Credit Dynamics of Various Entities in Russia: Impact of Oil Prices and Sanctions

Credit is one of the most important indicators leading boom/bust periods in the economy, therefore studying its dynamics (cycles) allows for early warning of overheating. At the same time credit is important driver of economic growth, and cases of credit constraints for the country as a whole or some of its sectors of companies often become a serious obstacle for growth and development. Sanctions, especially financial, represent one possible way of constraining country's economic prospects. Recent sanctions against Russia might seem relatively mild, but Russian economy was growing at lower and lower rates with oil prices decreasing at the same time. Our aim is to decompose the effects of sanctions and oil prices on the credit dynamics of various entities in Russia. At the first stage we identify credit cycles using threshold method. After the seasonal component is eliminated, we proceed with disaggregating time series we use for our study into long-run trend and cyclical components. For this we apply the Hodrick-Prescott (HP) filter. We favour its use compared to the Baxter-King filter since the latter cuts off some data at the beginning and the end of the times series, and since we have only 64 observations for the longest time series, we opted for HP filter that uses more information. Once the trend is accounted for, thresholds (of statistical nature) can be applied to determine the start and the end dates of the credit boom, denoting cyclical variations higher than average. More precisely, if lit is the deviation of the logarithm of real per capita credit from its long-run trend and if σ(li) is the standard deviation of the cyclical component of real per capita credit, then if on one or more particular sequential dates it is true that lit ≥ φσ(li) (φ is the threshold), we can claim that on this date(s) credit boom was observed. To check for robustness, alternative values of φ were used. At the second stage VECM models were constructed to account for the interaction of endogenous variables. In all cases, however, the coefficients and the share of explained variation due to endogenous variables in the equation for credit indicator were statistically insignificant, which allowed us to use a simple single-equation model for out-of-sample forecasting. The data until the 4th quarter 2013 were used for estimation, and then the data on exogenous variables were used for out-of-sample forecasting (dynamic). The difference until 4th quarter 2014 was used to measure the quality of the forecast and to judge if the quality of the forecast since 4th quarter 2014 could be used as a proxy of the sanctions effect. In order to determine the effect of oil prices, the out-of-sample forecast was also made using the average quarterly oil prices for 2012-2013. Additionally it should be mentioned, that the impulse response analysis in VECM models in most cases demonstrated significant response of GDP gap on changes in the oil prices. In the case of internal credit indicators the internal interest rate also showed response to the GDP gap proxy changes and to the changes in the oil prices. Most of the private sector external credits saw the boom coinciding with the time of the world financial crisis of 2008-2009, while for the government the boom in external borrowing was identified in 2012-2013. The government-affiliated companies and banks had another external credit boom at the end of 2014 – early 2015. It should be stressed that it is visible that financial sanctions have changed the composition of external borrowing from direct investments, bonds and credits to more short-term and less direct external financing. External credit was partly substituted for by internal credit as the boom at the end of the study period suggests. Again, the government has somewhat different timings of credit booms in relation to internal credits. Mostly total internal credit has the same timings of booms as credit in national currency, which is not surprising taking into account the dominant share of credit in national currency in the total outstanding credit, especially after the crisis of 2008-2009 when the banking system became aware of the necessity to deal with the currency mismatch between assets and liabilities. Results of decomposing the effects on external borrowing and domestic credit into effect related to decreasing oil prices and effect due to external (financial) sanctions show that sanctions are felt more compared to decrease in oil prices (in case of external borrowing), and that short-term borrowing decreased less for government companies and banks. Domestic credit market was also influenced by sanctions and decreasing oil prices to varying degrees as external credit was largely substituted by domestic credit where possible.