Stabilizing the Consumer Price Index (CPI) to protect the populace from the adverse effects of inflation necessitates appropriate measures at both political and economic governance levels. This study examines the impacts of imports (IM) and exports (EX) on inflation (CPI) in Afghanistan using data from 1990 to 2023. The findings from the Autoregressive Distributed Lag (ARDL) model indicate that both IM and EX significantly impact CPI in the short and long term. A robustness check employing the Kernel-based Regularized Least Squares (KRLS) machine learning technique further validates these results. The analysis confirms that international trade has a substantial and positive effect on CPI. Additionally, in the context of Afghanistan, political instability acts as a positive moderator, amplifying the influence of imports and exports on inflation. The study concludes that the country requires a reevaluation of its policies regarding exchange rates and economic growth to mitigate the negative effects of imports, exports, and political volatility on the stability of the CPI.

Afghanistan Economy, Imports, Exports, Inflation, Political Instability

Afghanistan’s economy, marked by its unique geopolitical landscape and internal challenges, has long been vulnerable to fluctuations in trade and inflation. In a country heavily reliant on imports, the dynamics of trade flows can have profound effects on the cost of living and stability of the economy. However, the influence of trade on inflation is not merely a matter of economic theory or practice: in Afghanistan, it is intricately tied to political issues. Large trade imbalances and a persistently massive gap between imports and exports define the nation’s economic environment, posing risks that could intensify inflationary pressures. Afghanistan’s trade deficit as of 2023 was at 6.798 USD billion, with USD 8.576 billion in imports and USD 1.777 billion in exports (Trading Economics, 2024). These numbers highlight Afghanistan’s dependence on imported products showing how easily external shocks could upset the stability of domestic prices.

The dynamics of inflation in Afghanistan are closely linked to the cost of imported commodities that accounts for a significant portion of the consumer price index (CPI). In this relationship, the exchange rate pass-through effect is crucial; as the Afghani currency depreciates in value relative to other major currencies such as the US dollar, the cost of imported goods rises sharply driving up the rate of inflation. According to a Da Afghanistan Bank study, there is a 73.79% correlation between inflation and the exchange rate, meaning that changes in currency values have a serious effect on domestic prices (Iqbal & Naqibullah, 2020). Regression research has also shown that over 44% of inflation fluctuations may be explained by changes in the exchange rate, highlighting the significance of keeping an eye on foreign exchange dynamics to comprehend inflationary patterns.

Although imports are frequently criticized for contributing to inflation, exports also have an impact that should not be disregarded. Afghanistan’s export market is modest and primarily consists of low-value goods like agricultural and textile products. These sectors’ performance has the potential to influence domestic supply, which in turn may have an impact on inflation. For example, a significant rise in demand for exports may lead to a reduction in the supply of domestically produced goods driving up their prices. External factors, such as global commodity pricing and supply chain disruptions, may further complicate this relationship. Afghanistan’s heavy reliance on imports for essentials like food and fuel makes it especially vulnerable to shifts in international markets (UNCTAD, 2023).

One should not ignore the broader economic context. Afghanistan’s economy has suffered greatly since the Taliban took power in August 2021, as evidenced by a precipitous drop in GDP and a major cutback in humanitarian assistance (Asian Development Bank, 2024). Political instability has the power to drastically increase or decrease the impact of trade on inflation by obstructing import and export routes, changing the course of policy and undermining investor confidence; it has contributed to the environment characterized by persistent inflationary pressures and high unemployment rates. These macroeconomic factors interact with trade dynamics in a complicated way, with imports and exports significantly affecting inflation rates.

Effective policymaking in Afghanistan requires an understanding of how trade dynamics affect inflation as the country navigates its way to economic recovery in the face of persistent geopolitical uncertainty. By using econometric modelling and empirical analysis, this paper seeks to conduct a thorough examination of these linkages and provide important insights that can guide policymakers’ attempts to stabilize prices and foster sustainable economic growth by clarifying how changes in imports and exports affect inflation rates in Afghanistan’s economic environment.

Many researchers have examined the connection between inflation and trade flows, especially in developing and emerging nations. Since international trade is the main channel through which external economic forces affect domestic price levels, the importance of this relationship should not be underestimated. Prior research on this subject has mostly concentrated on how imports and exports affect inflation in larger economic contexts, paying special attention to trade liberalization, price stability, and the function of currency rates. The idea that trade plays a crucial role in reducing inflationary pressures through its impact on competition, market efficiency and access to foreign goods is usually supported by empirical data. Trade flows can help reduce inflation by improving resource allocation, increasing the supply of goods, and stabilizing domestic prices through integration with international markets.

Hussain and Malik (2011) investigate the connection between Pakistan’s inflation and economic growth and discover a strong inverse relationship. According to their study, excessive rates of inflation reduce purchasing power and raise doubts about investment choices, which impedes economic progress. This fundamental knowledge emphasizes the necessity of understanding how the external trade, primarily imports and exports, can affect inflationary tendencies. To design a successful economic policy that would cope with detrimental effects of inflation, the authorities need to have a strong grasp of trade dynamics. Following up on this idea, Ghauri et al. (2022) investigate how exports contribute to economic growth. They contend that exports boost foreign exchange reserves, which in turn helps reduce inflationary pressures while also promoting economic growth. This implies a clear correlation between export performance and inflation management, suggesting that increasing export capacity can be a strategic reaction to rising prices. Inflation stability and exports are positively correlated, which highlights the significance of trade policies that give priority to export growth.

In a related study, Munir and Javed (2018) examine the export composition of South Asian nations and how it affects economic expansion. They maintain that nations with a larger proportion of manufactured exports typically see better growth results. This suggests that the kind of exports has a significant impact on how economies develop when inflation rates fluctuate, supporting Raheem et al. who have shown that not only the size but also content of exports matters when it comes to their capacity to counteract inflation. Ghauri et al. (2022) evaluate Pakistan’s export dynamics using ARDL cointegration analysis. According to their findings, exports have a favorable impact on economic growth. However, when demand exceeds supply, they may also exacerbate inflationary pressures. The intricacy of overseeing export-led growth strategies in the face of price increases is highlighted by this dual effect. Here, the relationship between export growth and domestic price stability becomes clear, indicating that in order to prevent inflation from getting worse, authorities must carefully balance these elements.

Amiti et al. (2024), moving from national contexts to global perspectives, examine the rise in the U. S. import prices during the COVID-19 pandemic. They discovered that, because of global supply shocks and increases in demand unique to the United States, higher import price had a major impact on domestic inflation levels. For developing nations like Pakistan, where reliance on imports might worsen inflationary tendencies during global disruptions, this conclusion is especially pertinent. Their research shows how domestic inflation rates can be directly impacted by overseas market situations.

The relationship between inflation and import prices in the USA is examined by Corrigan (2005), who points out that rising import costs can result in higher consumer prices. This is exactly what happens in many developing countries like Pakistan whose economies depend significantly on imports for necessities. Corrigan’s observations align with the findings of Amiti et al. as they both show how countries that rely on imports may become economically vulnerable because of changes in global import prices. Herawati and Sidik (2022) conducted an analysis of how interest rates and imports affected inflation in ASEAN nations between 2006 and 2019. The idea that dependence on imports can worsen inflationary pressures in emerging nations is supported by their findings, which show that increasing import costs may contribute substantially to domestic inflation. Corrigan’s discoveries regarding import price sensitivity highlight the necessity for developing nations to diversify their economies and reduce their dependence on imported goods.

Using a Granger causality method, Usman and Bashir (2022) investigate how imports affect China’s economic growth. They find a reciprocal relationship in which imports both influence inflation dynamics and spur growth, demonstrating how managing growth and inflation can be made more difficult by interwoven trade relationships. Given that this link is reciprocal, policymakers should take into account both the influence of imports on growth and its subsequent effects on inflation. Turning their attention to regional research, Jackson et al. (2023) examine the factors that contribute to inflation in Sierra Leone, highlighting important variables including import costs and currency rates that have a major impact on inflation rates. Their results are consistent with prior research showing that external factors are important for domestic price levels, implying that both regional contexts and trade influences should be taken into account when examining inflation dynamics. Andinuur (2013) investigates the interaction of Ghana’s economic development and foreign direct investment (FDI). According to the study, excessive inflation can discourage foreign investment, posing a cyclical dilemma for policymakers, even if FDI can reduce inflationary pressures by increasing productive capacity. This cyclical relationship implies that internal economic conditions can impact external investment flows and control over inflation becomes essential for drawing in foreign direct investment (FDI).

The existing research suggests that imports, exports, and inflation rates interact in a complicated way. Imports frequently cause inflation through higher prices and supply chain weaknesses; increases in exports are typically associated with rising inflation because of improved domestic demand and price adjustments. These connections, however, have not been fully examined in relation to Afghanistan, especially considering the nation’s persistent political unrest and the economic fallout from the U. S. withdrawal. Since Afghanistan’s economic resilience has been severely undermined by the sudden end of international aid, it is crucial to comprehend how trade dynamics affect inflation in this particular setting. This paper offers a thorough examination of the connections between imports, exports, and inflation in the context of Afghanistan. It does this in an effort to educate policymakers on practical methods for maintaining price stability and encouraging long-term economic growth in a country that has to deal with excessive reliance on trade.

Descriptive data are presented in Table 2 above, and the study shows the standard deviations of CPI, IM, EX, PI, EXR, GDP, and TAX, confirming that these indicators represent a moderate level of volatility. Furthermore, the Jarque-Bera statistic indicates that the expected observations have a normal distribution and proves that the P-Value is not statistically significant. Before doing any quantitative analysis, the stationarity of the time series must be verified in order to prevent pseudo-regression and confirm the veracity of the expected results. This study uses the ARDL technique to estimate the long and short-run effect of explanatory variables on CPI. One unique aspect of the ARDL technique is that it focuses on the interactions between identical sequences without requiring that their components be connected. To maintain the economic importance of the data, it should limit the inclusion of the factors in the framework to the first order. With too much information excluded, capturing the second-order variance could result in erroneous regression.

Table 3 displays the results of the ADF and Zivot & Andrews structural break unit root tests that we used. It is also possible to confirm that all parameters remain dependable stationary at the first difference. The date of the break is considered to be an endogenous study variable in the structural break unit root test, which was created by Zivot and Andrews (1992). The results of the Zivot and Andrews test are also included in Table 3, which indicates that all the variables are stationary at the first difference along with the associated break years. Clearly, the ARDL model is a good choice for the subsequent stages of analysis. Before applying ARDL technique, we might ascertain whether the variables are co-integrated and, if so, whether it is worthwhile to move further with the boundary test by calculating the estimated value of the F-Statistic and T-Statistics.

Table 4 depicts the outcomes of the boundaries cointegration test and lag selection criterion. One practical choice for determining the lag sequence is the AIC, which is used to perform boundary tests on the data, with the optimal lag order determined to be 1. The results at 1 % significance level, show that there is a strong co-integration relationship between the variables because the calculated F-Statistic and T-Statistic values of 7.636 and -4.291 greatly exceed the critical values.

We performed a sequential analysis prior to applying the ARDL technique to examine the short- and long-term effects of explanatory factors. In order to better understand the relationship between IM, EX, PI and CPI, benchmark regression and moderation effect exploration were required.

The findings of the benchmark regression analysis are shown in Table 5. Given that the IM and EX coefficients are positive and significant at 1% and 5% significance level, it may be concluded that rising IM and EX support CPI. Likewise, there is still a favorable association between PI and CPI. To account for their possible impact on CPI and separate the precise effects of IM and EX, control variables including EXR, GDP and TAX were added.

The moderating effect of Political Instability (PI) on the influence of IM and EX on CPI is displayed in Table 6. The moderating variable PI amplifies the effects of both IM and EX on CPI, as indicated by the positive coefficients for the interaction terms (IM*PI and EX*PI). In particular, this indicates that the positive effect of IM on CPI increases with PI levels, indicating that higher PI amplifies the advantages of imports on consumer costs.

Reliance on imported goods that often increases during brief periods of stability becomes even greater because of limited domestic output and prices rising in the absence of competition. High prices are also a result of ongoing inflationary pressures, speculative pricing, and currency volatility. Besides, CPI can be impacted by improved supply chain capabilities and growing customer preferences for imports, especially when a small number of companies control a large portion of the market.

Similarly, the positive impact of EX*PI on CPI might also be a result of higher demand for Afghan products in politically stable environments, which could raise prices because of higher production costs or more competition in the market. In the case of Afghanistan, these dynamics highlight the intricate connection between political circumstances and economic conduct.

Table 7 shows that Afghanistan’s Consumer Price Index (CPI) is negatively impacted by WTO membership, with a significant coefficient of -0.076, suggesting that greater trade activity results in reduced costs. On the other hand, the US exit is associated with an increase in the CPI, as indicated by a significant coefficient of 0.094, meaning that it causes instability and higher expenses. Other variables are controlled for in both models (with high adjusted R-squared values of 0.972 and 0.976), underscoring the significance of these processes in determining Afghanistan’s economic environment after significant political shifts.

The ARDL approach can be used to estimate the regression and draw conclusions about long-term impacts once the Boundary test has shown the presence of a long-term cointegration relationship among parameters. At this point, we analyze the long-term effects and investigate how IM, EX, and PI affect CPI over an extended period of time. Table 8 depicts the ARDL long- and short-term estimation outcomes. Long-term results indicate that IM and EX have a positive impact on CPI, with coefficients of 0.214 and 0.187, respectively. This means that a 1% increase in IM and EX will result in an increase in CPI of 0.214% and 0.187%, respectively. Stronger effects are shown by PI and EXR, with coefficients of 0.286 and 0.769, respectively, suggesting that strong inflationary pressures are linked to increased currency instability and depreciation. With TAX showing a slight positive effect (0.084), GDP likewise exhibits a positive correlation with a coefficient of 0.305 which are significant at 5% and 10% significance level, respectively. The long- and short-run estimations presented in this paper align with Amiti et al. (2024) and Herawati et al. (2022), where import and export positively affect CPI.

The results of additional residual diagnostic tests are also shown in Table 8, including the Jarque-Bera, Ramsey RESET, Breusch-Godfrey serial correlation LM, Durbin-Watson stat, and Breusch-Pagan-Godfrey heteroscedasticity test. The diagnostic tests show that the long-term estimating techniques have produced consistent, dependable results with useful policy implications. We employed the CUSUM approach to assess the prediction model’s durability in an effort to avoid aptitude model unsteadiness that results from parameter upheaval. The model’s structure is consistent, as shown by Fig. 4, with the residuals (CUSUM) between the borders.

Figure 4. 

COSUM

4.1. Robustness check: KRLS-based machine learning approach

The current study examines the causal link between chosen indicators using point-wise differentials and the KRLS machine learning algorithm. By examining the forecasters’ marginal effects at each stage, the previously indicated empirical method enables us to investigate the structural changes in Afghanistan’s economic management. The effects of IM, EX, PI, and EXR are shown to be positive and significant in Table 9. In keeping with the primary finding of the ARDL approach, the effects of GDP and TAX are likewise positive but not statistically significant. The model’s R²­ result, which is 0.9888, shows that the 98.88% adjustments may be justified by the current model’s analytical ability with the specific regression. Using derivatives, the percentiles obtained (25^th, 50^th, and 75^th) indicate the time-wise marginal impacts on the dependent variable (CPI) of the explanatory variables (IM, EX, PI, EXR, GDP, and TAX). The point-wise derivatives support the findings’ sensitivity, suggesting that the diverse marginal effects of the impacted indicators are absent. As an immediate consequence, we analyze the long-term modifications of the explanatory indicators, such as how their changes impact the difference in the regressed CPI.

Table 9.

Download as

CSV

XLSX

Results of Machine Learning Approach

Variables	Average	SE	P-Value	P25	P50	P75
IM	0.071	0.010	0.000	0.170	0.302	0.372
EX	0.045	0.021	0.042	0.035	0.060	0.061
PI	0.032	0.002	0.000	0.021	0.086	0.098
EXR	0.098	0.043	0.018	0.894	0.696	0.622
GDP	0.051	0.043	0.501	0.124	0.157	0.234
TAX	0.013	0.009	0.105	-0.198	0.208	0.162
Diagnostic
	Lambda	Tolerance	sigma	Eff.df	R2	Looloss
	0.0108	0.0428	0.0174	4.1633	0.9888	33.1309

According to the data displayed in Fig. 5, the marginal impact of IM primary indicates a positive and growing influence on Afghanistan’s CPI. It tends to fluctuate and stay positive at higher CPI levels. EX also has an increasingly positive impact on CPI. The CPI is positively impacted by PI and EXR, which are somewhat flat at the beginning, sharp in the middle, and then tend to slow down at the end. At the beginning, GDP had a negative impact on CPI; at the end, however, its effect turned positive. In a similar vein, TAX had a slight positive effect on the CPI at first, but later it became stronger. The results obtained are consistent with our primary model, the ARDL-based counterfactual assessment.

Figure 5. 

Marginal Effects Through Machine Learning Approach

In this study we aim to find out how imports (IM), exports (EX), and political instability (PI) affect Afghanistan’s Consumer Price Index (CPI), taking into account the impact of important macroeconomic factors like GDP, taxation (TAX), and exchange rates (EXR). The study covers the years 1990–2023, offering a thorough examination of the ways in which both domestic and international factors influence changes in the country’s price level. According to the findings, the currency rate volatility, GDP growth, and taxation policies all have a major impact on Afghanistan’s CPI. The analysis has shown that rising imports and exports have compounding impacts on local market pricing and availability of products and services. Political instability exacerbates price volatility, interferes with market systems and undermines economic stability. Together, these factors show that a complex interaction between fiscal policies, external trade activity, and sociopolitical issues shapes Afghanistan’s CPI.

Yang Jingjing: Supervision, Writing – review & editing, Validation, Resources, Project administration, Funding acquisition. Shah Mir Mowahed: Writing – review & editing, Writing – original draft, Methodology, Investigation, Formal analysis, Conceptualization, Software. Mohammad Wais Sharif Zada: Software, Validation, Visualization, Writing – review & editing, Investigation, Data curation.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

This work was supported by the research on impact of Hunan Talent policy to technology innovation. Funded by Hunan Provincial Social Science Foundation, Reference No.22YBA027.

Data is available up-on request.