The effects of electricity price changes on prices of other goods and services – evidence from Turkey

ABSTRACT

This article employs a Factor-Augmented Vector Autoregressive model to assess the effects of electricity price innovations on prices of other goods and services. Using monthly series from Turkish Domestic Producer Price Index (D-PPI) and Harmonized Index of Consumer Prices (HICP) components, the results from the analyses on D-PPI components suggest that (i) Machinery & Equipment (not elsewhere classified); Electrical Equipment; and Rubber & Plastic Products increase the most, while (ii) Tobacco Products; Crude Petroleum & Natural Gas; and Water Supply, Sewerage, Waste Management & Remediation Services increase the least. In addition, the results from the analyses on HICP components suggest that (iii) Housing, Water, Electricity, Gas & Other Fuels; Furnishings, Household Equipment & Maintenance; and Restaurants & Hotels increase the most, while (iv) Communications; Alcoholic Beverages, Tobacco & Narcotics; and Education increase the least.

KEYWORDS:

• Electricity prices
• inflation
• pass-through
• FAVAR

JEL CLASSIFICATION:

• Q43
• E31
• O13

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ To be precise, the standard deviations of the monthly inflation of the CPI electricity and HICP are both 3.38, it is 2.06 for HICP all items and 2.2 for D-PPI general in the period between 1996 and 2018.

² World Bank Data and Electricity Generation – Transmission Statistics of Turkey published by TEIAS.

³ BP Statistical Review of World Energy (2018, 6) reports that the Turkish electricity generation growth rate was 8% in 2017.

⁴ Electricity Market Development Report 2017 published by Republic of Turkey Energy Market Regulatory Authority reveals that the share of the natural gas in Turkish electricity production was 32.38% in 2017.

⁵ Gazprom (2018) indicates that in 2017, the Turkish share of the Russian natural gas exports accounts for 19% of Russia’s total natural gas exports to Western European countries.

⁶ We implemented a set of unit root tests in order to determine whether the series have unit root or not. Unit root tests indicate that the price growth series are all stationary. These tests are not reported here to save space.

⁷ There are no significant differences between Turkey’s methodology as reported by Eurostat and related international standards. TSI employs NACE, Rev.2 classification procedure by using cash prices excluding value-added tax (VAT) and all relevant taxes in the calculation of D-PPI data series, and the base year of the series is 2003. All of the data were compiled by survey results (see Turkish Statistical Institute CPI Metadata definition).

⁸ When the confidence interval contains the baseline, then we fail to reject the null hypothesis that there is no effect of electricity price innovations on that particular variable. In general, impulse response figures reveal that the shocks in the price of electricity increase all prices on different categories. Overall, the effect of shocks dies out after 5–7 months. Impulse response figures reported here are the supplementary material that is available from the authors upon request.

⁹ We also conduct the same analyses for the components of CPI. Since the results from these analyses are not statistically significant and/or economically not intuitive, we do not report them in this study. The source of this limitation possibly stems from the failure to perfectly match the components of the two CPI series whose base years are 1994 and 2003. Only seven components of the CPI whose base year is 1994 matched perfectly with the CPI series whose base year is 2003.

¹⁰ Electricity prices and prices of other products might be responding to other variables related to the state of the economy, monetary policy and exchange rates. Thus, we repeat the same exercises such that we incorporate these variables into the system. We also include three macroeconomic series which are Weighted Average Overnight Interest Rate, Industrial Production growth Rate and US Dollar Selling Rates as percentage change from the previous period that are obtained from The Central Bank of the Republic of Turkey and Bloomberg. Since there is only a slight change in terms of the order of magnitude and magnitude itself after incorporating these macroeconomic series into the system, these tables also indicate the robustness of our analyses.