Abstract
The study examines the effect of crop production technology intensification on wheat yield under drought conditions. The causal effect of fertilizers and herbicides usage was quantified based on a dataset of reports from 196 farms located in the Kulunda Steppe in Altai Krai. To quantify causal effects, the various methods were tested – Two Means Comparison tests and Linear regression. Due to the not experimental nature of the data, Propensity Score Matching was used to balance the sample based on the main indicative signs (covariates). We refined the effects of crop technologies intensification based on the balanced sample. The results showed that, according to all considered methods, intensification of agriculture even in the dry climatic conditions resulted in the expected growth of wheat yield. The average causal effect of intensive farming was +2.02 dt/ha. The results also showed that it is possible to obtain more correct estimates of causal effects based on balanced samples. Using simple approaches like Two Means Comparison methods lead to underestimation or overestimation. Finally, we highlighted some limitations and peculiarities of the Propensity Score Matching method. However, the Propensity Score Matching can be considered a good and prospective tool for developing digital services in agricultural analytics.