AGRARIS: Journal of Agribusiness and Rural Development Research Vol. 11 No. 2 July - December 2025. Pages: 200-214 RESEARCH ARTICLE Article history: Submitted : June 7th, 2024 Revised : May 22nd, 2025 Accepted : December 11th, 2025 Atikah Fajriyah Mubarok 1,*. Komalawati 2 and Agus Setiadi 1 Department of Agribusiness. Faculty of Animal Husbandry and Agriculture. Diponegoro University. Jl. Prof. Soedarto. Tembalang. Semarang. Indonesia Research Center for Cooperatives. Corporations, and People's Economy. National Research and Innovation Agency. Jl. Gatot Subroto. South Jakarta. Indonesia *) Correspondence email: atikahfajriyahm@gmail. Beef Price Volatility in Indonesia: Before. During, and After the COVID-19 Pandemic DOI: https://doi. org/10. 18196/agraris. ABSTRACT IndonesiaAos beef market has long exhibited structural volatility, which became more pronounced during major disruptions such as COVID-19. Analyzing these fluctuations across different periods is crucial for strengthening market resilience. This paper examines beef price volatility in Indonesia across three critical periods, before the pandemic . 7Ae2. , during the COVID-19 pandemic (March 2020 Ae June 2. , and after the pandemic (June 2023 onwar. , and identifies the key factors influencing it. Daily price data for beef, chicken meat, and eggs were obtained from the Indonesian Strategic Food Price Information Center (PIHPS) for 2017Ae2024 and analyzed using the GARCH . model and logarithmic regression with crisis dummy variables. The results show that beef price volatility increased significantly during the pandemic and remained high in the new normal period, confirming the long-term persistence of price shocks. Significant influencing factors include the COVID-19 pandemic, seasonal events (Ramadan and Eid al-Fit. , the price and volatility of chicken meat, and lagged beef imports from previous periods. The methodological contribution of this study lies in the use of daily data and a time-lag structure that captures short-term dynamics more These results underscore the need for structural reform, daily price monitoring systems, and adaptive market intervention to strengthen IndonesiaAos food security and market resilience. Keywords: Beef. COVID-19. GARCH. Price stabilization. Price volatility INTRODUCTION The COVID-19 pandemic has emerged as a major global disruption that reshaped economic systems worldwide, including in Indonesia (Rela et al. , 2022. Seshaiyer & McNeely. Whitehead & Kim, 2. Beyond its public health impacts, the pandemic triggered structural shocks in national food systems (Khedhiri, 2023. Nurhidayah & Djalante, 2022. Yudha & Roche, 2. , including beef as a particularly vulnerable commodity due to its price sensitivity (Calvia, 2024. Surni et al. , 2. Even before the COVID-19 pandemic, beef prices in Indonesia were highly volatile due to structural challenges, including heavy reliance on imports . ccounting for 32. 9% of total suppl. (Ministry of Agriculture, 2. , fragile domestic supply chains, and exchange rate ISSN: 2407-814X . 2527-9238 . Beef Price Volatility in Indonesia: Before. DuringA. (Mubarok et al. fluctuations (Hadi & Chung, 2. The pandemic further exacerbated these vulnerabilities, as social restriction policies . alled PPKM and PSBB in Indonesi. disrupted logistics, reduced beef production by 10. 18%, and triggered price spikes of up to 19. 9% in 2021 (Bai et al. Bairagi. Mishra, & Mottaleb, 2. Notably, beef prices remained consistently higher than those of other protein sources, such as chicken meat and eggs, throughout the pandemic. Price volatility patterns varied significantly across commodities. While chicken meat and egg prices followed similar trends, beef exhibited distinct fluctuations. Among the three, eggs displayed the highest volatility, with a Coefficient of Variation (CV) of 0. 12, compared to chicken meat . and beef . The pandemic period intensified these fluctuations: beef's CV rose to 0. 06Aitriple its pre- and post-pandemic stability level of 0. Similarly, chicken meat and eggs saw higher CVs during the pandemic . 08 and 0. 12, respectivel. compared to pre-pandemic levels . 07 for bot. and post-pandemic declines . 05 and 0. These findings underscore how large-scale restrictions and the transition to the "New Normal" amplified price instability across key food commodities. Given these dynamics, analyzing price volatility before, during, and after the pandemic is critical to understanding systemic vulnerabilities in IndonesiaAos beef supply chain. Such research can inform policies to enhance resilience against future shocks, ensuring more stable food prices and supply security. While several studies have explored beef price volatility, they have often done so in isolation, focusing on specific dimensions such as domestic production (Komalawati. Asmarantaka. Nurmalina, & Hakim, 2. , import policy, regional market patterns (Komalawati. Asmarantaka. Nurmalina, & Hakim, 2. , or global connectivity (Tanaka & Guo, 2. Few have investigated the full trajectory of volatility across three distinct periods, pre-pandemic, pandemic, and post-pandemic, using high-frequency . In fact, the post-pandemic period . 3Ae2. revealed a new pattern of volatility resulting from a combination of government interventions and economic recovery, which has yet to be systematically mapped. This study addresses research gaps by applying a multidimensional analytical framework using the GARCH model to assess the impact of external shocks . , the COVID-19 pandemi. relative to structural factors like import dependence and exchange rate. It also examines spillover effects from substitute commodities . hicken and egg. , using highfrequency daily data from the National Strategic Food Price Information Center (PIHPS) of Bank Indonesia . 7Ae2. to capture short-term volatility often missed in aggregate Post-pandemic data . 3Ae2. allows analysis of recovery-phase dynamics and policy impacts. The study aims to . analyze beef price volatility across pre-, during-, and postpandemic periods, . measure the relative roles of pandemic-related and structural shocks, and . provide long-term policy recommendations for price stabilization. Despite its comprehensive framework, this study has limitations, particularly the exclusion of climate-related variables due to unavailable daily-scale data, and a national-level focus aligned with macroeconomic indicators. Nonetheless, insights from global literature support interpretation of findings and policy relevance (Bozma. Urak. Bilgic, & Florkowski. Ultimately, this study enhances understanding of how external crises interact with ISSN: 2407-814X . 2527-9238 . Journal of Agribusiness 202 AGRARIS: and Rural Development Research structural vulnerabilities in shaping price volatility, highlighting that the pandemic amplified rather than caused instability. The use of high-frequency modeling emphasizes the need for anticipatory, multisectoral policy approaches aligned with resilient food system strategies (Montalbano & Romano, 2023. Namany. Govindan, & Al-Ansari, 2. RESEARCH METHOD This study employs HeatonAos . secondary data research methodology by utilizing institutional administrative records. The primary data for analyzing price volatility consist of daily consumer-level price data from July 10, 2017, to December 31, 2024, sourced from the National Strategic Food Price Information Center (PIHPS) of Bank Indonesia (Table . assess the influencing factors, additional monthly data from July 2017 to December 2024 were obtained from the Ministry of Trade. Bank Indonesia, and Statistics Indonesia (BPS) (Table The entire dataset is segmented into three periods: pre-pandemic (July 2017AeMarch 2. , pandemic (March 2020AeJune 2. , and post-pandemic . rom June 2023 onwar. , based on key government policy milestones such as the declaration of the national public health emergency, the implementation of Large-Scale Social Restrictions, and the end of the pandemic period through Presidential Decree No. 17 of 2023 (Wahyuni. Pujiharto. Azizah, & Zulfikar, 2. These temporal divisions allow for a more accurate analysis of price dynamics across different policy regimes. TABLE 1. TYPES AND SOURCES DATA No. Data Types Daily beef prices at the consumer level Data Sources National Strategic Food Price Information Center . alled as PIHPS in Indonesi. Bank Indonesia Daily chicken meat and egg prices at the National Strategic Food Price Information Center . alled as PIHPS in consumer level Indonesi. Bank Indonesia Exchange rate Bank Indonesia Inflation rate Bank Indonesia Import of Beef Central Bureau Statistics and Ministry of Trade Import of Feeder Cattle Central Bureau Statistics Beef production Central Bureau Statistics and National Food Agency Daily price data gaps due to holidays and weekends were addressed using an interpolation technique developed by Insukindro . Prior to analysis, the data were transformed using the natural logarithm function to reduce the influence of variability and stabilize the data (Raudys & Goldstein, 2. Beef price volatility was estimated using the GARCH model introduced by Bollerslev in 1986 (Komalawati et al. , 2. Before applying the GARCH model, the Augmented DickeyAeFuller (ADF) test was applied to ensure the stationarity of the mean process (Hassani. Yeganegi. Khan, & Silva, 2020. Pallotta & Ciciretti, 2. , while allowing for potentially non-stationary or near-integrated behavior in the conditional variance during periods of structural shocks. This test is essential to avoid spurious regression results, which may lead to biased estimations. Non-stationary data were differenced until stationarity was achieved. ISSN: 2407-814X . 2527-9238 . Beef Price Volatility in Indonesia: Before. DuringA. (Mubarok et al. Following the stationarity test, the appropriate mean equation was determined using autoregressive (AR), moving average (MA), or Autoregressive Integrated Moving Average (ARIMA) models. Model selection was based on the lowest values of the Akaike Information Criterion (AIC) and Schwarz Criterion (SC), as well as the highest log-likelihood value. The ARIMA model was required to satisfy several criteria to be deemed valid, including random residuals, parsimony, statistically significant parameter estimates, fulfillment of the stationarity condition . AR and MA coefficients less than on. , a convergent iteration process, and a low Mean Squared Error (MSE). The selected mean equation was then tested for heteroscedasticity using the ARCH-LM test (Lagrange Multiplier test for Autoregressive Conditional Heteroscedasticit. If the residuals exhibited heteroscedasticity, the model was subsequently estimated using the GARCH . This study employs the GARCH . model, which is widely used due to its ability to capture time-varying conditional variance (Yousef & Shehadeh, 2. The selection of GARCH . is strongly supported by the characteristics of the beef price time series, which visually exhibits a volatility clustering effect, where the GARCH . model is the most parsimonious and effective specification for capturing this effect (Joukar & Nahmens, 2. The GARCH . is specified as follows: E2t =0 1 A2t-1 1 E2t-1 Where yuayc2 represents the conditional variance in period t, yu0 is a constant, yuAycOe1 is the squared residual from the previous period . Ae. , and yuaycOe1 the conditional variance in period tAe1. Parameters yu1 and yu1 are the coefficient estimates of the GARCH terms, respectively. The coefficient reflects the ARCH effect and indicates the magnitude of short-term price volatility or how quickly volatility responds to recent shocks . , policy changes or supply disruption. (Engle, 1. The value of measures the impact of past volatility on current conditional volatility (Tanaka & Guo, 2. Higher value will suggest long-lasting volatility persistence, meaning prices struggle to stabilize even after the crisis. A value of close to one implies high volatility persistence over time. It is important to emphasize that the interpretation of the GARCH. parameters in this study focuses on volatility persistence rather than strict covariance stationarity. During periods characterized by large structural shocks, such as the COVID-19 pandemic and the post-pandemic adjustment phase, the conventional stationarity condition ( < . may not necessarily hold. In such contexts, values of approaching or exceeding unity reflect prolonged volatility persistence and slow dissipation of shocks rather than model Such volatility dynamics are commonly observed in markets affected by major structural changes, where non-stationary or near-integrated variance processes capture economically meaningful shifts in market conditions (Campos-Martins & Amado, 2. Previous studies further show that commodity markets exposed to crisis-driven disruptions exhibit highly persistent volatility dynamics that are economically relevant for understanding market instability and informing policy responses, particularly during periods of heightened uncertainty (Khan. Kayani. Khan. Mughal, & Haseeb, 2023. Zavadska. Morales, & Coughlan. Accordingly, the GARCH model in this study is employed as a volatility ISSN: 2407-814X . 2527-9238 . Journal of Agribusiness 204 AGRARIS: and Rural Development Research characterization and policy-relevant analytical tool, rather than as a strict second-moment stationary process (Horvyth. Trapani, & Wang, 2. The GARCH model is estimated using the Maximum Likelihood method and processed with EViews 9. To analyze the factors influencing beef price volatility in Indonesia, monthly data from July 2017 to December 2024 are used. The daily price and volatility data are aggregated into monthly data to enable further analysis. The relationship between influencing factors and volatility is estimated using a logarithmic regression model that include lagged endogenous and exogenous variables. The lagged variables included because agricultural markets require some periods to adjust from shocks in other markets or transport delays (Morales, 2. The formulation of the model used to examine the determinants of beef price volatility is as ycoycu. uayc ) = yuE0 yuE1 ycoycu . cEycaycOe1 ) yuE2 ycoycu. uaycOe4 ) yuE3 ycoycu. uaycOe24 ) yuE4 ycoycu. aycAycOe1 ) yuE5 ycoycu. ayaycOe4 ) yuE6 ycoycu. cEycOe1 ) yuE7 ycoycu. aycIyc ) yuE8 ycoycu. aycIyc ) yuE9 ycoycu. cEycaycOe1 ) yuE10 ycoycu. cEyceycOe1 ) yuE11 ycoycu. uaycaycOe1 ) yuE12 ycoycu. uayceycOe1 ) yuE13 yayceyce yuE14 yayceyca yuE15 yaycyyca yuNyc Expected sign 1 ,7 ,8 ,9 ,10 ,11 ,12 ,13 ,15 >0. 2 , 3 ,4 ,5 ,6 , 14 <0 In this model, ln(Et ) represents the natural logarithm of beef price volatility in period t, calculated from daily price fluctuations aggregated into monthly data. ln (Pbt-1 ) is the natural logarithm of the beef price in the previous month, while ln(Et-4 ) and ln(Et-24 ) denote the natural logarithms of beef price volatility at 4-month and 24-month lags, respectively. ln(IMt-1 ) refers to the natural logarithm of beef import volume in the previous month . onnes/mont. , and ln(FCt-4 ) represents the natural logarithm of feeder cattle imports four months earlier . ead/mont. ln(Qt-1 ) is the natural logarithm of beef production in the previous month . onnes/mont. ln(ERt ) indicates the natural logarithm of the exchange rate (IDR/USD), and ln(IRt ) is the natural logarithm of the inflation rate . ln(Pct-1 ) and ln(Pet-1 ) are the natural logarithms of chicken meat and egg prices in the previous month, respectively. ln(Ect-1 ) and ln(Eet-1 ) denote the natural logarithms of chicken meat and egg price volatility in the prior month. The model also includes dummy variables: Def for Ramadan and Eid al-Fitr. Dea for Eid al-Adha, and Dpc for the COVID-19 pandemic AAt is the error term, and 1 -15 are the estimated coefficients for each explanatory This log-log model allows the interpretation of estimated coefficients as elasticities, showing the percentage change in beef price volatility resulting from a one-percent change in each independent variable. Estimation is conducted using Ordinary Least Squares (OLS) and processed with EViews 9 software. RESULTS AND DISCUSSION Beef Price Volatility and Determinant Factors The prices of beef from 2017 to 2024, the prices of beef before, during, and after pandemic were then tested for stationarity using the Augmented Dickey-Fuller (ADF) test. ISSN: 2407-814X . 2527-9238 . Beef Price Volatility in Indonesia: Before. DuringA. (Mubarok et al. Table 2 presents the results of the unit root tests. Based on the table, most variables used in the volatility analysis were found to be stationary at first difference, except for price of beef after pandemic, as indicated by ADF test statistics that exceed the 1% critical value. TABLE 2 RESULTS OF UNIT ROOT TESTS USING AUGMENTED DICKEY-FULLER (ADF) TEST Variables Beef price (Pb ) Beef price before pandemic (Pbp ) Beef price during pandemic (Ppc ) Beef price post-pandemic . ) Level 84 . First Difference 70 . Critical values 1% The stationarity test was followed by the determination of the appropriate mean model. Table 3 presents the best mean equations used to estimate beef price volatility for the overall period . 7Ae2. , as well as for the subperiods before, during, and after the COVID-19 The results indicate that each time series segment required a different ARIMA The best-fitting ARIMA model for the overall 2017Ae2024 beef price volatility data is ARIMA . For the pre-pandemic period, the best model is MA. for the pandemic period, it is ARIMA. and for the post-pandemic period. ARIMA. is the most suitable. All four ARIMA models were subsequently tested for heteroscedasticity using the ARCH-LM test, which produced significant probability values . < 0. These results indicate the presence of ARCH effects in the residuals of the mean equations, confirming that the models are suitable for further volatility analysis using the GARCH approach. TABLE 3. THE ESTIMATION RESULTS OF BEEF PRICE VOLATILITY ANALYSIS Price Volatility Pre-pandemic price Pandemic period price volatility Post-pandemic volatility price Mean Model 00 . AR( . AR( . MA( . MA( . GARCH Model 31E-06 . 30E-07 47E-06 . 75E-07 . yu 11 . Log-likelihood AIC SIC ARCH-LM test 23 . Note: yu is the ARCH effect . hort-term shock responsivenes. yu is the GARCH effect . olatility persistenc. yu yu is the degree of long-term volatility persistence. (A) is the probability values Table 3 also presents the estimated parameters of the GARCH . models for beef price volatility across the full period . 7Ae2. , as well as the pre-, during-, and postISSN: 2407-814X . 2527-9238 . Journal of Agribusiness 206 AGRARIS: and Rural Development Research pandemic periods. The ARCH-LM test conducted on the GARCH . model returned nonsignificant F-statistic and Chi-square values . > 0. , indicating no remaining ARCH effects and confirming that the model is well-specified. The estimation results of the GARCH . model over the entire period . 7Ae2. indicate the ARCH coefficient (), which captures the short-term responsiveness of volatility to new shocks, is relatively large ( = 3. , suggesting that beef prices were highly sensitive to unexpected disturbances during this period. This shock sensitivity was particularly pronounced during the pandemic period, when increased to 1. 89, reflecting strong shortterm volatility responses associated with sudden supply chain disruptions and trade Prior to the pandemic, the estimated value was more moderate ( = 0. indicating relatively lower sensitivity to short-run shocks. In the post-pandemic period, declined substantially ( = 0. , implying that immediate price reactions to new shocks However, this does not necessarily indicate lower overall volatility, as volatility persistence remained high through the dominance of the GARCH component (). Across the full sample and during the pre-pandemic and pandemic periods, the predominance of relative to is consistent with volatility dynamics driven mainly by exogenous shocks rather than gradual variance accumulation, particularly during the height of market disruptions in 2020Ae2022. These results highlight the importance of distinguishing between short-term shock effects () and long-term volatility persistence () when interpreting GARCH-based volatility dynamics. The GARCH coefficient () represents the impact of past conditional variance on current volatility (Tanaka & Guo, 2. This study finds that the post-pandemic period is characterized by a value approaching 1, suggesting that price volatility remained persistent even after the initial shock had subsided. Moreover, during the post-pandemic period, was lower than , implying that variance persistence played a more dominant role than immediate external market shocks in influencing beef price movement. The sum of and reflects the degree of long-term volatility persistence. Among the four estimated models, only the post-pandemic period exhibited values approaching 1, indicating near-integrated volatility persistence. This finding aligns with previous studies by Zavadska et al. , which suggest that values of close to 1 imply a high risk of prolonged volatility. It suggests persistent volatility, meaning prices remain unstable post-crisis or beef prices may stay volatile in the long-term (Komalawati et al. , 2. due to structural weaknesses like import dependency and supply chain inefficiencies (Shobur et al. , 2. The high level of volatility persistence in beef is consistent with the findings of (Pipit. Pranoto, & Evahelda, 2. The results of the volatility estimation are illustrated in Figure 1, which shows the movement of beef price volatility in Indonesia from July 10, 2017 to December 31, 2024. The graph represents daily beef price volatility as the conditional standard deviation . , the square root of the conditional varianc. The pattern shown in Figure 1 is consistent with the GARCH model estimation results discussed above: beef price volatility was considerably ISSN: 2407-814X . 2527-9238 . Beef Price Volatility in Indonesia: Before. DuringA. (Mubarok et al. higher during the pandemic period compared to the pre-pandemic . 7Ae2. and postpandemic . 3Ae2. Note: Volatility is measured using conditional standard deviation and the values range from 0. 00000486 to 0. 01079, with most observations concentrated below 0. orange line shows the border of period FIGURE 1. DAILY BEEF PRICE VOLATILITY IN INDONESIA TABLE 4. FACTORS AFFECTING BEEF PRICE VOLATILITY IN INDONESIA Variables Coefficient Probability value Constanta Beef prices Beef price volatility t-1 Beef price volatility t-24 Imported beef t-1 Imported feeder cattle t-4 Production t-1 Exchange rate Inflation rate Chicken meat prices t-1 Egg prices t-1 Chicken meat price volatility t-1 Egg price volatility t-1 Ramadan and Eid al-Fitr Eid al-Adha COVID-19 pandemic 60*** R-squared Adjusted R-squared of regression F-statistic Prob(F-statisti. Durbin-Watson stat Note: Dependent variable = beef price volatility at period t. ***, **, * = significant at 0. 01, 0. 05, and 0. 1 respectively Table 4 presents the factors influencing beef price volatility, with an R-squared value of This indicates that approximately 61. 73% of the variation in beef price volatility in ISSN: 2407-814X . 2527-9238 . Journal of Agribusiness 208 AGRARIS: and Rural Development Research Indonesia can be explained by the independent variables included in the model. The moderate R-squared value reflects the inherent complexity of the factors that drive price volatility, which often involve a wide range of dynamic, interrelated, and mutually exclusive phenomena that cannot be captured by a single or limited set of variables (Algieri, 2021. Kieu. Luu, & Yoon, 2. In addition, the adjusted R-squared result of 0. 5024 suggests that around 24% of the variation in the dependent variable . eef price volatilit. can be explained by the independent variables, after adjusting for the number of explanatory variables used in the The analysis confirms that the COVID-19 pandemic significantly increased beef price volatility in Indonesia . oefficient = 1. p < 0. , primarily due to disruptions in supply chains, consumption shifts, and economic stress, consistent with studies in Turkey (Bozma et , 2. and the U. (Ramsey. Goodwin. Hahn, & Holt, 2. , although not all studies agree, likely due to differing data coverage (Hermawan et al. , 2. The R-squared value of 617 indicates that the independent variables in the model explain approximately 61. 7% of the variation in beef price volatility, while the remaining proportion is attributable to other factors not included in the model. Seasonal events also play a role: volatility spikes during Ramadan and Eid al-Fitr . oefficient = 1. p < 0. due to demand surges (Komalawati et al. , 2. , while Eid alAdha reduces volatility . oefficient = Ae1. p < 0. through increased beef supply, aligning with (Dewia. Nurmalina. Adhi, & Brymmer, 2. Volatility is also influenced by historical trends and substitution effects (Javadi. Ghahremanzadeh, & Soumeh, 2024. Zumbach, 2. Previous price volatility and the price and volatility of chicken (Anwar et al. , 2023. Xie. Zhu. Liu. Ye, & Liu, 2. at lag tAe1 . oefficient = 1. p < 0. significantly affect beef volatility, indicating strong inter-commodity linkages. A negative effect at lag tAe4 reflects market adjustment tied to cattle fattening cycles (Komalawati et al. , 2. , while lag-24 volatility is not significant. Discussion Beef price volatility is typically driven by climate-related production shocks (Godde. Mason-DAoCroz. Mayberry. Thornton, & Herrero, 2. , shifting consumption patterns (Magalhaes et al. , 2. , and currency fluctuations (Hadi & Chung, 2. The pandemic exacerbated these issues, amplifying volatility. Import policy instability and global market uncertainty (Du & Dong, 2. often have longer-lasting impacts than health crises alone. Unpredictable price swings threaten food security and political stability, as evidenced in Ethiopia and Ghana (Wossen. Berger. Haile, & Troost, 2. This study confirms persistent beef price volatility post-pandemic, with low but OO 1 in GARCH models. Heavy reliance on Australian imports intensified shocks and price surges 9% . and 24% . followed AustraliaAos post-drought recovery directly impacted Indonesia, a major importer of Australian beef. Strategic solutions include import diversification, domestic production strengthening . , feedlot. , and strategic reserves. Pandemic restrictions significantly disrupted supply chains . < 0. (Monge & Lazcano. ISSN: 2407-814X . 2527-9238 . Beef Price Volatility in Indonesia: Before. DuringA. (Mubarok et al. , revealing fragile logistics and fragmented information systems (Peel, 2. Recommendations include digital traceability and regional logistics hubs (Ijaz et al. , 2021. Sumrow. Hudson. Sarasty. Carpio, & Bratcher, 2. Other contributors to volatility include spillover effects from substitute commodities like chicken and eggs (Hermawan et al. , 2. , indicating consumer sensitivity and crosscommodity linkages. Lagged effects . our-period lag of beef price volatilit. and seasonal events like Ramadan affect beef prices, justifying a seasonal intervention calendar to coordinate geographically targeted actions. Domestic production showed weak response to price signals . > 0. , underscoring the need for resilience-focused policies . rice guarantee. The pandemic exposed systemic vulnerabilities, calling for machine learning-based early warning systems and scenario-based policy simulations. Though GARCH . captured volatility persistence (Adj. RA = 0. , future research should explore non-linear models, spatial econometrics, and non-pandemic shocks . eopolitics, climat. , as well as expand the scope to other key food commodities such as rice, cooking oil, chili, and seafood, to evaluate whether similar volatility patterns apply across markets. CONCLUSION This study finds that the COVID-19 pandemic served as a major shock that significantly increased beef price volatility in Indonesia. However, volatility did not subside in the postpandemic period, indicating that structural weaknesses, particularly import dependency and inefficient supply chains, are the dominant drivers of persistent price instability. The pandemic acted more as an amplifier than a root cause, reflecting persistent volatility dynamics rather than transitory shocks. Volatility was also shaped by seasonal demand surges (Eid al-Fitr. Eid al-Adh. , substitution effects from other protein sources . , and delayed responses from domestic production. The prolonged nature of volatility highlights a vulnerability within Indonesia's beef supply chain that requires urgent structural reforms beyond reactive crisis Policy actions must therefore focus not only on crisis mitigation but also on long-term These include the development of a real-time price monitoring system, strategic import diversification, and investment in domestic production infrastructure. Seasonal price stabilization programs and predictive analytics using big data should be institutionalized to anticipate future shocks and safeguard food security. Acknowledgments: The authors would like to express our deepest gratitude to The Agribusiness Study Program. Faculty of Animal Husbandry and Agriculture. Diponegoro University, for providing the necessary research facilities and academic support, and all parties who contributed directly or indirectly to the completion of this work. AuthorsAo Contributions: AFM: conceptualization ideas, collecting data, processing and analyzing data, writing original manuscript, editing the manuscript. KK: research design. ISSN: 2407-814X . 2527-9238 . Journal of Agribusiness 210 AGRARIS: and Rural Development Research conceptualization ideas, processing and analyzing data, reviewing and editing the manuscript. AS: research design, conceptualization ideas, finalizing and reviewing the manuscript. Conflict of Interest: The authors declare no conflict of interest. REFERENCES