Asymmetric Examination of Exchange Rate Dynamics: Are There Any Divergent Effects on Agricultural Growth

Abstract

This study aims to investigate the asymmetric effects of exchange rates on the growth of the agricultural, industrial, and service sectors in Türkiye, utilizing the Nonlinear Autoregressive Distributed Lag (NARDL) method for time series analysis. The analysis covers the period from 2005Q4 to 2023Q4. The results indicate significant differences in sectoral effects. Specifically, increases in exchange rates do not have a substantial long-term impact on the agricultural sector, while decreases in exchange rates tend to enhance agricultural output over the long term. In contrast, for the industrial and service sectors, increases in exchange rates result in a decline in production, whereas decreases in exchange rates lead to an increase in production. The findings also reveal that the service sector is the most affected by exchange rate fluctuations in Türkiye, as indicated by the coefficients. These results demonstrate that Türkiye, as an emerging market and developing country, experiences the impacts of exchange rate fluctuations on its agricultural sector differently than on its industrial and service sectors in the long term. The NARDL findings underscore the necessity for regulatory institutions to manage exchange rates more sensitively, taking into account the relative importance of each sector in the Turkish economy.

JEL Codes: F31, O24, Q17.

Plain language summary

Do Changes in Exchange Rates Affect Agricultural Growth Differently in Some Cases?

This study looks at how changes in exchange rates affect different parts of Türkiye’s economy—agriculture, industry, and services—between 2005 and 2023. Using a special method called NARDL, the researchers found that the effects are not the same for every sector. When exchange rates go down, agricultural production tends to increase over time. But when exchange rates go up, there is no strong effect on agriculture in the long run. For industry and services, it’s different: A rise in exchange rates leads to less production. A fall in exchange rates leads to more production. The service sector is the most sensitive to these changes. In short, exchange rate changes affect each sector in different ways. Policymakers in Türkiye should consider these differences when making economic decisions.

Keywords

exchange rates sectoral economic growth Mundell-Fleming model trade Türkiye

Introduction

Theoretically, it has been established that currency devaluation has a stimulating effect on economic activity. It is anticipated that the first spike in the price of foreign goods relative to domestic goods will lead to an excess demand for domestic goods. Moreover, the influence of currency depreciation on the total output levels of an economy has been a subject of debate among economists with diverse perspectives, primarily due to the existing uncertainty.

In its essence, currency devaluation fundamentally enhances exports and encourages a transition from imports to domestic goods, thereby boosting the demand for domestic products. In addition, countries may choose to devalue their currency to achieve competitive trade benefits and stimulate the growth of export-oriented sectors (Dornbusch & Leslie, 1988). This can ultimately propel a country toward a developmental path characterized by increased potential for long-term economic growth and employment (Dornbusch et al., 1994, 1995). As a result, with the aim of benefiting from expansionary aspects of devaluation, widespread devaluation can be seen following the collapse of the Bretton Woods system and through the adjustment policies of the International Monetary Fund (hereafter IMF) in countries such as Greece, Egypt, Argentina, and Türkiye in the years 2010, 2016, 2002, and 1994, respectively.

Although devaluation may originally benefit overall economic performance by boosting aggregate demand, its adverse impacts on the supply side result in an ambiguous outcome in terms of its influence on domestic output. The uncertainty surrounding this issue has captured the attention of development economists, who argue that the initial positive impact of devaluation will be undermined through various channels, such as distributional effects, account deficits, and real money balance effects (Guitian, 1976; Hanson, 1983). This is because the initial effect will not be enough to counterbalance the subsequent increase in import prices, leading to a decline in output (Krugman & Taylor, 1978).

Indeed, for countries like Türkiye, which rely on imported inputs for production and exports, this problem signifies a significant unresolved issue. Furthermore, the Turkish lira was pegged to the British pound and French francs until the Bretton Woods agreement, during which the value of the lira was fixed at 2.8 liras per United States (hereafter U.S.) dollar. Nevertheless, as the Bretton Woods system collapsed and floating exchange rates became increasingly prevalent, the lira experienced significant devaluation in 1960 and 1973. The value of the lira dropped from 2.8 to 9 liras per U.S. dollar in 1960 and further declined to 13 liras per U.S. dollar in 1973. The Turkish economy experienced a significant shift after the political and socio-economic crisis of 1994 to 1999, which ultimately led to the 2001 financial crisis. This crisis not only resulted in distortions of key macro-economic indicators such as output, exchange rate, and unemployment in Türkiye but also brought about the exchange rate to fall to 2.4 million per dollar. The Central Bank of the Republic of Türkiye (hereafter CBRT) has implemented necessary steps in response to the economic crisis, specifically targeting the monetary aspects of the markets, to maintain stability. Moreover, with the introduction of the lira reform, new lira replaced old Turkish lira at a rate of one new Turkish lira to 1 million old Turkish liras.

Following the crises of 2001, 2008, and the most recent COVID-19 epidemic, the CBRT has exerted considerable influence over the exchange rate market. Through open market operations, the CBRT attempted to prevent a sharp devaluation of the Turkish lira, albeit at the cost of depleting the central bank’s exchange rate reserves. Recently, there has been a surge in public statements made by government officials pertaining to the adoption of the “Beijing Consensus” (or Chinese Economic Model) as a means to enhance the competitiveness of the lira, with the ultimate aim of boosting the country’s exports and employment rates. The strong growth rates displayed by many countries with undervalued currencies in the early 2000s (Ridhwan et al., 2024) may have contributed to the adoption of such an approach. However, these tendencies raised inquiries regarding the pass-through of the exchange rate into industrial output and inflation levels in Türkiye. The concerns mainly revolved around the country’s reliance on intermediate goods, current account deficits, and foreign debt, which were already extensively debated.

Although economic theories are grounded in numerical data, countries’ economic systems are fundamentally supported by the social institutions formed by individuals. Consequently, countries with similar economic characteristics may not achieve identical outcomes by adopting comparable policies. Furthermore, the economic effects stemming from the political stability fostered by social structures should not be underestimated. In this context, the discussion surrounding the appreciation of the Turkish lira, often regarded as an undervalued currency, has become increasingly prevalent. Analyzing these effects from various scientific perspectives presents substantial research opportunities. The rationale behind this study is grounded in the previously mentioned reasons, with the aim of examining Türkiye within the context of its internal balances. A model that elucidates the effects of currency fluctuations on sectoral production has significant potential to provide valuable insights for a developing country like Türkiye and other emerging markets. Furthermore, the incorporation of variables such as exports, oil prices, and money supply, factors that substantially influence production across various sectors in Türkiye, offers a comprehensive perspective, enabling the research process to be conducted through a distinctive model.

In this context, the research questions are as follows:

Research Question 1: Do increases in the exchange rate have varying effects on sectoral growth in Türkiye?

Research Question 2: Do decreases in the exchange rate have varying effects on sectoral growth in Türkiye?

Research Question 3: What other variables influence sectoral growth in Türkiye within a model that incorporates the exchange rate?

Answering these research questions will significantly contribute to the development of the model, drawing on the theoretical foundation necessary for the research as well as previous similar studies. Additionally, the existing gaps in the literature will provide a compelling rationale for conducting this study.

Theoretical Framework

Building on the conceptual ambiguity surrounding the macroeconomic effects of exchange rate fluctuations, this section develops the theoretical foundations of the study by outlining the key transmission mechanisms through which currency devaluation (depreciation) and revaluation (appreciation) may influence aggregate and sectoral output.

First of all, in the context of the Mundell-Fleming model, a devaluation of the domestic currency results in increased output levels, potentially accompanied by higher interest rates depending on the reaction of monetary policy (Wang, 2020). In an open economy, this shift toward production can also be explained by the expectation of earning valuable foreign currency through exports. Bahmani et al. (2013) explain this situation through the Marshall-Lerner condition. Conversely, in the case of revaluation (appreciation), the opposite effect will be observed (Levy-Yeyati et al., 2013). In this scenario, the belief that export earnings in the local currency will decrease diminishes the incentive to engage in international risks. This perception may lead businesses to concentrate more on the safer domestic market rather than pursuing additional production for exports. As a result, companies might scale back or abandon plans for increased production, as the potential returns from exports may not warrant the associated risks.

Furthermore, recent studies have revisited this framework in light of contemporary challenges, including financial globalization and digital currencies. For instance, Enajero (Enajero, 2021) and Tenderere (Tenderere & Mishi, 2025) demonstrate that the traditional Mundell–Fleming predictions may be altered when economies experience volatile capital mobility and the presence of alternative monetary assets, notably cryptocurrencies. Similarly, Leightner (2024) and Magacho et al. (2022) empirically confirm that the model’s predictions differ across fixed and flexible exchange-rate regimes in small open economies characterized by advanced technology utilization, such as Australia and South Korea.

Additionally, the “J-Curve” theory elucidates that in the short run, currency devaluation will deteriorate a nation’s trade balance, whereas in the long run, after pre-existing trade agreements are alleviated, trade balance will progressively improve (Bremmer, 2007). The short-term deficits described by the J-curve approach, which are offset by long-term recovery, can also be attributed to production aimed at generating foreign currency earnings (Geldner, 2025). The existence of significant exchange-rate volatility not only hinders the recovery phase, undermining long-term advancements in sectors such as agriculture (Trofimov, 2024), but also induces disparities in readjustment across bilateral trading partners (Bosupeng et al., 2024; Montes et al., 2024).

In practice, the real-world effects of a domestic currency revaluation or devaluation on output may not always align with the theoretical predictions. This is a common occurrence in the field of economics where ambiguity often plays a significant role. Nevertheless, the theory’s limitations arise from its failure to consider the sensitiveness of domestic price levels to fluctuations in exchange rates (specifically devaluation in this context). Such volatility might possibly decrease overall demand and translate into substantial economic losses, manifesting as recurrent declines in industrial output and employment levels (Alejandro, 1963; Cooper, 1992). Contrary to Krugman and Taylor’s (1978) assertion that a reduction in demand as a result of devaluation lowers prices, adverse supply shocks completely negate the gains from devaluation and put upward pressure on price levels, depending on the response of nominal wages (Lizondo & Montiel, 2006; Van Wijnbergen, 1986).

Moreover, recent empirical research supports that the extent of exchange rate pass-through to domestic price levels is neither constant nor uniform across various economies (An et al., 2021; Beirne et al., 2024; Bhat et al., 2025). Consequently, the magnitude of pass-through evolves and fluctuates according to monetary policy credibility, inflation expectations, and structural characteristics of production (Colak et al., 2024; Lama et al., 2025).

Further, when the regional context is taken into consideration, the implementation of a strategy of competitive devaluation in an emerging economy might result in negative regional spillovers due to the “neighborhood effect” (Calvo & Reinhart, 1996; Rose & Glick, 1999). In a globally interconnected economy, when a country’s export competitor experiences devaluation, it can deteriorate the trade balance of the partner country. This can potentially trigger a speculative attack on the partner country’s currency, causing an excessive fluctuation of the domestic currency (S. Kim & Roubini, 2000). Should a country’s trader competitor resort to the “tit for tat” strategy as a policy response, the attempt to “beggar thy neighbour” (Smith, 2007, p. 377) via competitive devaluation may not necessarily yield a desired outcome.

The contemporary studies on foreign exchange spillovers reveal considerable interconnectedness among major and emerging-market currencies, implying that depreciation episodes in one economy can propagate volatility to others (Ha et al., 2025; Huynh et al., 2023). Additionally, the global monetary environment intensifies these effects, as expansionary policies in advanced economies or crisis-related depreciations, such as those observed during the COVID-19 period, can produce implicit pressures on exchange rates and capital flows in developing regions (Bhatia & Tuteja, 2024; Dées & Galesi, 2021).

In their study on the relationship between exchange rates and financial fragility, Eichengreen and Hausmann (1999) highlight the inability of developing economies to borrow (domestic or abroad) using their domestic currencies, regardless of the duration of the loan. This lack of ability is commonly referred to as the concept of “original sin,” leading to probable financial fragility and a currency mismatch for domestic investments. In a recent report by the United Nations Trade and Development [UNCTAD] (2023) organization, it was revealed that approximately 54.7% of short- and long-term external debt in developing economies is denominated in US dollars. This represents the highest level ever recorded since the outbreak of the COVID-19 pandemic, underscoring the significant impact of currency devaluation on the “balance sheets” of the banks in developing economies. The repercussions of the “balance sheet” effect will counteract the anticipated advantages of currency devaluation, and substantial unhedged exposures to short-term foreign currency debt will arise. This will culminate in widespread domestic credit rationing, an elevated amount of non-performing loans (NPLs), and a credit crunch for industrial producers (Hutchison & Glick, 2000).

Moreover, the aforementioned industrial producers with elevated foreign-currency exposure encounter substantial reductions in investment subsequent to depreciations, and banking systems with dollarized liabilities face increased non-performing loans and tighter lending standards (Bruno & Shin, 2017; Hofmann et al., 2022; Seraj & Coskuner, 2021). This mismatch between currency composition of assets and liabilities constrains monetary policy autonomy and heightens systemic risk, making exchange-rate flexibility less effective as a stabilization tool (Chakrabarti & Sen, 2023; International Monetary Fund [IMF], 2022).

In addition, Fernandez-Arias and Hausmann (2001) characterize short-term debt as unfavorable, drawing an analogy to “bad cholesterol,” owing to the fact that riskier countries attract less capital but receive a higher proportion of it in the form of speculative foreign direct investment (FDI). The susceptibility of developing economies to speculative attacks, exacerbated by structural problems, contributes to concerns regarding the potential reversal of the early advantages of moderate devaluation. This reverse effect can serve as a trigger for major capital outflows and ultimately bring about a complete economic collapse due to credibility concerns, possibly sparking a large-scale speculative attack. Krugman (1998) claims that the response of markets to devaluation strategies differs across developing and developed nations on account of problems surrounding “credibility.” These problems can cause a subsequent decline in the currency’s value, generating a vicious cycle of crisis-induced devaluation, scarcity of available funds, and financial bankruptcy for organizations and enterprises. According to Shang-Jin and Wu (2002), contractionary devaluation in developing countries is a result of “poor” governance, which is closely tied to the aforementioned credibility aspect.

Recent empirical research reinforces the potency of these credibility and debt composition channels. It is documented that emerging economies with larger shares of short-term foreign-currency debt are particularly prone to experience abrupt capital outflows and currency crises during periods of global stress (Bernoth & Herwartz, 2021; Dalgic, 2024; S. Liu et al., 2024; Zhang et al., 2024). Several analyses further show that investor skepticism pertaining to policy credibility amplifies speculative pressures, particularly when short-term debt is rolled over through foreign-denominated instruments (Chowdhury & Sundaram, 2023; Dhar, 2021; Inoguchi, 2025).

The theories presented indicate that the impact of exchange rate fluctuations on real output dynamics is significantly context-dependent, contingent upon trade openness, financial depth, and policy credibility. These interactions are particularly pertinent for emerging economies that rely extensively on imported inputs and foreign-currency borrowing, conditions that characterize Türkiye’s economic structure. The following section, therefore, surveys the related literature and identifies the empirical gaps that motivate this study.

Literature Review

Several studies utilizing Vector Autoregressive (VAR) frameworks document that exchange-rate depreciations affect output dynamics through trade, price, and financial channels, yet the sign and magnitude are context-dependent. In emerging markets, flexible exchange rates can either cushion or amplify shocks depending on external vulnerabilities and policy settings (Babubudjnauth & Seetanah, 2021; Fisera, 2024; Montane et al., 2021). Models that include control variables such as inflation, employment, trade, capital flows, and money supply observe heterogeneous impulse responses across countries and time periods, underscoring that structural features shape the transmission from exchange rates to Gross Domestic Product (GDP; Ahmed et al., 2023; Souza & de Mattos, 2022). Evidence from small open economies and commodity-dependent regions further shows that the interaction between exchange-rate movements and price formation (terms-of-trade and pass-through) is crucial for short-run output responses (Chen & Ward, 2025; Manopimoke et al., 2024). While the samples and variables examined vary, the findings also demonstrate variability. This further underscores the necessity of analyzing exchange rate fluctuations from different perspectives, as outlined in the theoretical framework. Similar situations are also observed in other studies within literature.

Studies utilizing the Vector Error Correction Model (VECM) have underscored the significant impact of exchange rates on economic growth and sectoral production in various research studies that differentiate between developed and developing countries (Kandil, 2019; Moses et al., 2020; Zengeni & Mahonye, 2019) . Notably, contrary to earlier studies, recent investigations incorporate the oil price variable into their models (Ben Salem et al., 2024; Chatziantoniou et al., 2023). Further, evidence demonstrates that in countries that have substantial export capacity, depreciation may enhance growth, whereas in economies depending on imported inputs or foreign-currency debt, it can lead to contractionary consequences owing to rising production costs and inflation pass-through (Bigerna, 2023; Ha et al., 2020). Overall, these findings justify a country-specific empirical assessment that explicitly models trade and cost transmission together.

Research conducted using the Generalized Method of Moments (GMM) has examined the relationships between exchange rate fluctuations, economic growth, and production across various models focusing on high-, middle-, and low-income countries (Demir & Razmi, 2022; Ebrahimi et al., 2020; Ferrara & Yapi, 2022; Gabriel et al., 2020). In addition, Bussière et al. (2020) report that exchange rate depreciation dampens growth in financially dollarized countries due to increased balance-sheet pressure. In the context of commodity exposure, Mous and Maznan (2024) find that exchange rate volatility diminishes industrial production during periods of elevated energy prices.

When examining various studies that have employed different panel regression models, researchers have identified the impact of exchange rate fluctuations on sectoral production in several contexts. For instance, Hardy (2023) documents that exchange rate volatility significantly weakens the credit capacity of firms in emerging economies by deteriorating their balance sheets, ultimately constraining investment and productivity. Caballero (2021) argue that exchange rate shocks are transmitted from firms to the broader macroeconomy through credit constraints, amplifying systemic risk in financially dollarized sectors. Evidence from Salomao and Varela (2020) further suggests that sectors with high import dependence exhibit stronger output contractions following currency depreciations, confirming that exchange rate dynamics are central to the transmission of external shocks into domestic production. Ramoni-Perazzi and Romero (2022) demonstrate that exchange rate depreciation raises input costs and inflationary pressures, disproportionately affecting manufacturing firms that are exposed to global value chains. These findings collectively indicate that the exchange rate is not merely a nominal variable but a key determinant of sectoral performance through its influence on cost structures and financial stability.

Studies examining the relationship between currency devaluation or exchange rates and economic and sectoral growth using Autoregressive Distributed Lag (ARDL) methods have included samples from various regions. For instance, Bahmani-Oskooee and Arize (2020b) provide evidence that exchange rate depreciation in African economies brings about asymmetric growth responses, emphasizing the relevance of structural vulnerabilities and the degree of import dependence. More recent studies, such as Nusair (2022) and Nosheen et al. (2023), show that exchange rate misalignments create persistent output fluctuations in Asian countries by affecting trade balances and capital flows. Ali et al. (2022) reveal that while depreciation pressures in Pakistan could contribute to short-term benefits in net exports, they ultimately pose challenges for long-term growth as production costs soar. The instability of the exchange rate, according to Sharaf and Shahen (2023), undermines the confidence of investors and, thereby, erodes fiscal space in Egypt.

In addition to studies conducted on various samples using diverse methodologies, it is also beneficial to examine research specifically focused on Türkiye, which is the focal point of this study. Özdemir et al. (2022) demonstrate that exchange rate depreciation does not impose a constraining effect on aggregate production in Türkiye, suggesting that depreciation may bolster tradable sectors via competitiveness gains. Further, Halicioglu et al. (2018), Bahmani-Oskooee and Durmaz (2020a), and Karamelikli and Bahmani-Oskooee (2021) document that exchange rate depreciation enhances industrial activity through augmenting net exports and initiating expansionary monetary effects. However, more recent sectoral studies, such as Ari et al. (2019) and Khalid et al. (2025), reveal that depreciation yields heterogeneous outcomes across industries, benefiting export-oriented manufacturing while adversely affecting sectors with high import dependency due to increased production costs. Complementary evidence provided by Demirkılıç (2021) and Tong and Durmaz (2024) indicates that in periods characterized by heightened financial fragility, currency depreciation reduces output by tightening external financing conditions and exacerbating balance sheet effects.

Literature Gap

The literature review indicates that studies conducted in various countries predominantly focus on measurements based on total production and GDP output. However, it is essential not to overlook the critical sectoral distinctions that are significant for developing countries, despite the fact that many studies concentrate on these nations. Additionally, research specifically conducted on Türkiye similarly tends to focus solely on total or industrial production. Since the sectoral distinctions made in industrial production can also be interpreted through a single sector at the macro level, this literature review reveals a significant gap. Moreover, examining the sectoral effects collectively, not only in terms of addressing different sectors but also regarding the variation of explanatory variables, would provide a substantial contribution to the literature. Furthermore, incorporating both the symmetric and asymmetric effects of the exchange rate into the model allows the study to address these important gaps in the existing literature. Indeed, the research model is supported by the literature through the inclusion of sectoral growth, exchange rates, money supply, exports, and oil prices, while it distinguishes itself by separately measuring and presenting the agricultural, industrial, and service sectors.

Türkiye in Context and Motivation

Türkiye exemplifies a distinctive macro-financial case among emerging economies due to its simultaneous exposure to currency volatility, structural external imbalances, large-scale migration pressures, and unprecedented natural disasters.

The COVID-19 epidemic represented a turning point for the Turkish economy, as the interplay of dwindling central bank reserves and expansionary credit-driven policies precipitated a succession of significant currency depreciations. Furthermore, the massive earthquake that struck 11 cities during the post-pandemic recovery phase caused damage amounting to hundreds of billions of dollars. These extraordinary shocks not only amplified exchange rate volatility but also intensified inflationary pressures and interrupted production networks.

The regional context for Türkiye, like many other aspects, is increasingly complex. In addition to competition, the wars and conflicts in neighboring countries are driving migration to Türkiye, significantly disrupting the balance of demand. Refugees and undocumented migrants from countries such as Syria, Ukraine, Russia, Afghanistan, and others are profoundly affecting the supply-demand equilibrium, which finally affects the monetary issues. This persistent migration-driven demand shock distinguishes Türkiye from other emerging economies, where exchange rate dynamics are typically examined under more stable demographic circumstances.

Türkiye’s macroeconomic fragility is further compounded by its extensive reliance on external borrowing. As of 2024, the country’s total external debt exceeds 550 billion dollars, and a substantial share of this debt is denominated in foreign currency (Central Bank of the Republic of Türkiye [CBRT], 2024).23 This structural attribute renders the Turkish economy particularly susceptible to exchange rate fluctuations, as currency depreciations promptly result in increased debt servicing costs, reduced credit accessibility, and aggravated financial instability. In contrast to studies assessing exchange rate-growth dynamics in countries with limited external debt exposure, Türkiye constitutes a unique instance where devaluation may inhibit rather than augment productive capacity.

The efficacy of the enacted “competitive Turkish lira” policies in Türkiye is disputable given multiple considerations. These obstacles encompass the issues inherent to being a developing country with an open economy, dependence on imported raw materials, concerns over credibility, and the influence of geographical location. Moreover, Türkiye’s reliance on imported intermediate goods for agricultural, industrial, and service production profoundly alters the pass-through of devaluation. Instead of fostering competitiveness, depreciation raises input costs, constrains production, and fuels inflationary spirals, hence challenging the assumptions of traditional growth-enhancing devaluation theories.

Having said these, Türkiye offers a compelling and analytically distinct context in which the effects of exchange rate depreciation cannot be evaluated solely through conventional theoretical frameworks alone. The confluence of external debt exposure, migration-induced demand shocks, natural disasters, and import-dependent production structures renders Türkiye an exceptional case for reassessing the effectiveness of devaluation policies. This study aims to empirically examine whether currency depreciation in Türkiye has adversely affected domestic output in the agricultural, industrial, and service sectors, which are notably susceptible owing to their considerable reliance on imported intermediate inputs and exchange rate fluctuations.

Methodology

The empirical model employed to examine the effects of exchange rate fluctuations on sectoral production within the Turkish economy is delineated in Equation 1.

lngd p_{t} = β_{0} + β_{1} lnex c_{t} + β_{2} lnex p_{t} + β_{3} lnoi l_{t} + u_{it}

(1)

In this model, the dependent variable is sectoral production (lngdp). Separate estimations have been performed for production (current, $) in the agriculture, forestry, and fishing sectors, the industrial sector, and the services sector. The primary independent variable in the model is the exchange rate (lnexc), while the control variables encompass sectoral exports (lnexp, and oil prices (lnoil). The exchange rate utilized in the model is the U.S. dollar exchange rate. An increase in the dollar rate (i.e., depreciation of the local currency) may enhance national income by stimulating the country’s exports; however, it may also result in a reduction in production within sectors that heavily depend on imported inputs due to the escalated costs of imports. Among the control variables, lnexp reflects the export levels of each sector, which is essential for the continuity of sectoral production in a globalized economic context. The final control variable, lnoil, is included in the model to account for energy costs, which are a critical input in production. Given that increases in oil prices elevate production costs, they are anticipated to exert a negative impact on production.

As highlighted in the existing literature, while numerous studies have investigated the impact of exchange rates on production within the Turkish economy, the potential nonlinear effects of exchange rates on production levels have been largely neglected. Market distortions, including price and wage rigidities, asymmetric information, and other similar factors, may lead to differential responses of economic variables to positive and negative shocks in another variable. Specifically, fluctuations in exchange rates may exert disparate effects on production levels. This study seeks to investigate the influence of exchange rates on sectoral production through the application of the Nonlinear Autoregressive Distributed Lag (NARDL) methodology, as established by Shin et al. (2014). This approach represents an advanced iteration of the Autoregressive Distributed Lag (ARDL) model introduced by Pesaran et al. (2001). The ARDL methodology presents several advantages; notably, it does not necessitate that the time series be stationary at the same order, allowing for the inclusion of series that are I(0) or I(1), provided they are not I(2). Additionally, this approach facilitates efficient and consistent estimation results, even when applied to small and limited sample sizes. These advantages have contributed to the widespread adoption of the ARDL approach in the exploration of relationships between variables.

In the NARDL approach, the effects of positive and negative shocks in the independent variable on the dependent variable are analyzed separately. The asymmetric long-run relationship between the variables is examined through the application of asymmetric cointegration regression, as presented in Equation 2. In this equation, $y_{t}$ represents the dependent variable, $u_{t}$ denotes the error term, and $β^{+}$ and $β^{-}$ signify the asymmetric long-term coefficients. Equation 3 defines and $x_{t}^{-}$ and $y_{t}^{-}$ as the positive and negative partial sums of the independent variable, respectively, while $x_{0}$ indicates the initial observation value of the independent variable. Furthermore, Equations 4 and 5 delineate the processes for the asymmetric positive and negative partial sums.

y_{t} = β^{+} x_{t}^{+} + β^{-} x_{t}^{-} + u_{t}

(2)

x_{t} = x_{0} + x_{t}^{+} - x_{t}^{-}

(3)

x_{t}^{+} = \sum_{i = 1}^{t} Δ x_{t}^{+} = \sum_{i = 1}^{t} \max (Δ x_{i}, 0)

(4)

x_{t}^{-} = \sum_{i = 1}^{t} Δ x_{t}^{-} = \sum_{i = 1}^{t} \max (Δ x_{i}, 0)

(5)

The model delineated by Equation 1 in this study is articulated in a nonlinear form in equation

\begin{matrix} Δ lngd p_{t} = α_{0} + α_{1} lngd p_{t - 1} + α_{2}^{+} {lnexc}_{t - 1}^{+} + α_{3}^{-} {lnexc}_{t - 1}^{-} \\ + α_{4} lnm 2_{t - 1} + α_{5} lnex p_{t - 1} + α_{6} lnoi l_{t - 1} \\ + \sum_{i = 1}^{p - 1} α_{7} Δ gd p_{t - i} + \sum_{i = 0}^{q} α_{8}^{+} Δ {lnexc}_{t - i}^{+} \\ + \sum_{i = 0}^{r} α_{9}^{-} Δ {lnexc}_{t - i}^{-} + \sum_{i = 0}^{s} α_{10} Δ lnm 2_{t - i} \\ + \sum_{i = 0}^{t} α_{11} Δ lnex p_{t - i} + \sum_{i = 0}^{v} α_{12} Δ lnoi l_{t - i} + ε_{t} \end{matrix}

(6)

In this context, the dependent variable is denoted as $Δ lngd p_{t}$ . The variables $lnex c^{+}$ and $lnex c^{-}$ signify the positive and negative shocks of the exchange rate on sectoral production, respectively. The parameters $α_{2}$ and $α_{3}$ represent the long-term coefficients associated with the positive and negative shocks of the exchange rate, while $α_{8}$ and $α_{9}$ correspond to the short-term coefficients for these shocks. p, q, r, s, t, and v indicate the optimal lag lengths, which have been determined in accordance with the Akaike Information Criterion (Shin et al., 2014).

The analysis period of this study encompasses the fourth quarter of 2005 through the fourth quarter of 2023. The selection of this analysis period was influenced by the availability of relevant data. The variables lngdp and lnexp, which have been adjusted for seasonal and calendar effects, were sourced from the Turkish Statistical Institute (TURKSTAT, 2024). The data for lnexc was obtained from the Central Bank of the Republic of Türkiye (CBRT), while the lnoil data, which reflects Brent oil prices, was retrieved from the St. Louis Federal Reserve database. There is a rationale behind choosing these variables. Several recent studies employing ARDL, VAR, and VECM frameworks have highlighted that oil prices are a key exogenous determinant influencing both exchange rates and sectoral output, as energy costs transmit into production and price dynamics (e.g., Ben Salem et al., 2024; Chatziantoniou et al., 2023; Mous & Maznan, 2024). These studies show that elevated energy prices intensify the contractionary effects of currency depreciation, especially in industrial and service sectors dependent on imported inputs. Similarly, exports are widely recognized as the primary channel through which exchange rate movements affect economic growth. Empirical evidence from both cross-country and Türkiye-specific analyses confirms this transmission mechanism: depreciation tends to stimulate export-oriented sectors while constraining those with high import dependence (e.g., Ari et al., 2019; Bahmani-Oskooee & Durmaz, 2020a; Karamelikli & Bahmani-Oskooee, 2021; Khalid et al., 2025; Özdemir et al., 2022). Accordingly, incorporating both oil price and export variables provides theoretical and empirical coherence with the established literature, ensuring that the models capture the trade and cost channels through which exchange rate dynamics influence sectoral growth.

Empirical Results

Table 1 presents the descriptive statistics of the data. The results unequivocally demonstrate the disparities in economic magnitude among sectors. The mean lngdp values indicate that during the period in question, the services sector (18.25) had the highest average production level, while agriculture (16.83) had the lowest. Further, an examination of the standard error values of the $lnex c^{+}$ and $lnex c^{-}$ variables shows that fluctuations in the positive direction were higher than those in the negative direction (1.01 and 0.16). The mean values of lnexp indicate an elevated volume in exports associated with services and industry, whereas the agricultural sector has shown relatively lower levels of export activity. Based on the standard error value of the lnoil, it can be noted that the volatility in this variable is more limited.

Table 1.

Descriptive Statistics.

Variable	Mean	Median	Max	Min	Std. dev.	Obs.
lngdp (Agriculture)	16.83	16.84	17.02	16.58	0.13	73
lngdp (Industry)	18.09	18.09	18.51	17.55	0.27	73
lngdp (Service)	18.25	18.24	18.78	17.80	0.28	73
$lnex c^{+}$	1.32	1.12	3.67	0.00	1.01	73
$lnex c^{-}$	−0.48	−0.49	−0.03	−0.69	0.16	73
lnexp (Agriculture)	14.12	14.11	14.85	13.63	0.22	73
lnexp (Industry)	17.44	17.45	17.98	16.90	0.27	73
lnexp (Service)	16.31	16.34	17.06	15.53	0.36	73
lnoil	4.29	4.31	4.79	3.37	0.33	73

Table 2 presents the correlation matrices for the data adopted in the estimation models. The correlation matrices show generally low to moderate relationships among the variables, suggesting a low risk of multicollinearity. The majority of correlation coefficients between lngdp and other variables are below .7. Although $lnex c^{+}$ and $lnex c^{-}$ exhibit a pronounced negative correlation, since they represent positive and negative exchange rate shocks, their simultaneous inclusion in the model is theoretically warranted.

Table 2.

Correlation Matrices.

Agriculture
	lngdp	$lnex c^{+}$	$lnex c^{-}$	lnexp	lnoil
lngdp	1
$lnex c^{+}$	0.23	1
$lnex c^{-}$	−0.26	−0.80	1
lnexp	0.05	0.67	0.65	1
lnoil	−0.09	−0.15	0.14	0.01	1
Industry
	lngdp	$lnex c^{+}$	$lnex c^{-}$	lnexp	lnoil
lngdp	1
$lnex c^{+}$	−0.04	1
$lnex c^{-}$	−0.07	−0.80	1
lnexp	0.68	−0.06	0.10	1
lnoil	0.26	−0.15	0.14	0.47	1
Services
	lngdp	$lnex c^{+}$	$lnex c^{-}$	lnexp	lnoil
lngdp	1
$lnex c^{+}$	0.21	1
$lnex c^{-}$	−0.27	−0.80	1
lnexp	0.25	0.25	−0.13	1
lnoil	0.40	−0.15	0.14	0.15	1

There is no evidence of significant linear correlations among other variables, indicating that multicollinearity tends not to be a concern. Therefore, the regression results can be considered reliable, and the independent effects of the explanatory variables can be interpreted with confidence.

The NARDL method assumes that the variables included in the model are exogenous and represent fluctuations arising from self-adjusting market dynamics. However, considering the sample period covered in the analysis, there is a risk that authorities may have implemented various interventions to ensure economic stability, which could potentially be a source of endogeneity. In this case, a causality analysis was conducted to investigate the direction of relationships among the variables. According to the Granger Causality Test results provided in the Appendix, it was concluded that there is a causality relationship from the independent variables in our model to lngdp at various statistical confidence levels, whereas no significant causality was found from lngdp to the independent variables. These results indicate that the variables are generally exogenous, hence there is no significant endogeneity risk.

Before proceeding with the NARDL estimation, the stationarity of the time series is assessed utilizing the Augmented Dickey-Fuller (ADF) and Phillips-Perron (PP) unit root tests. The findings from these tests are summarized in Table 3. Both unit root tests employed model specifications that included a constant and a constant plus trend. The results indicate that all variables, with the exception of the $lnex c^{-}$ variable, were stationary at their first difference, while the $lnex c^{-}$ variable was stationary at its level. Consequently, the requisite condition for conducting the NARDL estimations has been fulfilled.

Table 3.

ADF and PP Unit Root Tests.

Augmented Dickey-Fuller (ADF)
Variables	(Level)		(First difference)
Variables	c	c + t	c	c + t
lngdp (Agriculture)	−1.583	−2.671	−5.598***	−5.806***
lngdp (Industry)	−0.488	−2.893	−10.940***	−10.859***
lngdp (Service)	0.165	−3.048	−10.681***	−10.718***
$lnex c^{+}$	3.921	2.215	−6.385***	−7.231***
$lnex c^{-}$	−4.113***	−4.827***
lngdp (Agriculture)	0.823	−0.428	−13.153***	−13.233***
lngdp (Industry)	−1.231	−3.068	−9.222***	−9.152***
lngdp (Service)	−1.872	−2.925	−8.669***	−8.602***
lnoil	−2.267	−2.334	−6.861***	−6.809***
Phillips-Perron (PP)
	(Level)		(First difference)
Variables	c	c + t	c	c + t
lngdp (Agriculture)	−0.910	−4.706***	−12.140***	−12.979***
lngdp (Industry)	−0.689	−3.426	−11.165***	−11.081***
lngdp (Service)	−0.084	−3.260	−10.753***	−10.904***
$lnex c^{+}$	4.814	0.841	−6.385***	−7.213***
$lnex c^{-}$	−4.238***	−5.220***
lngdp (Agriculture)	−0.254	−2.429	−12.950***	−13.031***
lngdp (Industry)	−1.385	−3.068	−9.195***	−9.126***
lngdp (Service)	−1.853	−3.002	−8.685***	−8.616***
lnoil	−2.466	−2.537	−6.757***	−6.699***

Note. *** indicates the stationarity of the series at the 1% significance level.

Using data from the agriculture, industry, and services sectors, the model specified in Equation 6 was estimated, and the results are presented in Table 4. Initially, the bound test results ( $F_{PSS}$ ) indicate that the estimations for all three sectors suggest a cointegration relationship among the variables in the model. Furthermore, the asymmetry test ( $W_{LR}$ ) yields significant results for the agriculture and industry sectors, indicating the presence of an asymmetric long-term relationship in these sectors; however, no such relationship is identified in the services sector. Additionally, the Breusch-Godfrey serial correlation test ( $χ_{BG}^{2}$ ), Ramsey-Reset specification test ( $χ_{RAMSEY}^{2})$ , Breusch-Pagan-Godfrey heteroskedasticity test ( $χ_{BPG}^{2})$ , Jarque-Bera normality test ( $χ_{J - B}^{2}$ ), and CUSUM-CUSUMSQ tests presented in Table 4, Panel B predominantly indicate that there are no significant statistical issues affecting the estimations. Table 4, Panel A, displays the long-term coefficients derived from the estimations.

Table 4.

NARDL Long Term Coefficients and Diagnostic Tests.

Variables	Agriculture	Industry	Services
Panel A: Long-term coefficients
Variable	Coef. (Std.Er.) [Prob]	Coef. (Std.Er.) [Prob]	Coef. (Std.Er.) [Prob]
$lnex c^{+}$	−0.184 (0.035) [0.000]	−0.024 (0.049) [0.621]	0.075 (0.044) [0.102]
$lnex c^{-}$	−0.712 (0.276) [0.013]	−1.187 (0.358) [0.002]	−0.920 (0.291) [0.003]
lnexp	1.281 (0.019) [0.000]	0.522 (0.155) [0.002]	0.359 (0.077) [0.000]
lnoil	−0.307 (0.048) [0.000]	−0.155 (0.042) [0.000]	−0.115 (0.038) [0.004]
Panel B: Descriptive statistics
Test	Stat. [Prob.]	Stat. [Prob.]	Stat. [Prob.]
$F_{PSS}$	7.069***	4.772***	5.946***
$W_{LR}$	4.689**	5.688**	5.353**
$χ_{BG}^{2}$	1.692 [0.156]	0.217 [0.805]	0.549 [0.582]
$χ_{RAMSEY}^{2}$	1.372 [0.246]	2.951 [0.095]	0.814 [0.373]
$χ_{BPG}^{2}$	1.034 [0.454]	1.199 [0.304]	0.689 [0.831]
$χ_{J - B}^{2}$	0.964 [0.617]	0.304 [0.858]	0.735 [0.692]
CUSUM	Stable	Stable	Stable
CUSUM2	Stable	Stable	Stable

Note. *** and ** indicate the stationarity of the series at the 1% and 5% significance levels, respectively.

According to Table 4, the significant asymmetric exchange rate effect is pronounced in the agricultural sector, as reflected by the coefficient of $lnex c^{-}$ reported at −0.712. This indicates that a 1% shock in the $lnex c^{-}$ variable brings about an average long-term increase of 0.71% in agricultural production. The influence of the $lnex c^{+}$ variable is statistically significant, with a 1% shock in the $lnex c^{+}$ variable resulting in a 0.18% decrease in agricultural production. In addition, the statistically significant positive coefficient at the 1% level for the lnexp variable (1.281) demonstrates that a 1% rise in lnexp translates into a 1.2% increase in agricultural production. The significant negative coefficient of −0.307 for the lnoil variable signifies that a 1% increase in lnoil results in a detrimental effect of 0.3% on agricultural production.

The long-term coefficients derived from the examination of the industrial sector, however, indicate a distinct reaction to currency shocks. Although the coefficient of the $lnex c^{+}$ variable is negative (−0.024), it is not statistically significant. Nonetheless, the coefficient associated with the $lnex c^{-}$ variable is −1.187 and demonstrates statistical significance at the 1% level. This result suggests that a 1% deviation in the $lnex c^{-}$ variable results in a 1.19% spike in industrial production. The statistically significant positive coefficient of 0.522 for the lnexp variable implies a 1% shock in this variable produces an upward effect of 0.5% on industrial production. Additionally, a 1% shock in lnoil results in a decrease of industrial production by 0.15%.

In the context of the services sector, perturbations in $lnex c^{+}$ exert positive, though statistically insignificant, impact. A 1% rise in $lnex c^{-}$ and lnoil leads to a reduction of 0.9% and 0.11% in production within the services sector, respectively. Conversely, the significant positive coefficient for the lnexp variable (0.359) shows that a 1% spike in the lnexp variable boosts service sector production by 0.35%.

Table 5 displays the estimation results of the error correction model. The coefficient of the error correction term ( $ect (- 1))$ is negative and statistically significant across all three sectors analyzed. This finding suggests that any shock leading to a deviation from the long-term equilibrium will tend to revert back toward balance, thereby affirming the efficacy of the error correction mechanism. Upon analyzing the short-term coefficient estimates, it is evident that the directional effects of $lnex c^{-}$ , lnoil, and lnexp remain approximately consistent with those observed in the long run. In contrast, the $lnex c^{+}$ variable diverges from long-term outcomes solely in the positive effect it generates on first-period lagged estimates for the industrial and service sectors.

Table 5.

NARDL Error Correction Model Estimations.

Agriculture		Industry		Services
Variable	Coef. (Std.Er.) [Prob]	Variable	Coef. (Std.Er.) [Prob]	Variable	Coef. (Std.Er.) [Prob]
$d (lngdp (- 1)$	−0.555 (0.109) [0.000]	$d (lngdp (- 1)$	−0.175 (0.099) [0.086]	$d (lnex c^{+})$	0.091 (0.051) [0.084]
$d (lngdp (- 2)$	−0.447 (0.122) [0.000]	$d (lngdp (- 2)$	0.147 (0.098) [0.143]	$d (lnex c^{+} (- 1))$	−0.129 (0.048) [0.011]
$d (lngdp (- 3)$	−0.158 (0.119) [0.192]	$d (lnex c^{+})$	0.124 (0.057) [0.035]	$d (lnex c^{+} (- 2))$	0.021 (0.050) [0.668]
$d (lngdp (- 4)$	−0.351 (0.108) [0.002]	$d (lnex c^{+} (- 1))$	−0.235 (0.056) [0.000]	$d (lnex c^{+} (- 3))$	0.000 (0.050) [0.995]
$d (lngdp (- 5)$	−0.344 (0.101) [0.001]	$d (lnex c^{+} (- 2))$	0.065 (0.067) [0.341]	$d (lnex c^{+} (- 4))$	0.022 (0.049) [0.645]
$d (lnex c^{-})$	−0.331 (0.204) [0.113]	$d (lnex c^{+} (- 3))$	−0.126 (0.059) [0.042]	$d (lnex c^{+} (- 5))$	−0.048 (0.047) [0.320]
$d (lnex c^{-} (- 1))$	−0.264 (0.211) [0.219]	$d (lnex c^{+} (- 4))$	0.050 (0.056) [0.378]	$d (lnex c^{+} (- 6))$	−0.217 (0.053) [0.000]
$d (lnex c^{-} (- 2))$	0.034 (0.199) [0.864]	$d (lnex c^{+} (- 5))$	−0.203 (0.055) [0.000]	$d (lnex c^{-})$	0.559 (0.296) [0.067]
$d (lnex c^{-} (- 3))$	−0.569 (0.180) [0.003]	$d (lnex c^{-})$	−0.294 (0.266) [0.276]	$d (lnex c^{-} (- 1))$	0.815 (0.280) [0.006]
$d (lnex c^{-} (- 4))$	−0.856 (0.205) [0.000]	$d (lnex c^{-} (- 1))$	0.336 (0.247) [0.182]	$d (lnexp)$	0.303 (0.028) [0.000]
$d (lnex c^{-} (- 5))$	−0.627 (0.200) [0.003]	$d (lnex c^{-} (- 2))$	0.082 (0.259) [0.752]	$d (lnexp (- 1))$	−0.113 (0.035) [0.002]
$d (lnexp)$	−0.065 (0.046) [0.164]	$d (lnex c^{-} (- 3))$	0.577 (0.196) [0.005]	$d (lnexp (- 2))$	−0.056 (0.026) [0.039]
$d (lnexp (- 1)$	−0.512 (0.091) [0.000]		−0.622 (0.215) [0.006]	$d (lnexp (- 3))$	−0.035 (0.031) [0.275]
$d (lnexp (- 2)$	−0.399 (0.082) [0.000]	$d (lnex c^{-} (- 5))$	0.588 (0.195) [0.004]	$d (lnexp (- 4))$	−0.096 (0.026) [0.000]
$d (lnexp (- 3)$	−0.286 (0.074) [0.000]	$d (lnexp)$	0.500 (0.054) [0.000]	$d (lnoil)$	0.048 (0.020) [0.021]
$d (lnexp (- 4)$	−0.189 (0.071) [0.011]	$d (lnexp (- 1)$	−0.030 (0.068) [0.652]	$d (lnoil (- 1))$	−0.013 (0.021) [0.524]
$d (lnexp (- 5)$	−0.155 (0.055) [0.007]	$d (lnexp (- 2)$	−0.203 (0.062) [0.002]	$d (lnoil (- 2))$	0.089 (0.020) [0.000]
$d (lnoil)$	−0.037 (0.020) [0.072]	$d (lnexp (- 3)$	0.021 (0.046) [0.640]	$d (lnoil (- 3))$	−0.008 (0.020) [0.671]
$d (lnoil (- 1)$	0.051 (0.020) [0.015]	$d (lnexp (- 4)$	0.089 (0.042) [0.043]	$d (lnoil (- 4))$	0.059 (0.020) [0.012]
2008	−0.049 (0.012) [0.000]	$d (lnexp (- 5)$	−0.068 (0.048) [0.162]	2,020	−0.026 (0.029) [0.367]
$(ect (- 1))$	−0.355 (0.056) [0.000]	$d (lnoil)$	−0.008 (0.025) [0.735]	$(ect (- 1))$	−0.442 (0.069) [0.000]
		2,020	−0.062 (0.020) [0.004]
		$(ect (- 1))$	−0.379 (0.066) [0.000]

In the last stage of the analysis, the model employed in this study, which was initially estimated using a nonlinear method, is subsequently estimated using a linear approach. Table 6 presents the long-term coefficients and diagnostic test results derived from the ARDL methodology. The results of the bounds test ( $F_{PSS}$ ) indicate that the estimations for all three sectors confirm the presence of a cointegration relationship among the variables included in the model. Furthermore, the results from the Breusch-Godfrey serial correlation test ( $χ_{BG}^{2}$ ), Ramsey-Reset specification test ( $χ_{RAMSEY}^{2}$ ), Breusch-Pagan-Godfrey heteroskedasticity test ( $χ_{BPG}^{2}$ ), Jarque-Bera normality test ( $χ_{J - B}^{2})$ , and CUSUM-CUSUMSQ tests predominantly suggest that there are no significant statistical issues affecting the estimations.

Table 6.

ARDL Long Term Coefficient and Diagnostic Tests.

Variable	Agriculture	Industry	Services
Panel A: Long-term coefficients
Variable	Coef. (Std.Er.) [Prob]	Coef. (Std.Er.) [Prob]	Coef. (Std.Er.) [Prob]
lnexc	−0.200 (0.048) [0.000]	−0.147 (0.029) [0.000]	−0.114 (0.061) [0.066]
lnexp	1.364 (0.047) [0.000]	0.372 (0.118) [0.002]	1.199 (0.041) [0.000]
lnoil	−0.466 (0.135) [0.001]	−0.072 (0.036) [0.050]	−0.264 (0.152) [0.087]
Panel B: Descriptive statistics
Test	Stat. [Prob.]	Stat. [Prob.]	Stat. [Prob.]
$F_{PSS}$	5.734***	4.957***	3.861**
$χ_{BG}^{2}$	2.338 [0.108]	0.226 [0.798]	1.537 [0.225]
$χ_{RAMSEY}^{2}$	0.194 [0.660]	2.740 [0.105]	0.731 [0.396]
$χ_{BPG}^{2}$	1.113 [0.371]	1.576 [0.108]	1.564 [0.127]
$χ_{J - B}^{2}$	0.973 [0.614]	0.404 [0.816]	1.133 [0.567]
CUSUM	Stable	Stable	Stable
CUSUM2	Stable	Stable	Stable

Note. *** and ** indicate the stationarity of the series at the 1% and 5% significance levels, respectively.

An analysis of the long-term coefficients reveals that, the lnexc and lnoil variables display an adverse impact across all three sectors, whereas the lnexp variable exhibits a more profound positive pass-through in comparison with the other variables. The statistical significance of the lnexc and lnoil variables’ influence over the services sector is deemed marginal.

Table 7 presents the estimation results of the error correction model derived from the ARDL approach. The error correction term ( $ect (- 1))$ is negative and statistically significant across all three sectors, thereby affirming the efficacy of the error correction mechanism. The short-term coefficient estimates reveal a positive pass-through of lnexc into lngdp for the industrial sector. The lnexp variable negatively impacts the agricultural and industrial sectors, while positively affecting the service sector; the results for the lnoil variable parallel those of the lnexp variable.

Table 7.

ARDL Error Correction Model Estimations.

Agriculture		Industry		Services
Variable	Coef. (Std.Er.) [Prob]	Variable	Coef. (Std.Er.) [Prob]	Variable	Coef. (Std.Er.) [Prob]
$d (lngdp (- 1)$	−0.489 (0.111) [0.000]	$d (lngdp (- 1)$	−0.137 (0.123) [0.271]	$d (lnexp)$	0.291 (0.029) [0.000]
$d (lngdp (- 2)$	−0.358 (0.119) [0.004]	$d (lngdp (- 2)$	0.256 (0.117) [0.034]	$d (lnexp (- 1))$	−0.163 (0.032) [0.000]
$d (lngdp (- 3)$	−0.111 (0.126) [0.383]	$d (lnexc)$	0.123 (0.049) [0.017]	$d (lnexp (- 2))$	−0.088 (0.032) [0.009]
$d (lngdp (- 4)$	−0.296 (0.120) [0.017]	$d (lnexc (- 1))$	−0.095 (0.053) [0.078]	$d (lnexp (- 3))$	−0.016 (0.025) [0.514]
$d (lngdp (- 5)$	−0.308 (0.110) [0.007]	$d (lnexc (- 2))$	0.041 (0.053) [0.446]	$d (lnexp (- 4))$	−0.061 (0.025) [0.017]
$d (lnexp)$	−0.038 (0.048) [0.425]	$d (lnexc (- 3))$	0.067 (0.050) [0.191]	$d (lnoil)$	0.073 (0.025) [0.006]
$d (lnexp (- 1))$	−0.240 (0.074) [0.002]	$d (lnexp)$	0.535 (0.061) [0.000]	$d (lnoil (- 1))$	−0.026 (0.027) [0.336]
$d (lnexp (- 2))$	−0.121 (0.067) [0.077]	$d (lnexp (- 1))$	0.072 (0.070) [0.314]	$d (lnoil (- 2))$	0.092 (0.027) [0.001]
$d (lnoil)$	−0.034 (0.019) [0.079]	$d (lnexp (- 2))$	−0.201 (0.068) [0.005]	2,020	0.040 (0.010) [0.000]
$d (lnoil (- 1))$	0.041 (0.019) [0.041]	$d (lnexp (- 3))$	0.043 (0.052) [0.403]	$(ect (- 1)$	−0.110 (0.027) [0.000]
$d (lnoil (- 2))$	−0.033 (0.019) [0.082]	$d (lnexp (- 4))$	0.129 (0.046) [0.007]
2008	−0.035 (0.012) [0.004]	$d (lnoil)$	−0.006 (0.028) [0.824]
$(ect (- 1)$	−0.134 (0.027) [0.000]	2,020	−0.052 (0.022) [0.024]
		$(ect (- 1)$	−0.592 (0.114) [0.000]

Discussion

Upon analyzing the estimation results from NARDL and ARDL models, it is evident that the effects of positive and negative exchange rate shocks on sectoral output exhibit significant variation across the three sectors under consideration. In the context of agricultural sector production, a prolonged depreciation of the currency rate diminishes output levels, whereas an appreciation of the exchange rate enhances output. Conversely, an examination of the industrial and service sectors illustrates that the depreciation of the exchange rate seldom has a major long-term impact on sectoral production; yet it augments sectoral output in the short run for one quarter. This sectoral heterogeneity aligns with classical theoretical expectations derived from the Marshall–Lerner condition and subsequent open-economy extensions (see Bahmani et al., 2013; Levy-Yeyati et al., 2013).

The findings indicate that the effects of exchange rate fluctuations in Türkiye differ between the long-term and short-term perspectives and can be analyzed through two distinct approaches. These findings support the argument that in economies with strong import dependence, exchange-rate appreciation enhances production through the cost channel, whereas depreciation raises input costs, particularly in agriculture (see Ben Salem et al., 2024; Chatziantoniou et al., 2023). Türkiye exhibits a significant dependence on imported inputs for production, as illustrated by the sectoral reliance rates presented in Table 8, which is derived from data obtained from the TURKSTAT. Such results challenge the validity of devaluation-based growth strategies in highly import-dependent economies, echoing concerns pertaining to contractionary devaluation and credibility constraints emphasized in previous studies (see Eichengreen & Hausmann, 1999; Krugman, 1998; Shang-Jin & Wu, 2002). An appreciation depreciation of the exchange rate stimulates sectoral production by lowering the cost of imported inputs, whereas a depreciation elevates the cost of imported inputs primarily in the agriculture sector, thereby influencing sectoral output via the cost channel.

Table 8.

Imported Intermediate Goods Across Sectors.

Sectors	Intermediate goods usage (%)	Distribution of production across sectors (%)
Crop and animal production, hunting	15	97.78
Agriculture (total)	15
Food Products; beverages, and tobacco products	23	18.33
Textiles, wearing apparel, leather	19	15.84
Electricity, gas, steam, and air conditioning supply	41	13.34
Basic metals	66	9.11
Industry (total)	43.6
Land transport and transport via pipelines	19	16.76
Retail trade, except of motor vehicles and motorcycles	9	10.59
Accommodation and food service activities	11	8.03
Services (total)	13.79

Source. Turkish Statistical Institute [TURKSTAT] (2012).

The empirical results demonstrate that this relationship holds true across all three sectors, with a more pronounced effect observed in the industrial and services sectors. Consequently, variations in the exchange rate, especially appreciation, are beneficial to sectoral production across all three sectors by not only decreasing costs but also mitigating risk and uncertainty. Nevertheless, the transmission of depreciation in the industrial and services sectors continues to be ambiguous. This situation can be interpreted as the positive impact of competitive advantage owing to exchange rate depreciation in exports counterbalanced by the adverse consequences of excessive import reliance, yielding an absence of a discernible net effect.

This finding prompts important questions on the efficacy of devaluation-centric growth strategies in countries heavily reliant on imported intermediate products, such as Türkiye, and indicates that devaluation fails to function as an effective instrument for growth and production. Additionally, the rise in production, recorded during periods of enhanced stability in the Turkish lira, indicates that fostering production in Türkiye is fundamentally reliant on sustaining cost stability and reducing uncertainty.

This observation is consistent with the high dependency rates on imported inputs in these sectors, as detailed in Table 8. For example, the proportion of imported intermediate goods in total inputs is 43.6% in the industrial sector, escalating to 90% in the manufacture of chemicals and chemical products and 81% in the manufacture of computer, electronic, and optical products. The agricultural and services sectors exhibit relatively similar rates of imported intermediate goods utilization. However, the more significant negative impact of exchange rate fluctuations on sectoral output in the services sector is attributed to the varying capacity to substitute imported inputs with domestic alternatives. Indeed, approximately 25% of the output in the services sector is derived from land transportation and pipeline transport, as well as retail trade, excluding motor vehicles and motorcycles. This segment predominantly utilizes high-tech products or other imported inputs, which often lack domestic substitutes. Conversely, the agricultural sector benefits from the availability of local substitutes for relatively simpler production inputs, and the depreciation period for imported machinery and equipment in this sector is extended. Consequently, it is comparatively easier to mitigate the impact of disruptions on agricultural input imports resulting from fluctuations in exchange rates. Furthermore, classical economic theory asserts that a depreciation (or appreciation) of the exchange rate is expected to confer a competitive advantage (or disadvantage) in exports, thereby resulting in an increase in economic production. However, based on the empirical evidence gathered in this study, this theory appears to hold true only in the short term for the industrial and services sectors in Türkiye during the period under analysis.

When examining the econometric results pertaining to the control variables incorporated into the model, it becomes apparent that the findings largely align with established economic theory. Specifically, exports produce a significant positive influence on output in all industries. Furthermore, shifts in global oil prices, another control variable, negatively influence sectoral production. Beyond the general findings, the models re-estimated with interactive dummy variables provide important insights into the structural differences between crisis and non-crisis periods. During the 2007 to 2009 global financial crisis and the 2020 COVID-19 pandemic, the adverse effects of exchange-rate appreciation on industrial and service sector output became markedly stronger, while the agricultural sector demonstrated relative resilience. These results suggest that exchange rate shocks amplify existing structural vulnerabilities during turbulent periods, particularly in sectors with high import dependency and financial exposure. In contrast, during non-crisis periods, exchange-rate depreciation appears to support production through cost stability and improved expectations, emphasizing that policy interventions should differ across economic conditions. Considering the results obtained, it is evident that Türkiye is negatively impacted by currency depreciation, which contradicts the predictions of classical economic theory. Consequently, it is advisable to implement structural policies aimed at decreasing reliance on imported inputs. Additionally, the findings concerning oil prices further substantiate this situation, highlighting the necessity of measures to mitigate energy costs and reduce dependence on external energy sources. Therefore, structural reforms that reduce import dependence and strengthen policy credibility appear crucial for stabilizing production under exchange-rate volatility (see Chakrabarti & Sen, 2023; IMF, 2022).

Conclusion

This study examined the asymmetric effects of exchange rate fluctuations on the agricultural, industrial, and service sectors of Türkiye over the period 2005 to 2023 by employing NARDL and ARDL models. The findings reveal that the transmission of exchange rate movements to sectoral output is highly uneven. Currency appreciation was found to enhance production in all sectors by lowering the cost of imported inputs and reducing uncertainty, whereas depreciation generally exerted contractionary effects—particularly in the agricultural sector—due to cost pressures arising from import dependence. The results further indicate that the short-term expansionary effects of depreciation on industrial and service sectors dissipate in the long run, suggesting that devaluation does not serve as a reliable instrument for sustained growth in an import-dependent economy such as Türkiye. From a policy perspective, these results underscore the importance of structural measures aimed at reducing production costs and dependency on imported intermediate goods. Enhancing domestic input capacity, improving energy efficiency, and strengthening monetary policy credibility would mitigate the adverse impacts of exchange rate volatility on production. Furthermore, the differentiated effects observed between crisis and non-crisis periods highlight the need for flexible and sector-specific policy responses. In this regard, exchange rate management alone is insufficient to foster sustainable sectoral growth without broader institutional and structural reforms that enhance resilience to external shocks.

Theoretical Implications

The Mundell-Fleming, Marshall-Lerner, and J-curve theories remain applicable to Türkiye, as confirmed by this research, which considers sectoral differences within the industrial and services sectors. This study contributes to the economic theory literature by highlighting these sectoral variations. The theoretical contributions of this research extend beyond this; it effectively addresses a gap in the existing literature. The findings presented have the potential to enhance the scientific impact of the research. The sectoral growth context explored by the research model can pave the way for empirical tests in different countries in future studies. Consequently, it will be possible to observe the effects of exchange rates across various sectors, not only in Türkiye but also in other developing nations. But, specifically in Türkiye, the exchange rate is significantly influenced not only by market dynamics but also by policy interventions. This may be considered a source of endogeneity, which could have affected our estimations. Furthermore, different variables could be incorporated into the research based on the input dependencies of the countries being studied. For instance, oil prices represent a significant input variable for Türkiye due to its considerable external dependence on energy. Additionally, the prices of fuels required for agricultural machinery in Türkiye are directly related to this dependency. Therefore, it is recommended that future studies include variables that may influence sectoral growth levels in countries from social, economic, political, and other perspectives, alongside exchange rates. Finally, the role of exports in the model is underscored by the critical importance of export-based growth for developing countries. Overall, the research assumes a supportive, contributory, and suggestive role for theory, literature, and future models.

Practical Implications

There are several important practical lessons to be drawn from this research for Türkiye, which could potentially be applied to other countries with similar characteristics in the agricultural, industrial, and service sectors. Considering the historical development of Türkiye’s agricultural sector, it can still be said that it holds significant agricultural production potential. Although the percentage of arable land is decreasing due to global warming, it is entirely possible to increase agricultural production through appropriate measures. On the other hand, despite Türkiye’s past agricultural advantages, it currently imports a substantial volume of agricultural products. This situation has led to difficulties for agricultural producers due to competitive pricing. Imported inputs, such as fertilizers and feed, further increase the risk posed by exchange rate fluctuations. The findings of this study explicitly demonstrate that an appreciation in the exchange rate contributes to a rise in sectoral output, whilst a depreciation in the exchange rate results in a plummet in sectoral output, offering a solid framework that underpins the aforementioned situation. These results suggest that a stable exchange rate is crucial for promoting an effective development framework in Türkiye’s agricultural production. Further, it is imperative that long-term policies should be formulated to reduce the agricultural sector’s reliance on imported inputs. Policies ought to concentrate on encouraging the production of agricultural inputs consisting of fertilizers, feed, medicines, and seeds in Türkiye, with a particular emphasis on promoting research and development activities associated with these inputs. Steps to further incentivize and support agricultural production as well as policies that facilitate the work of agricultural producers are crucial. These steps may include reintroducing low-cost fuel policies for farmers, providing maintenance support based on land size, and offering additional incentives based on the production of specialized agricultural products. The combined effect of these measures can further encourage growth in the agricultural sector. In terms of the industrial and service sectors, it is pertinent to recognize that the appreciation of the exchange rate has altered sectoral output to an extent that contradicts Classical Economic Theory. This situation, as previously stated, can be attributed to the significant reliance on imported inputs in emerging economies such as Türkiye. In addition, growth policies that involve devaluing the local currency, especially in the industrial and service sectors, are of inadequate relevance considering their substantial reliance on imported inputs. When examining the industrial and service sectors, the constricting effects of exchange rate increases present significant problems for Türkiye. The country’s social structure largely consists of individuals who have migrated from rural areas to cities, leaving behind agriculture to seek employment opportunities in industry, with many of these workers earning minimum wage. The sustainability of employment opportunities for this workforce is dependent on the expansion and increased production of the industrial and service sectors. However, rising exchange rates pose a major obstacle to this. Economic incentive regimes such as the inward processing regime, which is applied particularly in export-oriented sectors to provide cheaper inputs for industrialists, may not be a viable solution. This is because rising exchange rates increase the cost of inputs for local producers, even if they are tax-exempt. At this point, fiscal measures to stabilize the exchange rate or a focus on domestically substitutable products are necessary. Policymakers should conduct thorough reviews of products that can be sourced domestically and develop new projects that enable industrialists to obtain the inputs they need more affordably from within the country. On the other hand, the high profit expectations of capital also pose another risk, as exchange rate increases lead to a shift from production to exchange rate-based financial investments. This creates a second alternative for capital outflows from production, exacerbated by high interest rates that have been raised to reduce the exchange rate. As seen, additional variables related to social, political, and economic factors necessitate offering different recommendations for the production of Türkiye’s agricultural, industrial, and service sectors.

Footnotes

Appendix

Table 1.

Granger Causality Test.

	Agriculture		Industry		Services
	F-stat	[Prob]	F-stat	[Prob]	F-stat	[Prob]
$H_{0}$ : $lnex c^{+}$ does not Granger Cause lngdp	1.005	[0.371]	2.175	[0.122]	2.657	[0.078]
$H_{0}$ : lngdp does not Granger Cause $lnex c^{+}$	0.110	[0.895]	0.651	[0.525]	0.990	[0.377]
$H_{0}$ : $lnex c^{-}$ does not Granger Cause lngdp	6.386	[0.003]	4.608	[0.013]	4.340	[0.017]
$H_{0}$ : lngdp does not Granger Cause $lnex c^{-}$	1.954	[0.150]	1.450	[0.242]	1.278	[0.285]
$H_{0}$ : lnexp does not Granger Cause lngdp	1.050	[0.355]	2.605	[0.081]	4.711	[0.012]
$H_{0}$ : lngdp does not Granger Cause lnexp	0.044	[0.956]	0.739	[0.481]	1.482	[0.235]
$H_{0}$ : lnoil does not Granger Cause lngdp	0.253	[0.777]	2.980	[0.058]	0.744	[0.479]
$H_{0}$ : lngdp does not Granger Cause lnoil	0.378	[0.686]	0.835	[0.438]	0.490	[0.614]

ORCID iDs

Çağlar Orak

Musa Bayır

Ahmed Yusuf Sarıhan

Ethical Considerations

There are no human participants in this article.

Consent to Participate

Informed consent is not required.

Consent for Publication

All authors have approved the submission and publication of this manuscript in this journal.

Author Contributions

The authors have contributed equally to this work. All authors read and approved the final manuscript.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

Data will be made available on request.

Disclosure Statement

The authors certify that they have NO affiliations with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers’ bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or non-financial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this manuscript.

References

Ahmed

Aizenman

Saadaoui

Uddin

G. S.

(2023). On the effectiveness of foreign exchange reserves during the 2021-22 U.S. monetary tightening cycle. Economics Letters, 233, 1–31. https://doi.org/10.1016/j.econlet.2023.111367

Alejandro

C. F.

(1963). A note on the impact of devaluation and the redistributive effect. Journal of Political Economy, 71(6), 577–580. https://www.jstor.org/stable/1828441

Ali

Wahab

Khan

(2022). The effects of devaluation of currency on economic growth: The mediating role of Current Account Deficit. Global Economic Review, ii(I), 145–156. https://doi.org/10.31703/ger.2022(VII-I).12

Wynne

M. A.

Zhang

(2021). Shock-dependent exchange rate pass-through: Evidence based on a narrative sign approach for Japan. Journal of International Money and Finance, 118, 1–16. https://doi.org/10.1016/j.jimonfin.2021.102462

Ari

Cergibozan

Cevik

(2019). J-curve in Turkish bilateral trade: A nonlinear approach. International Trade Journal, 33(1), 31–53. https://doi.org/10.1080/08853908.2018.1521316

Babubudjnauth

Seetanah

(2021). An empirical analysis of the impacts of real exchange rate on GDP, manufacturing output and services sector in Mauritius. International Journal of Finance & Economics, 26(2), 1657–1669. https://doi.org/10.1002/ijfe.1869

Bahmani

Harvey

Hegerty

S. W.

(2013). Empirical tests of the Marshall-Lerner condition: A literature review. Journal of Economic Studies, 40(3), 411–443. https://doi.org/10.1108/01443581311283989

Bahmani-Oskooee

Arize

A. C.

(2020b). Asymmetric response of domestic production to exchange rate changes: evidence from Africa. Economic Change and Restructuring, 53(1), 1–24. https://doi.org/10.1007/s10644-018-9240-y

Bahmani-Oskooee

Durmaz

(2020a). Asymmetric cointegration and the J-curve: Evidence from commodity trade between Turkey and EU. Empirica, 47, 757–792. https://doi.org/10.1007/s10663-019-09455-4

10.

Beirne

Renzhi

Panthi

(2024). Exchange rate pass-through in emerging Asia and exposure to external shocks. Economic Analysis and Policy, 81, 1608–1624. https://doi.org/10.1016/j.eap.2023.10.033

11.

Ben Salem

Nouira

Saafi

Rault

(2024). How do oil prices affect the GDP and its components? New evidence from a time-varying threshold model. Energy Policy, 190, 1–23. https://doi.org/10.1016/j.enpol.2024.114162

12.

Bernoth

Herwartz

(2021). Exchange rates, foreign currency exposure and sovereign risk. Journal of International Money and Finance, 117, 1–38. https://doi.org/10.1016/j.jimonfin.2021.102454

13.

Bhatia

Tuteja

(2024). Contagion and linkages across international currencies. International Review of Financial Analysis, 94, 1–14. https://doi.org/10.1016/j.irfa.2024.103301

14.

Bhat

J. A.

Bhat

S. A.

Parray

W. A.

(2025). Nonlinearity in exchange rate pass-through across BRICS: Role of business cycle and inflation. International Economics and Economic Policy, 22(1), 8. https://doi.org/10.1007/s10368-024-00635-7

15.

Bigerna

(2023). Energy price shocks, exchange rates and inflation nexus. Energy Economics, 128, 1–53. https://doi.org/10.1016/j.eneco.2023.107156

16.

Bosupeng

Naranpanawa

J.-J.

(2024). Does exchange rate volatility affect the impact of appreciation and depreciation on the trade balance? A nonlinear bivariate approach. Economic Modelling, 130, 1–25. https://doi.org/10.1016/j.econmod.2023.106592

17.

Bremmer

(2007). The J curve: A new way to understand why nations rise and fall. Simon & Schuster.

18.

Bruno

Shin

H. S.

(2017). Global Dollar Credit and carry trades: A firm-level analysis. Review of Financial Studies, 30(3), 703–749. https://doi.org/10.1093/rfs/hhw099

19.

Bussière

Gaulier

Steingress

(2020). Global Trade flows: Revisiting the exchange rate elasticities. Open Economies Review, 31, 25–78. https://doi.org/10.1007/s11079-019-09573-3

20.

Caballero

(2021). Corporate dollar debt and depreciations: All’s well that ends well? Journal of Banking & Finance, 130, 106–126. https://doi.org/10.1016/j.jbankfin.2021.106185

21.

Calvo

Reinhart

C. M.

(1996). Capital flows to Latin America: Is there evidence of contagion effects? (Policy Research Working Paper Series 161).

22.

Central Bank of the Republic of Türkiye (CBRT). (2024, July 20). Database. Retrieved July 20, 2024, from https://www.tcmb.gov.tr/

23.

Chakrabarti

Sen

(2023). Foreign currency borrowing and risk exposure of firms: An emerging market economy viewpoint. Journal of Policy Modeling, 45(6), 1246–1261. https://doi.org/10.1016/j.jpolmod.2023.09.004

24.

Chatziantoniou

Elsayed

A. H.

Gabauer

Gozgor

(2023). Oil price shocks and exchange rate dynamics: Evidence from decomposed and partial connectedness measures for oil importing and exporting economies. Energy Economics, 120, 1–20. https://doi.org/10.1016/j.eneco.2023.106627

25.

Chen

Ward

(2025). Output divergence in fixed exchange rate regimes. Journal of International Economics, 157, 1–24. https://doi.org/10.1016/j.jinteco.2025.104115

26.

Chowdhury

Sundaram

J. K.

(2023). Chronicles of debt crises foretold. Development and Change, 54(5), 994–1030. https://doi.org/10.1111/dech.12786

27.

Colak

Erden

Ozkan

(2024). Time-varying exchange rate pass-through over 2005–2021 using dynamic model averaging. International Review of Economics & Finance, 96, 1–20. https://doi.org/10.1016/j.iref.2024.103514

28.

Cooper

R. N.

(1992). Currency devaluation in developing countries. In Letiche

J. M.

(Ed.), Economic theory, econometrics, and Mathematical Economics, international economic policies and their theoretical foundations (pp. 742–770). Academic Press. https://doi.org/10.1016/B978-0-12-444281-8.50036-2

29.

Dalgic

H. C.

(2024). Financial dollarization in emerging markets: An insurance arrangement. International Economic Review, 65(3), 1189–1219. https://doi.org/10.1111/iere.12686

30.

Dées

Galesi

(2021). The global financial cycle and US monetary policy in an interconnected world. Journal of International Money and Finance, 115, 1–28. https://doi.org/10.1016/j.jimonfin.2021.102395

31.

Demir

Razmi

(2022). The real exchange rate and development: Theory, evidence, issues, and challenges. Journal of Economic Surveys, 36, 386–428. https://doi.org/10.1111/joes.12418

32.

Demirkılıç

(2021). Balance sheet effects of foreign currency debt and real exchange rate on corporate investment: Evidence from Turkey. Emerging Markets Review, 47, 1–21. https://doi.org/10.1016/j.ememar.2021.100796

33.

Dhar

(2021). Identification of extreme capital flows in emerging markets. International Review of Economics & Finance, 71, 359–384. https://doi.org/10.1016/j.iref.2020.09.002

34.

Dornbusch

Goldfajn

Valdes

R. O.

Edwards

Bruno

(1995). Currency crises and collapses. Brookings Papers on Economic Activity, 1995(2), 219–293. https://doi.org/10.2307/2534613

35.

Dornbusch

Leslie

H. H., F.

, (1988). The Open Economy: Tools for policymakers in developing countries. The World Bank. Oxford University Press.

36.

Dornbusch

Werner

Calvo

Fischer

(1994). Mexico: Stabilization, reform, and no growth. Brookings Papers on Economic Activity, 1994(1), 253–315. https://doi.org/10.2307/2534633

37.

Ebrahimi

Ghaderi

Saeed

(2020). Differences in exchange rate effects on production growth: The Case of Iranian Industries. Studies in Business and Economics, 15(3), 16–35. https://doi.org/10.2478/sbe-2020-0042

38.

Eichengreen

Hausmann

(1999). Exchange rates and financial fragility (NBER Working Paper No. w7418). https://ssrn.com/abstract=196288

39.

Enajero

(2021). Cryptocurrency, money demand and the Mundell-Fleming model of International Capital Mobility. Atlantic Economic Journal, 49(1), 57–69. https://doi.org/10.1007/s11293-021-09701-w

40.

Fernandez-Arias

Hausmann

(2001, March). Foreign Direct Investment: Good cholesterol? (IDB Working Paper No. 348). https://doi.org/10.2139/ssrn.1817215

41.

Ferrara

Yapi

(2022). Measuring exchange rate risks during periods of uncertainty. International Economics, 170, 202–212. https://doi.org/10.1016/j.inteco.2022.04.001

42.

Fisera

(2024). Exchange rates and the speed of economic recovery: The role of financial development. Economic Systems, 48(1), 1–46. https://doi.org/10.1016/j.ecosys.2023.101165

43.

Gabriel

de Santana

Jayme

R. L.

Jr da Costa Oreiro

(2020). Manufacturing, economic growth, and real exchange rate: Empirical evidence in panel data and input-output multipliers. PSL Quarterly Review, 73(292), 51–75. https://doi.org/10.13133/2037-3643_73.292_3

44.

Geldner

(2025). Exchange rates and trade balance dynamics: A quantile regression analysis. Applied Economics, 57, 1070–1104. https://doi.org/10.1080/00036846.2024.2311067

45.

Guitian

(1976). The effects of changes in the exchange rate on output, prices and the balance of payments. Journal of International Economics, 6(1), 65–74. https://doi.org/10.1016/0022-1996(76)90023-4

46.

Kose

M. A.

Otrok

Prasad

E. S.

(2025). Global macro-financial cycles and spillovers. Journal of Banking & Finance, 178, 1–32. https://doi.org/10.1016/j.jbankfin.2025.107512

47.

Marc Stocker

Yilmazkuday

(2020). Inflation and exchange rate pass-through. Journal of International Money and Finance, 105, 1–45. https://doi.org/10.1016/j.jimonfin.2020.102187

48.

Halicioglu

Bahmani-Oskooee

Mohammadian

(2018). On the asymmetric effects of exchange rate changes on domestic production in Turkey. Economic Change and Restructuring, 51, 97–112. https://doi.org/10.1007/s10644-017-9201-x

49.

Hanson

J. A.

(1983). Contractionary devaluation, substitution in production and consumption, and the role of the Labor Market. Journal of International Economics, 14, 179–189. https://doi.org/10.1016/0022-1996(83)90029-6

50.

Hardy

(2023). Foreign currency borrowing, balance sheet shocks, and real outcomes. Journal of International Money and Finance, 139, 1–40. https://doi.org/10.1016/j.jimonfin.2023.102969

51.

Hofmann

Patel

S. P.

(2022). Original sin redux: a model-based evaluation (BIS Working Papers 1004). https://doi.org/10.2139/ssrn.5354655

52.

Hutchison

Glick

(2000, January). Banking and currency crises: How common are twins? (Pacific Basin Working Paper Series 99-07). https://doi.org/10.2139/ssrn.288723

53.

Huynh

T. L. D.

Nasir

M. A.

Nguyen

D. K.

(2023). Spillovers and connectedness in foreign exchange markets: The role of trade policy uncertainty. Quarterly Review of Economics and Finance, 87, 191–199. https://doi.org/10.1016/j.qref.2020.09.001

54.

Inoguchi

(2025). The impact of global shocks on sovereign risk: Role of domestic factors. Economic Systems, 49(2), 1–42. https://doi.org/10.1016/j.ecosys.2024.101277

55.

International Monetary Fund (IMF). (2022). Assessing systemic financial stability risks due to Fx mismatches. Review of The Institutional View on The Liberalization and Management of Capital Flows, 1-35. https://doi.org/10.5089/9798400204593.007

56.

Kandil

(2019). Variation in nominal and real effective exchange rates: Evidence across developed and developing countries. International Review of Economics, 66, 181–219. https://doi.org/10.1007/s12232-019-00322-z

57.

Karamelikli

Bahmani-Oskooee

(2021). The Turkey-US commodity trade and the asymmetric J-curve. Economic Change and Restructuring, 54, 943–973. https://doi.org/10.1007/s10644-020-09298-1

58.

Khalid

Civcir

Özdeşer

(2025). Asymmetric effects of exchange rate volatility on Turkish-German commodity trade. Journal of International Trade & Economic Development, 34(5), 1004–1047. https://doi.org/10.1080/09638199.2024.2356780

59.

Kim

Roubini

(2000). Exchange rate anomalies in the industrial countries: A solution with a structural VAR approach. Journal of Monetary Economics, 45(3), 561–586. https://doi.org/10.1016/S0304-3932(00)00010-6

60.

Krugman

(1998). Currency crises. http://web.mit.edu/krugman/www/crises.html

61.

Krugman

Taylor

(1978). Contractionary effects of devaluation. Journal of International Economics, 8(3), 445–456. https://doi.org/10.1016/0022-1996(78)90007-7

62.

Lama

Budha

B. B.

Prabhesh

Rana

(2025). Exchange rate pass-through in Nepal. Journal of Asian Economics, 97, 1–19. https://doi.org/10.1016/j.asieco.2025.101897

63.

Leightner

J. E.

(2024). Small-Country Mundell–Fleming (IS/LM/BP) model predictions under both fixed and flexible exchange rates: Evidence from Australia and S. Korea. Journal of Risk and Financial Management, 17(11), 1–20. https://doi.org/10.3390/jrfm17110495

64.

Levy-Yeyati

Sturzenegger

Gluzmann

P. A.

(2013). Fear of appreciation. Journal of Development Economics, 101, 233–247. https://doi.org/10.1016/j.jdeveco.2012.11.008

65.

Liu

Shen

(2024). Sudden stop with local currency debt. Journal of International Economics, 148, 1–35. https://doi.org/10.1016/j.jinteco.2024.103888

66.

Lizondo

J. S.

Montiel

(2006, February 15). Contractionary devaluation in developing countries: An analytical overview (IMF Working Paper No. 88/51). https://ssrn.com/abstract=884824

67.

Magacho

Ribeiro

Rocha

(2022). Economic complexity and price competitiveness dependence: Empirical evidence using panel models. International Journal of Development Issues, 21(1), 142–158. https://doi.org/10.1108/IJDI-07-2021-0141

68.

Manopimoke

Nookhwun

Pattararangrong

(2024). Exchange rate in emerging markets: Shock absorber or source of shock? Journal of International Money and Finance, 148, 1–45. https://doi.org/10.1016/j.jimonfin.2024.103148

69.

Montane

Libman

Zack

(2021). Contractionary depreciations in Latin America during the 2000s. Brazilian Journal of Political Economy, 41(4), 723–744. https://doi.org/10.1590/0101-31572021-3196

70.

Montes

G. C.

da Silva Rodrigues Júnior

Bastos

J. C. A.

Batista

L. M.

(2024). Effects of monetary policy credibility and the open economy trilemma on monetary policy efficiency. International Journal of Finance & Economics, 29(2), 1236–1258. https://doi.org/10.1002/ijfe.2737

71.

Moses

T. K.

Victor

O. U.

Uwawunkonye

E. G.

Fumilade

O. S.

Nathaniel

(2020). Does exchange rate volatility affect economic growth in Nigeria? International Journal of Economics and Finance, 12(7), 54–71. https://doi.org/10.5539/ijef.v12n7p54

72.

Mous

N. S.

Maznan

N. H.

(2024). The impact of Monetary Policy Indicators on Economic Growth in Iraq (2007-2022). Journal of International Crisis and Risk Communication Research, 7(12), 950–962. https://doi.org/10.63278/jicrcr.vi.2213

73.

Nosheen

Chohan

Iqbal

Wohar

(2023). Asymmetric response of domestic production to exchange rate changes: Evidence from southeast Asian countries. Asian-Pacific Economic Literature, 37(2), 54–75. https://doi.org/10.1111/apel.12395

74.

Nusair

S. A.

(2022). The asymmetric effects of exchange rate changes on output: Evidence from Asian countries. International Trade Journal, 36(4), 324–349. https://doi.org/10.1080/08853908.2021.1940392

75.

Özdemir

Topcu

Emirmahmutoğlu

(2022). Bilateral J-curve between Turkey and the Eurozone. Mehmet Akif Ersoy University Journal of Economics and Administrative Sciences Faculty, 9(8), 1420–1432. https://doi.org/10.30798/makuiibf.1097414

76.

Pesaran

M. H.

Shin

Smith

R. J.

(2001). Bounds testing approaches to the analysis of level relationships. Journal of Applied Economics, 16(3), 289–326. https://doi.org/10.1002/jae.616

77.

Ramoni-Perazzi

Romero

(2022). Exchange rate volatility, corruption, and economic growth. Heliyon, 8(12), 1–12. https://doi.org/10.1016/j.heliyon.2022.e12328

78.

Ridhwan

M. M.

Ismail

Nijkamp

(2024). The real exchange rate and economic growth: A meta-analysis. Journal of Economic Studies, 51(2), 287–318. https://doi.org/10.1108/JES-10-2022-0548

79.

Rose

Glick

(1999). Contagion and trade: Why are currency crises regional? Journal of International Money and Finance, 18(4), 603–617. https://doi.org/10.1016/S0261-5606(99)00023-6

80.

Salomao

Varela

(2020). Exchange rate exposure and firm dynamics (CEPR Discussion Paper No. DP12654). https://ssrn.com/abstract=3112278

81.

Seraj

Coskuner

(2021). Real exchange rate effect on economic growth: comparison of fundamental equilibrium exchange rate and Balassa–Samuelson based Rodrik approach. Journal of Applied Economics, 24(1), 541–554. https://doi.org/10.1080/15140326.2021.1977083

82.

Shang-Jin

(2002). Negative alchemy? Corruption, composition of capital flows, and currency crises (NBER Working Papers 8187). National Bureau of Economic Research, Inc. https://doi.org/10.7208/chicago/9780226185057.003.0011

83.

Sharaf

M. F.

Shahen

A. M.

(2023). Asymmetric impact of real effective exchange rate changes on domestic output revisited: Evidence from Egypt. International Trade, Politics and Development, 7(1), 2–15. https://doi.org/10.1108/ITPD-09-2022-0020

84.

Shin

Greenwood-Nimmo

(2014). Modelling asymmetric cointegration and dynamic multipliers in a nonlinear ARDL framework. In Sickles

R. C.

Horrace

W. C.

(Eds.), Festschrift in Honor of Peter Schmidt (pp. 281–314). Springer. https://doi.org/10.2139/ssrn.1807745

85.

Smith

(2007). An inquiry into the nature and causes of the wealth of Nations. MetaLibri.

86.

Souza

R. D. S.

de Mattos

L. B.

(2022). Oil price shocks, exchange rate and uncertainty: Case of Latin American economies. Applied Economics Letters, 29(10), 880–886. https://doi.org/10.1080/13504851.2021.1896670

87.

Tenderere

Mishi

(2025). Interest rates and capital flows: An application of the Mundell-Fleming model in emerging markets. Africa’s Public Service Delivery & Performance Review, 13, 1–25. https://doi.org/10.4102/apsdpr.v13i1.904

88.

Tong

Durmaz

(2024). Bilateral trade of Turkey-US: Can Euro explain this relationship? Journal of International Trade & Economic Development, 33, 1691–1707. https://doi.org/10.1080/09638199.2023.2288907

89.

Trofimov

I. D.

(2024). The J-curve effect in agricultural commodity trade: An empirical study of South East Asian economies. Review of Economics and Political Science, 9(4), 358–381. https://doi.org/10.1108/REPS-05-2023-0038

90.

Turkish Statistical Institute (TURKSTAT). (2012, July 20). Supply and use tables, input-output tables. Retrieved July 20, 2024, from https://data.tuik.gov.tr/Bulten/Index?p=Arz-ve-Kullanim-Tablolari%2C-Girdi-Cikti-Tablolari-2012-24922

91.

Turkish Statistical Institute (TURKSTAT). (2024, July 20). Database. Retrieved July 20, 2024, from https://www.tuik.gov.tr/

92.

United Nations Trade and Development (UNCTAD). (2023). A world of debt: A growing burden to global prosperity. United Nations Global Crisis Response Group.

93.

Van Wijnbergen

(1986). Exchange rate management and stabilization policies in developing countries. Journal of Development Economics, 23(2), 227–247. https://doi.org/10.1016/0304-3878(86)90117-3

94.

Wang

(2020). The Mundell–Fleming model. In Wang

(Ed.), The Economics of Foreign Exchange and Global Finance (pp. 149–172). Springer.

95.

Zengeni

Mahonye

(2019). Exchange rate impact on output and inflation: A historical perspective from Zimbabwe. African Journal of Science Technology Innovation and Development, 11(3), 347–358. https://doi.org/10.1080/20421338.2019.1575539

96.

Zhang

Steiner

de Haan

Yang

(2024). Capital flow reversals and currency crises: Do capital flow types matter? Review of International Economics, 32(4), 1787–1823. https://doi.org/10.1111/roie.12753