GLOBAL GOAT MEAT PRODUCTION: A TIME SERIES ANALYSIS OF CONCENTRATION TRENDS IN LEADING COUNTRIES

INTRODUCTION

 Food security and nutrition are becoming an increasingly important element of people’s basic needs as the world’s population continues to grow. As the population is increasing the demand of animal derived proteins is also increasing and thus to have a healthy life it is necessary to have a good quality protein intake. This is ensured by animal feeds with proteins of high biological value, essential amino acids and digestibility. However, it is not enough with the intake of animal protein in many countries due to socio-economic, food accessibility and ethical issues (Kurtoğlu and Uzundumlu, 2023). With the increase in world population and the need to properly feed people, it is necessary to diversify the sources of proteins and optimize sustainable processes to produce them. This is especially true in developing countries where these premium proteins are available from some 7.46 million tons of goat meat from 1.13 billion goats (FAOSTAT, 2025).

 Goats (Capra hircus) are represented by about 600 breeds and are important for the livelihoods of millions of people in different climate zones, showing their versatility in desert, mountainous and tropical regions by providing the same amount of meat, milk, hides, fiber and manure with minimal input (Danso et al., 2024). They exhibit greater resilience than other larger farm animals in the face of climate change, and they are environmentally sustainable as they use less water and feed (Navarrete-Molina et al., 2024). Goat meat is a good source of high-quality protein, iron, zinc, and vitamin B12 that helps human immune system, energy metabolism and muscle development (Sindhura et al., 2023).

 The world’s livestock production has recently undergone a profound structural change due to rapid urbanization and the integration of science and technology. This evolution in the 21st century is increasingly characterized by larger scale operations and more professionalized enterprises. In this transformation, trade agreements promoting supply, demand and trade, led to an increase in the share of meat and meat products in total agricultural and food exports, from 6% in 1960 to 12% in the 2010s (Serrano and Pinilla, 2014). With increasing per capita income and urbanization rates, the demand for meat and meat products has increased over time, first in the West and later in emerging economies. (Chung et al., 2020). Demand for animal-based diets rises with increasing incomes and hence a shift in preference (Law et al. 2020). According to global projections of the OECD-FAO (2021), global consumption of meat, eggs and milk is expected to rise by 23-30% between 2000 and 2020, with increases of 2-6% for developed countries and up to 60% for developing countries.

According to data from 2024, beef accounted for 72.41% of global red meat production, lamb for 11.98%, goat for 8.03%, and buffalo for 7.58%. In that year, world beef production reached 69.75 million tons, with per capita production at 8.60 kg. Argentina led the way with 69.54 kg, followed by Brazil with 48.29 kg and the USA with 35.58 kg. In the same year, global lamb production reached 11.54 million tons, with per capita production at 1.42 kg. In Mongolia per capita production amounted to 38.35 kg, in Iceland to 20.25 kg and in Turkmenistan to 16.53 kg. In 2024, the world goat meat production was 7.74 million tons and the per capita consumption of goat meat was 0.95 kg worldwide. Mongolia came in first with 18.35 kg, followed by Chad with 8.16 kg, UAE with 5.42 kg, Central African Republic with 4.08 kg, and Mauritania with 3.68 kg (FAOSTAT, 2025). Countries with high levels of production also have high levels of consumption. Goats are important source of protein in semi-arid and desert areas where growth of cattle is limited. They are especially raised for a source of livelihood.

 Meat is usually preferred to be beef rather than lamb and goat meat. This is mainly due to the characteristic odor of lamb and goat meat, which can be affected by factors such as breed, diet and genetics (de Lima Júniore et al., 2016).

 According to data from 2024 FAO, the major producers of goat meat are China (32.77%), India (22.25%), Pakistan (7.25%), Nigeria (3.77%) and Bangladesh (3.15%). The data shows that the world production of goat meat increased from 1.1 million tons in 1961 to 7.5 million tons in 2024. The increase is 6.78 times. In the 2020s, China, India and Pakistan together accounted for 65% of the global sheep population (FAOSTAT, 2025), this share was 30% in the 1960s, 50% in the 1990s and 60% in the 2010s.

 In recent years, researchers have predicted the future production of sheep or goats or their products using different forecasting models. Nair et al. (2021) predicted the small ruminants population of India using the ARIMA model. Ordu and Zengin (2020) used ARIMA, ES and STLF models to predict the red meat production in Türkiye in the best way. Besides, Nimbalkar et al. (2019) developed a model on Indian goat meat export. Palabicak, and Binici (2023) investigated the sheep production in Turkey using the ARIMA model. (Primawati et al., 2023) employed LSTM and Prophet as advanced methods of predicting the production of goat’s milk with data from two farms in Indonesia.

 This paper is an analysis of the structural concentration in goat meat production, using concentration indicators (CR and HHI) and the forecast of its future evolution by ARIMA time series.  The analysis of the five largest producers (China, India, Pakistan, Nigeria and Bangladesh) and prediction of goat meat production in 2025–2030 based on 1961–2024 data is unique and useful to understand the dynamics of the national and global production and their potential effect on resilience in protein supply.

MATERIALS AND METHODS

 The basic goat production data used in this study were obtained from the Food and Agriculture Organization (FAO). We also used various national and international articles and studies to support our findings and enrich the discussion.

 In this study, goat meat production quantities obtained from the FAO were calculated for 10-year periods according to the HHI and CR₅ indices, and the market structure was examined. The ARIMA model was applied with SAS to produce predictions on goat meat output between 2025 and 2030. The data utilized included time series data from 1961 to 2024. But because certain countries didn't have enough data, projections were created using both the ARIMA and ARIMAX methods. Dummy variables were included in the model for Pakistan in 1996, 2001, and 2006, for Nigeria in 2011, and for Bangladesh in 1980 and 2008. We chose the ARIMAX model because it did a better job of making predictions, especially when it came to the model's application requirements, such as how accurate and reliable it was at predicting future trends compared to the normal ARIMA technique. Studies with the ARIMA and ARIMAX models were conducted using the SAS 9.4 statistical program, and the Microsoft Excel program was used for preparing figures and tables and performing the necessary calculations.

Classification of market structures with HHI and CR: We calculated the HHI and CR for each ten-year period to measure the production-based competition levels of countries. The following table (Krugman and Wells, 2021; Karakaya and Uzundumlu, 2025) classifies market structures according to the HHI and CR values.

Table 1. Market structures according to HHI and CR

Source: Krugman and Wells (2021).

HHI, HHI^-1 and CR calculation formulas: Some formulas were used in the HHI, HHI^-1 and CR calculations for countries such as China, India, Pakistan, Nigeria and Bangladesh. Pavic et al. (2016) and Karakaya and Uzundumlu (2025) used the following formulas for HHI and HHI^-1.

(1)

(2)

  (3)

i= number of producer or exporter countries

n= number of n countries considered in the competition, in this study n was taken as five.

MS₁ = Percentage share of the country ranked first in world production or export

MS₂ = Percentage share of the country ranked second in world production or export

MS₃ = Percentage share of the country ranked third in world production or export

MS₄ = Percentage share of the country ranked fourth in world production or export

MS₅ = Percentage share of the country ranked fifth in world production or export

ARIMA and ARIMAX Models: The Autoregressive Integrated Moving Average (ARIMA) model is a statistical method used to analyze time series data from past years to make predictions for future periods (Filder et al., 2019). Developed by Box and Jenkins in 1970, this model yields successful results, especially in short-term forecasts, compared to more complex models (Ariyo et al., 2014). When applied to stationary data, it can generate predictions of future values through specific statistical processes. ARIMA is widely used, particularly for the time series analysis of economic indicators (Box et al., 2015). The model's high accuracy in future predictions, ability to apply exponential smoothing methods, and structured training process incorporating the Box-Jenkins methodology have made it one of the prominent methods in the literature (Ospina et al., 2023).

 The general representation of the models is ARIMA (p, d, q). Here, p and q are the degrees of the autoregressive (AR) and moving average (MA) models, respectively, and d is the degree of differencing. An ARIMA model operates under the assumption of stationarity, meaning that the time series has a constant mean and variance over time. If the series is not stationary, the d in ARIMA refers to the differencing process. Because the first difference of the series is taken with respect to the previous year, the t₁ series consists of one less number of values compared to the t series. If the series is still not stationary, differencing can be applied multiple times, and the result is reflected in the d value in the ARIMA (𝑝,,) model (Karakaya and Uzundumlu, 2025).

 The ARIMAX model is an extension of the ARIMA model, which is a statistical method used for time series forecasting. The X at the end represents additional independent variables, “exogenous variables,” that can influence the forecast. The model includes additional independent variables, represented by the X at the end, which stands for exogenous variables (Adu et al., 2023). In the literature, the ARIMAX model is typically represented as ARIMAX (p,d,q) with exogenous variables (X), signifying that the conventional ARIMA (p,d,q) framework is augmented by the inclusion of external explanatory variables.

d=0: y_t = Y_t      (4)

d=1: y_t = Y_t - Y_t-1     (5)

d=2: y_t = (Y_t - Y_t-1) - (Y_t-1 - Y_t-2) = Y_t - 2Y_t-1 + Y_t-2  (6)

 AR model defines the estimation process by relating the current value in the time series to its past values. A p-order autoregressive model, AR(p), also known as an ARIMA (p,0,0) model, has the following general form (Mangiwa et al., 2025):

  (7)

Using the backshift operator, the model is expressed as follows:

(8)

 The Moving Average (MA) model is an approach that assumes the value of a time series at a specific point in time is influenced by the error term at that moment and the weighted error terms from previous periods. A q-order moving average model, MA(q), or ARIMA (0,0,q) model, is generally expressed as follows:

  (9)

The model can also be written using the backshift operator as follows:

(10)

 The Autoregressive Moving Average model (ARMA (p,q)) is a time series model formed by the combination of AR and MA models without a differencing operation. Its general form is expressed as follows:

  (11)

The model can also be written with the backshift operator as follows:

   (12)

The ARIMA(p,d,q) model is a generalized form of the ARMA(p,q) model for the analysis of non-stationary time series. In this model, data that do not exhibit stationarity are made stationary through a d-order differencing operation.

  (13)

 The ARIMAX(p,d,q) model further extends the ARIMA model by incorporating exogenous variables that may influence the dependent time series. The general form of the ARIMAX model is expressed as:

 (14)

The mathematical expression of the ARIMA and ARIMAX model is as follows:

Yₜ : value of the variable at time t

Xₜ : exogenous (external) variable affecting the time series

Yₜ₋ᵢ: past values at time t–i (i = 1, 2, ..., p)

φᵢ: estimated coefficients of the AR model for the i^th lag period

θ_i: estimated coefficients of the MA model for the i^th lag period

β: coefficient of the exogenous variable

φₚ(B) : p-order autoregressive (AR) polynomial

θ_q(B) : q-order moving average (MA) polynomial

d: order of differencing required to achieve stationarity

(1 − B)ᵈ : d-order differencing operator

B : backshift operator (for example, BYₜ = Yₜ₋₁)

ε_ₜ : error term (white noise) at time t.

 The SAS statistical program is used to perform the following ARIMA or ARIMAX model steps:

1. Data Collection and Organization: For scientific research to be accurate and data integrity to be ensured, the data must be accurate, complete, and understandable (Miller and Spiegel, 2025). In this study, FAOSTAT data were checked for each country and organized in Excel format.

2. Stationarity Test: ARIMA models require stationary data, and therefore the mean, variance, and autocorrelation structure of the data should not vary over the same period (Bawdekar et al., 2022; Khalili, 2025). Graphically, data is stationary when there is no continuous trend and early lags in the ACF and PACF graphs are high and diminish rapidly after multiple lags (Peng et al., 2022). Moreover, unit root tests like Augmented Dickey–Fuller (ADF), Phillips–Perron (PP), and Kwiatkowski–Phillips–Schmidt–Shin (KPSS) can determine data stationariness (Ellis, 2025). In this study, the stationarity of the data was determined by examining the trend graph, ACF and PACF graphs, and ADF tests.

3. Model Setup: In ARIMA modeling, the SAS statistical program systematically identifies the most suitable alternative models using the Minimum Canonical Correlation Analysis (SCAN) and Extended Sample Autocorrelation Function (ESACF) methods (SAS 2025). In this process, different combinations of p and q values ​​are tested separately, and the model providing the highest fit for each prediction scenario is selected.

4. Model Selection: The Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) were used to compare the best models (Karakaya and Uzundumlu, 2025). These criteria are widely used to evaluate the performance of a model by balancing its suitability and complexity. In this study, the selection of the most suitable model was based not only on AIC and BIC values but also on the prediction deviations for the 2019–2024 period. Accordingly, the model with the lowest information criterion values ​​and the smallest deviation in the 2019–2024 predictions was determined to be the most suitable model.

5. Diagnosis and Control: The Autocorrelation Function (ACF) and Partial Autocorrelation Function (PACF) are critical in determining the appropriate parameters of the ARIMA model (Peychinov et al., 2025). To evaluate the accuracy of the model, ACF, PACF, and Inverse Autocorrelation Function (IACF) analyses were performed on the residuals to check whether there was any dependency structure in the residuals of the selected model and whether the residuals exhibited white noise characteristics (Mishra et al. 2021). The presence of white noise characteristics in the residuals indicates that the model is valid and reliable, while in cases where this characteristic is not present, the model may need to be re-engineered to obtain more accurate predictions. We used the conditional least squares (CLS) estimation method to determine how statistically significant the model parameters and model coefficients were. We also used autocorrelation checks of residuals to determine the reliability of our model, the Ljung-Box test to determine if the residuals behaved like white noise, and residual normality tests to check for normality.

 In this case, alternative models are considered, and if this problem exists in other models as well, alternative models are tested considering ACF and PACF analyses to determine the best models. The MA component was identified using the ACF graph, and the AR component was identified using the PACF graph.

6. Interpretation and Evaluation: Based on the data obtained, goat meat production estimates for the 2025–2030 period were made according to the selected models, and the resulting values ​​were graphically visualized and interpreted using Microsoft Excel.

RESULTS

Model Selection and Diagnostic Analysis for Global and Leading Producers: In this section, we studied different models associated with ARIMA and ARIMAX to estimate goat meat production and tried to find the models that produced the best results for our data. We used the AIC and BIC criteria to determine the best-fitting models, keeping the models as simple as possible. To understand how close our predictions were to reality, we studied the deviation between the predictions of the years 2019-2024 and the actual values by adding dummy variables for the periods where breaks occurred, and we reduced the deviations with the ARIMAX model analysis. In addition to AIC, BIC, and actual value prediction control, we employed diagnostic tests to verify residual normality, autocorrelation, and white noise and confirmed that the residuals were normal and independent. Therefore, we concluded that our models are robust and that the 2030 projections are reliable and valid. The selected models (Table 2) performed better than all other alternatives considered for the world, the five leading producers, and other countries.

ARIMA model selection for world goat meat production forecasting

Table 2. Ranking Criterion Tests Based on p and q Values (AIC, BIC, DW, and MFE) for the World

According to the lowest values of AIC and BIC and passing all tests, the best model of China and the world turned out to be ARIMA(1,1,1). In India, the ARIMA(1,1,2) model had the lowest AIC and BIC, but it was not selected because the ARIMA(1,1,6) model passed all the diagnostic tests. The best model for Pakistan, Bangladesh, and Nigeria was ARIMAX (1,1,0) in the sense that it passed all diagnostic tests and had lower deviations. The inclusion of dummy variables to capture structural breaks was statistically justified (p < 0.05) in these models. This indicates that these models are effective in capturing the overall data pattern and the structural changes in these countries. The best model for the rest of the world was ARIMA (3,1,3) and this was confirmed with diagnostic tests.

World goat meat production projections: Figure 1 shows the goat meat production quantities worldwide for the period from 1961 to 2030 and the projections for this production.

Figure 1. World goat meat production and production estimates between 1961–2030 (1000 tons).

 From the 1960s to the 1970s, global goat meat production fluctuated between 1.1 and 1.3 million tons. Since the 1980s, factors such as population growth, rural development initiatives, and increased demand for animal protein in developing countries have led to a rise in global goat production to 1.8 million tons. In the 1990s, with the dissolution of the Soviet Union and Yugoslavia and the incorporation of newly independent countries, goat meat production reached 2.7 million tons annually. In the 2000s, driven by population growth and increased consumer awareness through advertising and other activities, global goat production rose to 5.4 million tons by 2010. In 2024, global production is expected to reach 7.74 million tons, approximately 7.03 times higher than in the past 64 years, with an average annual growth rate of 3.14%. The production is expected to be stable with a steady increase from 7.89 million tons in 2025 to 8.81 million tons in 2030, with an average production of 8.37 million tons in the 6 years and a compound annual growth rate of 2.19% on the basis of ARIMA (1,1,1).

China goat meat production projections: Goat meat production quantities in China for the period from 1961 to 2030 and projections for this production are presented in Figure 2.

Figure 2. China's goat meat production and estimated production quantities between 1961–2030 (1000 tons).

 China's goat meat production has increased significantly each decade from 1961 to 2020, driven by technological advancements. Production rose from 40,000 tons to 95,000 tons between 1961 and 1970, and then to 200,000 tons by 1980. Growth accelerated after 1987, reaching 520,000 tons between 1981 and 1990, and then doubling to 1.29 million tons between 1991 and 2000. From 2001 to 2010, production went up to 1.99 million tons, and from 2011 to 2020, it went up to 2.41 million tons. Over 63 years, goat meat production has experienced a compound annual growth of 6.61%, resulting in a production of 2.54 million tons in 2024, 56.37 times the production in 1961. According to the ARIMA (1,1,1) model, production is projected to increase by 0.72% annually from 2024 to 2030, reaching 2.65 million tons in 2030, and the average production between 2025 and 2030 is projected to be 2.60 million tons.

India goat meat production projections: Figure 3 shows India's goat meat production quantities and production estimates for the period between 1961–2030.

Figure 3. India's goat meat production and estimated production quantities between 1961–2030 (1000 tons).

 Goat meat production in India has steadily increased from 1961 to 2020, rising from approximately 240,000 tons in the 1960s to 300,000 tons in the 1970s, 400,000 tons in the 1980s, and approximately 0.5 million tons in the 1990s. Between 2001 and 2010, it reached 700,000 tons and then 1.2 million tons in the 2020s. The compound annual growth rate (CAGR) is 3.13%, and it will reach 1.72 million tons in 2024, which is 7.33 times the production level in 1961. Estimates by ARIMA (1,1,6) show that production will gradually increase from 1.81 million tons in 2025 to 2.35 million tons in 2030, growing at a CAGR of 5.33%. Production is projected to average 2.03 million tons during this period.

Goat meat production and projections in Pakistan: Figure 4 shows the goat meat production quantities in Pakistan for the 1961–2030 period and estimates for this production.

Figure 4. Pakistan's goat meat production and estimated production quantities between 1961–2030 (1000 tons).

 Goat meat production in Pakistan increased steadily from 1961 to 2020. The figure increased from 52.7–78.0 metric tons in the 1960s to 164.0 metric tons in the 1970s and 295.0 metric tons in the 1980s. In 1995, production was close to 435 thousand tons, but by 2000, it had dropped to 300 thousand tons. In the 2000s, a comparable trend was noticeable, with a high of 496 thousand tons in 2004 and a subsequent drop to 380 thousand tons in 2010. Production bounced back in the 2020s, increasing at a CAGR of 3.85% to 561 thousand tons in 2024, or around 10.83 times the 1961 level. With a predicted compound yearly growth rate of 0.93%, ARIMAX(1,1,0) forecasts indicate that production would rise gradually to 634 thousand tons by 2030. The projected average production between 2025 and 2030 is 572 thousand tons.

Goat meat production and projections in Bangladesh: Figure 5 shows Bangladesh's goat meat production quantities and production estimates for the period between 1961–2030.

Figure 5. Bangladesh's goat meat production and estimated production quantities between 1961–2030 (1000 tons).

 Bangladesh produced around 224 thousand tons of goat meat in the early 2020s, up from 30 thousand tons in the 1960s. Achievements include 43 thousand tons in the 1970s, 73 thousand tons in the 1980s, and 127 thousand tons in the 1990s. By 2024, production had increased 9.17-fold from 1961 to 243,825 tons, a 3.58% compound annual growth rate. ARIMAX (1,1,0) projects production will reach 267,882 tons by 2030 with a 1.58% compound annual growth rate from 2024 to 2030. This period averaged 257,482 tons of production.

Nigeria's goat meat production and projections: Figure 6 shows Nigeria's goat meat production quantities and production estimates for the period between 1961–2030.

Figure 6. Nigeria's goat meat production and estimated production quantities between 1961–2030 (1000 tons).

 Nigeria's goat meat production increased from 5-15 thousand tons in the 1960s to almost 60 thousand tons in the 1970s and more than 120 thousand tons in the 1980s.  It grew steadily, reaching 221 thousand tons in the 1990s and 287 thousand tons in the 2000s, with some oscillations in succeeding years.  By 2024, production had reached 292,406 tons, or 51.16 times the 1961 level, with a compound annual growth rate (CAGR) of 6.45%.  ARIMAX (1,1,0) model anticipates a steady climb to 320.917 tons by 2030, with an average annual output of 307.770 tons and a predicted CAGR of 1.56% from 2024 to 2030.

Goat meat production and projections of Other Countries: Figure 7 shows the goat meat production quantities and production estimates of other countries for the period between 1961–2030.

Figure 7. Goat meat production of Other Countries and estimated production quantities between 1961–2030 (1000 tons).

 The average production of goat meat by countries other than the top five was between 730 and 830 thousand tons in the 1960s, more than 950 thousand tons in the 1970s, and 1.2 million tons in the 1980s. Although relatively stable in the 1990s, production rose to approximately 1.40 million tons in 2000, 1.90 million tons between 2001 and 2010, and 2.23 million tons in the early 2020s. In 2024, production reached 2.38 million tons, 3.24 times the 1961 level, with a compound annual growth rate (CAGR) of 1.88%. The ARIMA (3,1,3) model estimates that production will reach 2.59 million tons by 2030, with an average annual production of approximately 2.52 million tons and a compound annual growth rate (CAGR) of 1.40% for the years 2024-2030.

Competition Status in Goat Meat Production: Table 9 contains statistical data evaluating the competitive level and market structure of world goat meat production over the years.

Table 9. Competitiveness of world goat meat production over the years

*2025-2030 years were forecasted with ARIMA and ARIMAX model

**The totals of leading countries and other countries were calculated as the world total in 2025-2030.

 While approximately 152 countries worldwide were engaged in goat farming in the 1960s, this number increased with the dissolution of the Soviet Union and Yugoslavia in the 1990s, and today the number of goat-producing countries has reached 170–175. Countries specialized in goat production, such as China, India, and Pakistan, consistently rank at the top in goat meat production, with China and India being by far the two most dominant actors, accounting for more than half of the total market. Pakistan and Nigeria are also observed to have moderate production shares in the market. Iran, Türkiye, Ethiopia, and Bangladesh have had a say in some periods and are present in the market with lower shares. The structure shows that the goat meat market is oligopolistic with few strong players dominating it. The calculated CR₅ values are increased from 41.63% to 68.84% and estimated to be 69.72% for 2025–2030. This indicates that the market is concentrated, as the HHI index has increased from 0.06 to 0.18. This scenario provides a structure with little or no competition but no monopoly. This means that market power in goat meat production is concentrated in a few countries.

DISCUSSION

 Reportlinker (2025) states that total sheep and goat meat production will be 15.7 million tons, while our study suggests that goat meat alone will be 8.10 million tons, and Demir (2025) indicates that sheep meat will be 11.92 million tons. It is estimated that total sheep and goat meat production will exceed 20 million tons in 2026. According to FAOSTAT (2025), global goat meat production increased by an average of 2.40% annually between 2013-2018, and rose to 2.85% between 2019-2024, likely due to the slaughter of more male goats. When the positive and negative aspects of this increase worldwide are briefly considered, it is noteworthy that goat meat, as a healthy protein source, holds an important place in the imports of developed countries, with the USA, France, Germany, and the UK being among the largest importing countries, as stated in FAOSTAT (2025). This contributes to export revenue for exporting countries and to health awareness for developed importing countries. Small ruminant farming has benefits of low start-up cost, fast liquidity (Onyango et al., 2015) and cost effectiveness in terms of feed and maintenance costs compared to cattle farming (Bettencourt et al., 2015). But this potential from an economic perspective brings serious environmental responsibilities as well. The impact of overgrazing on methane emissions (Samad et al., 2025) and its increasing share in the total emission profile as observed in the case of Pakistan (Mehmood et al., 2022) are noteworthy. According to Fielding (2022), uncontrolled grazing in arid pastures disturbs plant development. On the other hand, Fonseca et al. (2023) showed that controlled grazing in forested areas of the Mediterranean can reduce the risk of fire and improve the quality of the soil. All these findings demonstrate the need to achieve a sensible balance between nutritional, economic, and environmental aspects of small ruminant farming worldwide.

 The latest forecast indicates that China will continue to be among the world’s top producers of goat meat from 2025 to 2030. Output will probably level off at around 2.55-2.65 million tons a year but that is a large proportion of the world’s supply. Although the output is down a bit from its historic highs, China's dominance is still a major factor in the global market. With the share of China in the world production of sheep and goat meat (48% in 2020-2024 (FAOSTAT, 2025)) and given the current trends in goat-specific growth rates, the country is expected to retain its large market share and remain the leading supplier in the future. China’s expansion in goat meat production since 1960 caused an important shift in the sector, with the country accounting for 35% of global production by 2000. FAOSTAT (2025) data shows that despite a decrease in China's goat population from 137 million in 2019 to 117 million in 2024, carcass yield has increased from 13-14 kg to 18 kg, allowing the country to continue supplying approximately 30% of the world's goat meat. As highlighted by Wang et al. (2024), this yield increase is driven by state-supported genetic improvement efforts and modernization policies. But Visser (2025) stated that genetic selection and production-enhancing interventions aimed at high yields in small ruminants in China may increase meat, milk, and fiber yields in the short term but may lead to undesirable consequences such as mastitis, reproductive losses, and genetic diseases in the long term. Wang et al. (2024) stated that the country needs more goat and sheep meat but has not made much progress in meat production efficiency and therefore remains significantly dependent on imports. If China does not implement significant improvements in areas such as genetic innovation, feed management, and regional production support, its share of global goat meat production is likely to continue decreasing in the coming years. While the recent population decline in China is expected to negatively impact demand, in this country with a continuously growing GDP (World Bank 2025), consumer demand per capita has almost doubled due to increased incomes and increased awareness through education. Furthermore, demand for healthier foods is increasing among the elderly population (Wang et al. 2025). In rural areas, the declining population, coupled with modernization, has increased migration from rural to urban areas, leading to increased plant diversity in agricultural areas (Luo et al. 2025). Although the decrease in the number of producers in rural areas has caused a decline in the goat population, both genetic and technological advancements, as well as increased plant diversity, have resulted in increased yields in goat meat production. Thus, producers who migrated from rural areas to cities are now consumers. In this situation, both increased yields and demand pressure will increase production, and unless there is an adverse natural event, it is predicted that goat meat production in China will increase, albeit on a small scale, in the next 6 years.

 This study show that growth would continue and that India will have a 24.38% share between 2025 and 2030. Production grew from 1.2 million tons in the early 2020s to 1.72 million tons in 2024. The level of global production in 2024 is 7.33 times higher than it was in 1961, which means that it has grown at a rate of 3.21% per year since then. The ARIMA(1,1,6) model also says that production will reach 2.35 million tons by 2030, and the short-term CAGR will speed up to 5.33%.The results are similar to what Nair et al. (2021) predicted, which was that goat populations will reach 178.3 million by 2050, allowing meat production to continue. As noted by Gadekar et al. (2023), the states of West Bengal, Bihar, Maharashtra, Rajasthan, and Karnataka account for approximately 60% of India's goat meat production. Key reasons for the high demand for this meat include the use of true halal slaughtering methods, the presence of approximately 4,000 traditional slaughterhouses and 100 modern ones, geographical proximity to the Middle Eastern market, and the leaner, almost organic carcasses. These advantages make India one of the world's leading exporters of sheep and goat meat. On the other hand, the annual per capita meat consumption in the country is around 7 kg, significantly below the world average; this indicates that the domestic market still has growth potential. Demand is increasing and herds are growing; therefore, India’s goat sector is ready to ramp up production. However, this growth will need to be sustained through continued structural improvements.

 Pakistan was the third greatest producer of goat meat in the world from 2021 to 2024, accounting for approximately 7.25% of the total, and we think it will rise further to 7.28% from 2025 to 2030. Our forecast suggests that production will rise from 575,000 tons in 2025 to 635,000 tons in 2030. This means that the average will be about 605,678 tons, and the growth rate will be about 2.01% each year. Nouman and Khan (2015) studied the production of goat meat and estimated that it would be about 732,000 tons in 2020 and climb to 805,000 tons by 2028. This means that it would expand by 1.2% each year, which is more than what the ARIMA model said. Akram et al. (2022) say that red meat production would grow by 2.45% between 2020 and 2025.

 The ARIMAX (1,1,0) model says that Nigeria's goat meat production will expand by an average of 1.56% each year, from 295,573 tons in 2025 to 320,917 tons in 2030. On the other side, Akram (2022) determined that the compound growth rate was 6.69% from 1961 to 2021. We can tell that we will require 359,373 tons in 2025 and 496,614 tons in 2030 at this rate. Akram's prognosis, on the other hand, may not take into account changes in the economy's structure or outside forces because it assumes that the former growth path would continue without any problems. However, such factors are considered in the ARIMAX model, which allows more precise predictions by indicating accurate breaks. According to Reportlinker (2025), this figure will only increase at a rate of 0.2% per year, from 400,000 t in 2021 to 404,000 t in 2026. Different studies show different results with different methodologies and data sources. The methodology and data sources were controlled; therefore, our model adopted a more realistic, trend-based, and conservative view.

 The ARIMA model predicts that goat meat production in Bangladesh will increase from 247,000 tons in 2025 to 268,000 tons in 2030, with a compound annual growth rate of 1.58% for the period 2025-2030. These projections are in line with the historical trend indicating an average growth rate of approximately 0.4% per year during 2017-2026, with the market size reaching 240,000 tons by 2026. Other studies suggest more moderate growth paths. The levels of output they estimate are in line with our results. The agreement between the results of these different analytical techniques increases the credibility of our ARIMAX model that provides a reliable data-driven forecast for future industry trends.

 Five countries dominate global production, but the 'rest of the world' is not standing still. However, the share of 31.16% (2021-2024) of this remaining group of countries should fall to 30.28% by 2030, hiding a competitive battle between emerging producers. Chad and Ethiopia, for example, have proved that they are able to recover. Chad’s share of the world market has increased impressively, by 5.85%, to 2.05% since 2019. Ethiopia’s share remained unchanged at 1.79%. In contrast, Sudan has a tougher road ahead with output down 1.1% despite a 1.52% share of the market. The cases of Türkiye and Mongolia are also nuanced, both being important players (1.48% and 1.28%, respectively), but their growth trajectories are brought to a halt by abrupt production shocks of 2024, especially Mongolia’s steep plunge of 6.58%. The divergent trends in output growth in some parts of Africa and instability in the Middle East and Central Asia signal a world market that will be less stable and more competitive beyond the top five producers.

 The forecasts from the study are thought to be more accurate, especially in the short term. But in long-term forecasts, things like climate change, pandemics, policy changes, conflicts, or natural calamities like fires might make the predictions less accurate. If these kinds of things happen, the prediction outcomes could be very different from what was expected.

Conclusion: For a long time, China, India, and Pakistan were the three countries that made the most goat meat. Iran, Ethiopia, and Türkiye were also in the top five until 1990. Chad, Ethiopia, Sudan, Türkiye, and Mongolia all became substantial producers in the 2020s, coming in sixth to tenth. In the 1960s, China only had 6.28% of the global market. It is likely to stay around 31.27% between 2025 and 2030, which will maintain it in the lead. It is expected that China's production will go up from 2.54 million tons in 2024 to 2.65 million tons in 2030. The country's compound annual growth rate (CAGR), on the other hand, is expected to fall from 6.61% to 0.70%. Structural problems in manufacturing and an increasing dependence on imports are eating away at the country’s global competitiveness and could stunt growth in a highly competitive environment. India is known for irregular output, but being the second largest producer of goat meat in the world, it is expected to increase production from 1.72 million tons in 2024 to 2.35 million tons in 2030. This rapid growth is expected to drive India’s share of the global market to 24.38 percent. It is expected that Pakistan's production will continue between 575,000 and 635,000 tons from 2025 to 2030. This statistic will give it a 7.28% share of the global market. Nigeria is forecast to produce between 295,000 and 320,000 tons, and Bangladesh between 245,000 and 270,000 tons. On average, over the period 2025-2030, Nigeria will represent approximately 3.70% of world production and Bangladesh 3.09%. The fact that these five countries account for more than 70% of the world’s goat meat production indicates that the market is not highly competitive. This high concentration shows the urgent need to prepare for new ideas, climate change, and diseases. This is especially the case in Africa and other developing countries, where production is growing. Countries that can reduce production costs and tackle issues such as climate change, carbon emissions, and the use of new technologies will likely succeed in the global goat meat market in the future.

Author Contributions: Conceptualization, A.S.U.; data curation, D.T. and G.D.; methodology, A.S.U.; formal analysis, A.S.U.; investigation, A.S.U.; D.T. and G.D.; resources, A.S.U.; D.T. and G.D.; writing—original draft preparation and visualization, A.S.U.; supervision, A.S.U; validation, A.S.U.; writing—review and editing, A.S.U.

All authors have read and agreed to the published version of the manuscript.

REFERENCES

Adu, W. K., P. Appiahene and S. Afrifa (2023). VAR, ARIMAX and ARIMA models for nowcasting unemployment rate in Ghana using Google trends. J. Electron. Syst. Inf. Technol., 10: Art. 12. doi.org/10.1186/s43067-023-00078-1.

Akram, M. N., M. Amin, A. Yasin and M. Z. Aslam (2022). Future trends of red meat production in Pakistan: time series analysis. J. Anim. Plant Sci., 32(2): 489-495. doi.org/10.36899/JAPS.2022.2.0446.

Ariyo, A. A., A. O. Adewumi and C. K. Ayo (2014). Stock price prediction using the ARIMA model. Proceedings of the 2014 UKSim-AMSS 16th Int. Conf. Pc. Mdl.  Sim. (Cambridge, UK). doi.org/10.1109/UKSim.2014.67.

Bawdekar, A. A., B. R. Prusty and K. Bingi (2022). Sensitivity analysis of stationarity tests’ outcome to time series facets and test parameters. Math. Probl. Eng., 2022: Art. 2402989. doi.org/10.1155/2022/2402989.

Bettencourt, E. M. V., M. Tilman, V. Narciso, M. L. S. Carvalho and P. D. S. Henriques (2015). The livestock roles in the wellbeing of rural communities of Timor-Leste. Rev. Econ. Sociol. Rural, 53(1): 63–80. doi.org/10.1590/1234-56781806-94790053s01005.

Box, G. E. P., G. M. Jenkins, G. C. Reinsel and G. M. Ljung (2015). Time series analysis: Forecasting and control. 5th Ed. John Wiley & Sons, New Jersey, USA.

Chung, M. G., K. Kapsar, K. A. Frank and J. Liu (2020). The spatial and temporal dynamics of global meat trade networks. Sci. Rep., 10(1): Art. 16657. doi.org/10.1038/s41598-020-73591-2.

Danso, F., L. Iddrisu, S. E. Lungu, G. Zhou and X. Ju (2024). Effects of heat stress on goat production and mitigating strategies: a review. Animals, 14(12): Art. 1793. doi.org/10.3390/ani14121793.

de Lima Júnior, D.M., F.F. de Carvalho, F.J. da Silva, A.H.D.N. Rangel, L.P. Novaes and G.D.S. Difante (2016). Intrinsic factors affecting sheep meat quality: a review. Rev. Colomb. Cienc. Pecu. 29(1): 3-15. https://doi.org/10.17533/udea.rccp.v29n1a01.

Demir, G. (2025). Sheep meat production trends between 2023–2027: ARIMA analysis for leading countries and Türkiye. M.Sc. thesis (unpublished). Grad. Sch. Nat. Appl. Sci. Atatürk Univ., Erzurum.

Ellis, B. A. (2025). Unit root tests in R. Technical Paper, The Jethro Project (TJP). Retrieved May 5, 2025, from www.jethroproject.com.

FAOSTAT. (2025). Meat production. Retrieved June 6, 2025, from www.fao.org/faostat/en/#data/FS.

Fielding, D. (2022). Unravelling the terminology and impacts of rotational grazing – what evidence is there for environmental benefits? Climate-Positive Farming Reviews, Technical Report, The James Hutton Institute (JHI), Scotland.

Filder, T. N., M. M. Muraya and R. M. Mutwiri (2019). Application of seasonal autoregressive moving average models to analysis and forecasting of time series monthly rainfall patterns in Embu County, Kenya. Asian J. Probab. Stat., 4(2): 1-15. doi.org/10.9734/ajpas/2019/v4i230114.

Fonseca, F., M. Castro, L. Alves, J. Castro and T. de Figueiredo (2023). Impacts of extensive sheep grazing on soil physical and chemical quality in open mountain forests, NE Portugal. Span. J. Soil Sci., 13: 11632. doi.org/10.3389/sjss.2023.11632.

Gadekar, Y. P., A. K. Shinde, R. Banerjee and S. B. Barbuddhe (2023). Present status and future prospects of meat production from small ruminants: An overview. Indian J. Small Rum., 29(2): 185-197. doi.org/10.5958/0973-9718.2023.00052.1.

Karakaya, E. and A. S. Uzundumlu (2025). Kiwi production forecasts for the leading countries in the period 1983–2027. Appl. Fruit Sci., 67(2): Art. 66.  doi.org/10.1007/s10341-025-01289-8

Khalili, M. Y., M. R. Sevinç, M. A. Palabıçak and G. Sevinç (2025). Future forecasting of grain maize production in Türkiye with ARIMA model. Intl. J. Soc. Hum. Sci., 12(115): 12-20. doi.org/10.5281/zenodo.14787149

Krugman, P. R. and R. Wells (2021). Economics. 6th Ed. Macmillan International Higher Education, London.

Kurtoğlu, S. and A. S. Uzundumlu (2023). Egg Production Forecasts for Afyon Province for the Period 2018–2025. 1st Ed. Gece Publishing, Ankara, Türkiye.

Law, C., I. Fraser and M. Piracha (2020). Nutrition transition and changing food preferences in India. J. Agric. Econ., 71(1): 118-143. doi.org/10.1111/1477-9552.12322.

Luo, Q., C. Zhao, G. Luo, C. Li, C. Ran, S. Zhang, L. Xiong, J. Liao, C. Du, Z. Li, Y. Xue, M. Long, M. Li, X. Shen, X. Zhang, S. Yang and X. Bai (2025). Rural depopulation has reshaped the plant diversity distribution pattern in China. Resour. Conserv. Recycl., 215: Art. 108054. doi.org/10.1016/j.resconrec.2024.108054.

Mangiwa, R. D. D., R. Siregar, S. D. Sari and N. Agustina (2025). Application of the ARIMA model in forecasting ethereum prices. Parameter J. Mat. Stat. Ter., 4(1): 81-94. doi.org/10.30598/parameterv4i1pp81-94.

Mehmood, H. Z., H. Afzal, A. Abbas, S. Hassan and A. Ali (2022). Forecasts about livestock production in Punjab-Pakistan: implications for food security and climate change. J. Anim. Plant Sci., 32(5): 1347-1355. doi.org/10.36899/JAPS.2022.5.0541.

Miller, G. and E. Spiegel (2025). Guidelines for research data integrity (GRDI). Sci. Data, 12(1): Art. 95. doi.org/10.1038/s41597-024-04312-x.

Mishra, P., A. Yonar, H. Yonar, B. Kumari, M. Abotaleb, S. S. Das and S. G. Patil (2021). State of the art in total pulse production in major states of India using ARIMA techniques. Curr. Res. Food Sci., 4: 800-806. doi.org/10.1016/j.crfs.2021.10.009.

Nair, A. S., M. Thirunavukkarasu, A. Pandian, G. Senthilkumar and C. Balan (2021). Small ruminant population in India-a forecast for the future. Indian J. Small Rumin., 27(1): 24-30. doi.org/10.5958/0973-9718.2021.00021.0.

Navarrete-Molina, C., C. A. Meza-Herrera, A. De Santiago-Miramontes, L. M. Valenzuela-Núñez, R. I. Marin-Tinoco, M. A. Soto-Cardenas and M. A. Sariñana-Navarrete (2024). Dairy goat production: socioeconomic, environmental, and cultural importance across time (1970–2022) and possible scenarios (2050). Resources, 13(12): Art. 177. doi.org/10.3390/resources13120177.

Nimbalkar, C. A., S. D. Shinde, A. J. Nalawade and S. B. Patil (2019). Forecasting of sheep and goat meat export from India. Int. J. Curr. Microbiol. Appl. Sci., 8(7): 1365-1372. doi.org/10.20546/ijcmas.2019.807.16

Nouman, S. and M. A. Khan (2015). Modeling and forecasting of beef, mutton, poultry meat and total meat production of Pakistan for year 2020 by using time series ARIMA models. Int. J. Biotechnol. Allied Fields, 3(10): 154-171.

OECD/FAO (2021). Agricultural Outlook 2021-2030. OECD Publishing, Paris. Retrieved May 21, 2025, from doi.org/10.1787/19428846-en.

Onyango, D., G. Oyoko, R. Too and R. Masake (2015). The contribution of livestock to the South Sudan economy. Technical Report, IGAD Center for Pastoral Areas and Livestock Development (ICPALD), Nairobi, Kenya. Retrieved May 11, 2025, from www.igad.int/attachments/714.

Ordu, M. and Y. Zengin (2020). A comparative forecasting approach to forecast animal production: a case of Turkey. Livest. Stud., 60(1): 1-10. doi.org/10.46897/lahaed.719095.

Ospina, R., J. A. Gondim, V. Leiva and C. Castro (2023). An overview of forecast analysis with ARIMA models during the COVID-19 pandemic: Methodology and case study in Brazil. Mathematics, 11(14): Art. 3069. doi.org/10.3390/math11143069.

Palabıçak, M. A. and T. Binici (2023). Forecasting livestock production in developing countries by using time series: a case study of Türkiye. Appl. Ecol. Environ. Res., 21(2): 157-172. doi.org/10.15666/aeer/2106_57855806.

Pavic, I., F. Galetic and D. Piplica (2016). Similarities and differences between the CR and HHI as an indicator of market concentration and market power. Br. J. Econ. Manag. Trade, 13(1): 1-8. doi.org/10.9734/BJEMT/2016/23193.

Peng, R. Y., J. Wang, J. Zhai, Y. Zhang, H. Lu, H. Yi, Y. Yan, T. Peng, Z. Sharav and C. Chen (2022). Driving effect of multiplex factors on human brucellosis in high incidence region, implication for brucellosis based on one health concept. One Health, 15: Art. 100449. doi.org/10.1016/j.onehlt.2022.100449.

Peychinov, T., A. Karaivanova and T. Mecheva (2025). Proceedings of the 14th Int. Sci. Conf.  Eng., Techn. Sys.  Sci. (Plovdiv, Bulgaria). doi.org/10.3390/engproc2025100029.

Primawati, A., I. S. Sitanggang and D. A. Astuti (2023). Proceedings of the 3rd Int. Conf.  Smart Ct., Auto.  Intel. Comp. Sys. (Bali, Indonesia).

Report Linker (2025). Global mutton and goat meat market size volume per capita by country. Retrieved June 3, 2025, from www.reportlinker.com.

SAS (2025). SAS/ETS 13.2 user’s guide: The ARIMA procedure. SAS Institute Inc., Cary, North Carolina, USA.

Samad, H. A., E. K. Eshwaran, S. P. Muquit, L. Sharma, H. Arumugam, L. Kant, Z. Fatima, K. Sharun, M. A. Parameshwarappa, S. K. Latheef, V. S. Chouhan, V. P. Maurya, G. Singh and K. Kaniyamattam (2025). Sustainable livestock solutions: Addressing carbon footprint challenges from Indian and global perspectives. Sustainability, 17(5): Art. 2105. doi.org/10.3390/su17052105.

Serrano, R. and V. Pinilla (2014). Changes in the structure of world trade in the agri‐food industry: the impact of the home market effect and regional liberalization from a long‐term perspective, 1963–2010. Agribusiness, 30(2): 165-183. doi.org/10.1002/agr.21355.

Sindhura, A. and M. C. Sarma (2023). Significance of goat meat and meat products in human nutrition. 1st Ed. Choudhary Publishing Media, India; pp. 1-10.

Visser, C. (2025). The impact of genetic selection for increased production on fitness traits of small ruminants. Small Rumin. Res., 247: Art. 107491. doi.org/10.1016/j.smallrumres.2025.107491.

Wang, X., W. Shen, P. Wu, C. Wang, J. Li, D. Wang and W. Yue (2024). How food consumption trends change the direction of sheep breeding in China. Animals, 14(21): Art. 3047. doi.org/10.3390/ani14213047.

Wang, F., F. Shen and Y. Cai (2025). When the abundance ends: Economic transformation, population aging, and shrinking lifecycle surplus in China. J. Econ. Ageing, 30: Art. 100544. doi.org/10.1016/j.jeoa.2024.100544.

World Bank (2025). Advancing reforms, enhancing prospects: China Economic Update. Retrieved May 3, 2025, from data.worldbank.org/indicator/NY.GDP.MKTP.CD?locations=IN.