 (3): 1083-1107. https://doi.org/10.1007/s00181-011-0518-4. Imamverdiyev, N., Anders, S., Glauben, T., Perekhozhuk, O., Prehn, S. (2015): Determinants of Trade Duration of Kazakhstan’s Wheat Exports. International Association of Agricultural Economists (IAAE), August 9-14, Milan, Italy. https://doi.org/10.22004/AG.ECON.211788. Jaghdani, T.J., Johansen, U., Thakur, M., Đurić, I. (2020): Assessment of persistence of business/ trade relationships along the selected food chains of different European countries and sectors (Deliverable D5.3): https://ec.europa.eu/research/participants/documents/downloadPublic?documentIds=080166e5e84f1fcd&appId=PPGMS. Jaghdani, T.J., Götz, L., Ghodsi, M. (2024a): Wheat trade between Russia and Iran: a discontinuous trend. Russian Analytical Digest (RAD) (318): 19-26. https://doi.org/10.3929/ethz-b-000701771. 22 Jaghdani et al. | Ger J Agr Econ 73 (2024), No. 3 Jaghdani, T.J., Johansen, U., Thakur, M., Glauben, T. (2024b): Salmon trade duration: the application of firm-level trade transaction data from the Norwegian salmon industry. Agribusiness 40 (2): 325-348. https://doi.org/10.1002/agr.21900. Johnson, L.L. (2018): An Introduction to Survival Analysis. In: Gallin, J.I., Ognibene, F.P., Johnson, L.L. (Eds.): Principles and Practice of Clinical Research. 4th ed.: 373-381. Academic Press, London, UK. https://doi.org/10.1016/B978-0-12-849905-4.00026-5. Kaplan, E.L., Meier, P. (1958): Nonparametric Estimation from Incomplete Observations. Journal of the American Statistical Association 53 (282): 457-481. https://doi.org/10.1080/01621459.1958.10501452. Ketokivi, M., Mahoney, J.T. (2020): Transaction Cost Economics As a Theory of Supply Chain Efficiency. Production and Operations Management 29 (4): 1011-1031. https://doi.org/10.1111/POMS.13148. Kinnunen, P., Guillaume, J.H.A., Taka, M., D’Odorico, P., Siebert, S., Puma, M.J., Jalava, M., Kummu, M. (2020): Local food crop production can fulfil demand for less than one-third of the population. Nature Food 1( 4): 229-237. https://doi.org/10.1038/s43016-020-0060-7. Lee, T.M., Chi, P.Y., Chang, K.I. (2020): Duration and determinants of fishery trade patterns: Evidence from OECD countries. Marine Policy 118: 103977. https://doi.org/10.1016/j.marpol.2020.103977. Liefert, W.M., Liefert, O. (2015): The rise of the former soviet union region as a major grain exporter. In: Transition to Agricultural Market Economies: The Future of Kazakhstan, Russia and Ukraine: 27-38.: CABI International, Oxfordshire, UK. https://doi.org/10.1079/9781780645353.0027. Luo, Y., Bano, S. (2020): Modelling New Zealand dairy products: evidence on export survival and duration. Australian Journal of Agricultural and Resource Economics 64 (3): 605-631. https://doi.org/10.1111/1467-8489.12372. Marinciu, C.M., Şerban, G., Săulescu, N. (2022): Performance of some Romanian winter wheat cultivars under organic agriculture conditions. Annals of the University of Craiova 52 (1): 247-254. https://doi.org/10.52846/AAMC.V52I1.1340. Melitz, M.J. (2003): The impact of trade on intra-industry reallocations and aggregate industry productivity. Econometrica 71 (6): 1695-1725. https://doi.org/10.1111/1468-0262.00467. Miskelly, D.M. (2019): Analytical Testing to Support Australian Wheat Export Quality. Cereal Foods World 64 (1). https://doi.org/10.1094/CFW-64-1-0002. O’Rourke, K.H. (1997): The European grain invasion, 1870-1913. Journal of Economic History 57 (4): 775-801. https://doi.org/10.1017/S0022050700019537. Obashi, A. (2010): Stability of production networks in East Asia: Duration and survival of trade. Japan and the World Economy 22 (1): 21-30. https://doi.org/10.1016/J.JAPWOR.2009.06.002. Peterson, E.B., Grant, J.H., Rudi-Polloshka, J. (2018): Survival of the Fittest: Export Duration and Failure into United States Fresh Fruit and Vegetable Markets. American Journal of Agricultural Economics 100 (1): 23-45. https://doi.org/10.1093/ajae/aax043. Rickman, G.E. (1980): The Grain Trade under the Roman Empire. Memoirs of the American Academy in Rome 36: 261. https://doi.org/10.2307/4238709. Singer, J.D., Willett, J.B. (1993): It’s About Time: Using Discrete-Time Survival Analysis to Study Duration and the Timing of Events. Journal of Educational Statistics 18 (2) 155-195. https://doi.org/10.3102/10769986018002155. Svanidze, M., Götz, L., Djuric, I., Glauben, T. (2019): Food security and the functioning of wheat markets in Eurasia: a comparative price transmission analysis for the countries of Central Asia and the South Caucasus. Food Security 11 (3): 733-752. https://doi.org/10.1007/s12571-019-00933-y. Svanidze, M., Götz, L., Serebrennikov, D. (2022): The influence of Russia’s 2010/2011 wheat export ban on spatial market integration and transaction costs of grain markets. Applied Economic Perspectives and Policy 44 (2): 1083-1099. https://doi.org/10.1002/aepp.13168. 23 Jaghdani et al. | Ger J Agr Econ 73 (2024), No. 3 Tebaldi, E., Gobjila, A. (2018): Romania - Systematic Country Diagnostic : background note - agriculture. https://documents.worldbank.org/en/publication/documents-reports/documentdetail/698251530897736576/romania-systematic-country-diagnostic-backgroundnote-agriculture. Tutz, G., Schmid, M. (2016): Modeling Discrete Time-to-Event Data. Springer International Publishing, Basel, Switzerland. https://doi.org/10.1007/978-3-319-28158-2. Wang, P., Tran, N., Wilson, N.L.W., Chan, C.Y., Dao, D. (2019): An Analysis of Seafood Trade Duration: The Case of ASEAN. Marine Resource Economics 3 (1): 59-76. https://doi.org/10.1086/700599. Wrigley, C.W. (2009): Wheat: A Unique Grain for the World. In: Khan, K., Shewry, P.R. (Eds.): WHEAT: Chemistry and Technology. 4th ed.: 1-17. AACC International, Inc., St. Paul, MN, USA. https://doi.org/10.1094/9781891127557.001. Yugay, S., Götz, L., Svanidze, M. (2024): Impact of the Ruble exchange rate regime and Russia’s war in Ukraine on wheat prices in Russia. Agricultural Economics 55 (2): 384-411. https://doi.org/10.1111/AGEC.12822. Contact Author Tinoush Jamali Jaghdani, PhD Department of Agricultural Markets Leibniz Institute of Agricultural Development in Transition Economies (IAMO) Theodor-Lieser-Str. 2, 06120 Halle, Germany e-mail: [email protected] 24