Investigating the Duesenberry's ratch effect hypothesis in Iran using three approaches: ARDL, BMA_ADL and LSTM

Consumption as the most stable variable of GDP, plays an important role in the economy. According to consumption theories, the form of the consumption function in the short and long run changes the coefficients of macro-variables multiplier, This has an effect on how macro policies affect economic variables. In this regard, in the present study, an attempt has been made to empirically test the hypothesis of the ratch effect of Duesenberry consumption using annual time series data of the Iranian economy during the period 1976-2020. For this purpose, the Autoregressive Distributed Lag (ARDL), BMA_ADL and deep learning method LSTM (Long short term memory) has been used. results indicate that this hypothesis is not consistent for Iran. In other words, empirical evidence shows that, contrary to Dusenberry's theory, the slope of the consumption function in the short term is higher than the slope of the long-run function. As a result, the Multiplier coefficient of demand management policies in the short run is more than the long run. This pattern of consumption behavior can be called the precautionary behavior of households, which originates from Iran's economic conditions.