The simultaneous effect of credit on sustainable food crop production and agricultural economic growth in Indonesia

In general, Indonesian farmers still face limited capital in their production. Therefore, credit is needed as additional capital, which is expected to help increase their agricultural business, especially rice production. This study analyzes how credit influences rice production and agricultural economic growth simultaneously. Secondary data are a time series over 12 years (2010-2021) from BPS Statistics Indonesia and Bank Indonesia. The Statistical Analysis System/Econometric Time Series (SAS/ETS) program version 9.4 is used to estimate this study using two-stage least squares (2SLS). The results show that credit simultaneously influences rice production sustainability and economic growth in the agricultural sector through credit interest rates. A simulation model that combines an increase in agricultural credit distribution of 20% and an interest rate of 5% has a greater impact on increasing rice production and economic growth in the agricultural sector compared to a simulation model that only provides a credit interest rate of 5% or 3%.


Introduction
In Indonesia, the agricultural sector is crucial for the nation's economic growth.In terms of GDP contribution, labor absorption, and meeting societal food needs, the agricultural sector is essential for sustainable development.The availability of adequate food for the public is necessary to strengthen the food security in Indonesia.However, Indonesian farmers still lack business capital for their production.Therefore, farmers often seek loans or credit to compensate for their lack of business capital.Credit is necessary for farmers to pay labor costs, input costs (e.g., purchasing seeds, fertilizers, and pesticides), and marketing expenses (e.g., transportation and storage).Credit is expected to help farmers purchase the best inputs at the right time to increase agricultural productivity [1,2], thereby increasing their income [3].Increasing agricultural productivity is expected to enhance the economic growth of the agricultural sector.Farmers need credit to purchase production inputs, such as seeds, fertilizers, and pesticides, as well as for marketing activities, including storage, transportation [4,5] and paying labor wages.To assist farmers in accessing credit, the government launched a credit program for productive businesses, including the agricultural sector, with interest rates below market rates, such as The KUR (People's Business Credit) program.
Figure 1 shows that the distribution of credit in the agricultural sector increased from IDR 90.99 trillion in 2010 to IDR 415.51 trillion in 2021 (an average annual growth rate of 15.48%).By the end of 2021, the percentage of credit in the agricultural sector was approximately 16.88% of all credit.The increasing of lending in the agricultural sector is in line with the increase in GDP from IDR 956.1 trillion in 2010 to IDR 1403.7 trillion in 2021 with an average growth of 3.56% per year.Meanwhile, rice production as a representative food crop from 2010 to 2016 increased by 2.52% per year, then decreased from 76.66 million tons in 2017 to 55.26 million tons in 2021.Studies have indicated that credit positively impacts agricultural production.Research using multiple regression analysis with the Cobb-Douglas production function explained that credit had a positive impact on wheat crop productivity in Punjabi, Pakistan, even though the impact of credit was relatively small [6].Another study shows that farmers who obtain loans could increase their income more than farmers who do not obtain credit [7].A study [8] shows that credit effectively increases wheat and sugarcane production and improves the livelihood of farmers in Faisalabad, Pakistan.In contrast to previous studies, this study uses secondary data to examine how credit influences rice production, which ultimately impacts economic growth in the agricultural sector.In addition, the study examines how credit affects rice production and economic growth of the agricultural sector in various scenarios.

Type and source of data
This study used secondary data with a span of 12 years (2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018)(2019)(2020)(2021).The data consist of the GDP of the agricultural sector, credit of the agricultural sector, inflation, rice production, harvested land area, agricultural labor, total household food consumption, and the interest rate of credit.These data were obtained from Statistics Indonesia BPS and Bank Indonesia.

Specification model
Specification models of credit and agricultural economic growth are developed based on the theory and previous studies.
=  0 +  1   +  2   +  3 (1) (3) Equation model (1) explains that the value of credit disbursed is determined by interest rates (SKB t), inflation (INFL t), and rice production (PROD t).Meanwhile, equation (2) identifies that rice production is not only influenced by credit (KRED t) but also by other factors such as harvested land area (LAP t) and agricultural labor (TKP t).Furthermore, equation (3) implies that apart from the level of rice production (PROD t) and credit (KRED t), agricultural growth is also affected by other factors such as inflation (INFL t), consumption (KON t), and the previous year's agricultural sector GDP (LPDB t-1).

Identification and estimation
The formulated model of credit, rice production, and agricultural economic growth consists of three endogenous variables, five exogenous variables, and one lag variable; therefore, the total number of variables in the model was nine.The model is over-identified, so it can be estimated using the Two-Stage Least Squares (2SLS) method.Data processing was performed using the SAS/ETS Version 9.4 for Windows.

Validation
The validation values used in this study are the Root Mean Square Percentage Error (RMSPE) and U (or Theil's Inequality Coefficient).RMSPE was used to calculate the percentage difference between the expected and actual values of endogenous variables throughout the observation period.The lower the RMSPE number, the better the prediction of the study's endogenous variables.The statistical value of U ranges from 0 to 1.If U = 0, then the endogenous variable prediction is perfect or close to the actual value.If U = 1, then the prediction of the endogenous variable is not realistic.The better the prediction of endogenous variables, the lower the RMSPE and U values [9].

Simulation
The simulated scenarios used to examine the impact of loans on rice production and agricultural economic growth are as follows.1.The interest rate for credit in the agricultural sector at 5%. 2. The interest rate for credit in the agricultural sector at 3%. 3. The realization of agricultural sector loans increased by 20% with an interest rate of 5%

Determinants of credit, rice production, and economic growth in the agricultural sector
The credit equation model used in this study consists of interest rates, inflation, and production levels.The model estimation result was statistically sound based on the coefficient of determination R2 = 0.7616 and an F-test of 0.0071.The value of R2 = 0.7616 shows that 76.16% of the explanatory variables for lending rates, inflation, and production explain credit.The remaining 23.84% were explained by factors not included in the model.
Furthermore, the variable that significantly affects credit value is the interest rate.This indicates that the interest rate determines the value of the credit disbursed in the agricultural sector.The negative sign of the interest rate indicates that the lower the interest rate, the more likely it is to increase the value of disbursed credit.Conversely, a higher lending rate reduces the value of the credit extended to the agricultural sector.The inflation variable has no significant effect on credit value.The positive parameter of inflation indicates that the higher the inflation, the higher is the disbursed credit value.The case with the production variable is insignificant for credit value.The positive parameter of the production variable implies that the higher the production, the higher is the value of credit.This means that the higher production level produced by farmers can affect the level of profit and income, which has implications for farmers' purchasing power and capacity to obtain credit.The Cobb-Douglass production function approach is used in this study, where the rice production equation is influenced by labor, land area, and credit.The production equation model is relatively good, as shown by the coefficient of determination R2 = 0.9937 and F-test (0.0001).The value of R2 = 0.9937 indicates that 99.37% of the explanatory factors, including labor, land area, and credit, can explain production variables, while the remaining 0.63% can be explained by additional variables not included in the model (See Table 2).The significant variables influencing rice production are land area and credit.The positive value of the land area variable shows that the greater the land area, the higher is the rice production.The positive credit variable suggests that the greater the credit value disbursed to farmers, the greater is their farming output.Credit can be used to improve the quality of farming inputs, not only to boost farm production [10], but also to improve the technical efficiency of farmers [11,12].However, the labor variable had no significant effect on rice production.The negative parameter of the labor variable indicates that the higher the amount of labor used, the lower is the level of production.This is thought to be related to the high level of wages in using labor, which will increase costs compared to the output produced.
At the macro level, the equation model of economic growth in the agricultural sector was influenced by the levels of production, consumption, credit, inflation, and economic growth in the previous year.Table 3 shows that the coefficient of determination R2 = 0.9860 and the F test < 0.0001.The equation model of economic growth in the agricultural sector (GDP) was relatively good.The value of R2 = 0.9860 indicates that 98.60% of the agricultural sector GDP variable can be explained by the explanatory variables consisting of labor, land area, and credit, whereas the remaining 1.4% is explained by other variables not included in the model.
Rice production, inflation, and credit are significant variables influencing economic growth in agriculture.The positive sign of the credit variable means that a greater value of credit for the agricultural sector can lead to an increase in agricultural economic growth.Similarly, a positive production variable indicates that increased production by farmers can boost agricultural economic growth.A negative inflation variable indicates that increasing inflation reduces economic growth.This indicates that higher inflation rates can decrease people's purchasing power as the prices for products and services rise [13].This can reduce people's capacity and ability in production and consumption activities, which has implications for decreasing economic growth.Furthermore, there is no obvious effect of consumption or lag in the economic growth variables on the growth of the agricultural sector.Higher consumption tends to impede agricultural economic growth, as indicated by the estimated negative parameter of the consumption variable.Therefore, if consumption does not follow a rise in rice production, the growth of the agricultural sector will slow.The negative sign of previous agricultural economic growth indicates that the higher economic growth rate the year before will probably result in a slower growth rate at this time.

The credit impact on rice production and agricultural economic growth
Based on Table 4, the RMSPE value of the credit equation model had an error rate of 2.10 %.The model of rice production and the agricultural economic growth respectively have an error rate of 0.06 % and 2.71 %.This indicated that the model is valid and can be simulated.Furthermore, based on the U-Theil coefficient value criteria, the three variables-credit, rice production, and economic growth in the agricultural sector-have values close to zero (0).This indicates that the model is good, so simulations can be carried out on the model.Furthermore, a simulation of the credit interest rate at 3 % can increase the credit distribution by 10.90 % (See Table 6).This encourages an increase in rice production of 0.31 %, which, in turn, can improve economic growth in the agricultural sector by 12.85 %.However, the value of changes in the impact of interest rate credit at 3% on rice production and economic growth in the agricultural sector is relatively smaller than that of a credit interest rate of 5 %.This is because the interest rate is another factor that plays an important role in the profitability of financial institutions [14-16].The next simulation is the realization of credit at 20 % and a credit interest rate of 5 %.Based on Table 7, the disbursement of credit increases by 19.58 %, which in turn can encourage farmers to increase rice production by 1.26 %.This affects the increase in economic growth in the agricultural sector by 51.25 %.It means that lower credit interest rates can increase farmers' access to credit.The amount of credit received can be utilized by farmers to increase working capital for purchasing production inputs and investment capital for buying equipment or renting land.It is because farmers can only buy production inputs such as seeds, pesticides, and fertilizer due to limited capital so they cannot expand their business.It is hoped that low credit interest rates will help farmers pay loan installments more easily and turn their business cash around to increase sustainable production.The use of credit is expected to encourage the increased production and productivity of farmers' businesses [17,18], which has implications for economic growth in the agricultural sector.

Conclusions and suggestions
Based on these results, credit can simultaneously affect rice production and agricultural economic growth when credit is influenced by interest rates.The simulation model of reducing credit interest rates to 5% or 3% increases agricultural credit distribution, rice production, and agricultural economic growth.
Meanwhile, the simulation model of increasing credit distribution in the agricultural sector by 20% and

Table 1 .
Determinants of credit

Table 2 .
Determinants of rice production

Table 3 .
Determinants of economic growth in the agricultural sector

Table 4 .
Model validation results of credit, rice production, and economic growth of Indonesia's agricultural sector in 2010-2021Table5shows that a credit interest rate of 5 % can increase credit disbursements by 18.18 %.This explains why greater opportunities for farmers to access credit can encourage a 0.52% increase in rice production and increase the agriculture economic growth by 21.42 %.

Table 5 .
Scenario impact of interest rate reduction at 5 %

Table 6 .
Scenario impact of interest rate reduction at 3 %

Table 7 .
Scenario impact of increasing credit by 20 % and decreasing interest rate at 5 %