The procurement of natural gas has been a sticking point for European economies over the last two years. The Russian invasion of Ukraine in February 2022 triggered a dramatic jump in gas prices and a collapse in gas flows from Russia to Europe, placing the gas market at the centre of the policy debate. Yet, unlike oil, natural gas represents an unknown from a macroeconomic perspective. There is scant evidence on how demand and supply factors affect gas prices, or on the influence that gas prices in turn have on inflation and economic activity. Early analyses suggested that a 30% contraction in energy supplies could reduce output by up to 14% in Europe within two years (Garicano et al. 2022, Bachmann et al. 2022, Lan et al. 2022). In retrospect, many observers noted that things turned out to be less dramatic (Moll 2023). In a recent paper (Alessandri and Gazzani 2023), we provide direct evidence on this question by exploiting two tools that are widely used in empirical macroeconomic studies: ‘narrative’ identification and Bayesian vector auto-regression (VAR) models.
Using news to identify exogenous shocks to gas supply
Like all prices, gas prices move under the combined influence of both demand- and supply-side factors. The relevance of supply shocks is all too obvious after the Ukraine war; but large price shifts can also arise, for instance, when changes in business cycle conditions or temperatures shift households’ and firms’ demand patterns. To identify supply shocks, we parse a large dataset of daily news about the main European gas trading hub, the Dutch Title Transfer Facility (TTF), over the 2010-2022 period.
We focus on dates in which the prices of gas futures recorded large swings, and resort to a careful line-by-line analysis of the underlying daily news to separate events that are clearly symptomatic of shifts in the supply of natural gas to those relating to changes in demand and other confounding factors. The news do not give us a series of ‘supply shocks’ per se, but they allow us to construct a series that correlates with supply shocks and, as such, can be used in a regression model to study the impact of those shocks on the economy. Table 1 shows a subset of the dates in which the market was predominantly driven by supply-side news.
Table 1 Selected gas supply events
In the earlier sample we find evidence of sizable supply restrictions after the Fukushima accident (2011) and the first Ukrainian crisis (2014). The shocks became larger and more volatile in 2022. The exceptional price spikes that followed the invasion of Ukraine (on 24 February 2022) and the shutdown of the Yamal pipeline (2 March 2022) were partly compensated by accommodative signals from Gazprom; but prices climbed higher in the summer due to Russia’s decision to accept payments only in rubles and to the gradual shutdown of the Nord Stream pipeline. In Figure 1 we compare gas prices (orange line) to a counterfactual series obtained by cumulating the price changes observed around supply-side events (blue line). Supply shocks capture very accurately the steep rise and subsequent decline in prices observed between 2020 and 2022.
Figure 1 Gas prices and gas supply shocks
Notes: Real price of European natural gas (orange line) and cumulative instrument for the gas supply shocks from our narrative identification (blue line)
Macroeconomic and financial impact of gas shocks in the euro area
A simple regression analysis suggests that gas supply shocks cause significant adjustments in European financial markets:
a contraction in supply is followed by a sudden drop in equity prices and a rise in option-implied stock market volatility (VSTOXX), consistent with a negative reappraisal of the profitability and risk profile of European firms. Studying the impact of the shocks on output and inflation is harder because of the relative scarcity of data (macroeconomic indicators are only observed at monthly or quarterly frequencies) and the volatility caused by the Covid pandemic. To get around these problems we resort to a more sophisticated econometric model, namely a Bayesian VAR.
Figure 2 reports the responses of gas prices, industrial production, and core consumer prices in the euro area to a gas supply shock. Gas prices rise immediately by construction in response to the shock. Inflation is unchanged in the month when the shock takes place but rises persistently thereafter, reaching a peak roughly two years after the shock. A similar pattern emerges for industrial production, whose response is flat at zero for about one year and then declines slightly. Quantitatively, a 5% increase in gas leads to a 0.2% peak increase in core consumer prices and a 0.6% fall in industrial production.
Figure 2 The impact of gas supply shocks
Notes: Impact of an increase in real gas prices caused by a restriction in gas supplies on core inflation (HICP core) and industrial production (IP) in the euro area.
Gas versus oil shocks
Do gas shocks resemble oil shocks? We investigate this issue using a larger empirical model that captures the co-movement between oil prices, gas prices, and a range of financial and macroeconomic indicators.
The model reveals two interesting facts. One is that energy prices respond far more rapidly to oil shocks. The other one is that core prices respond more to gas shocks, even though it takes longer for this response to fully materialise. In other words, the implications of gas shocks are subdued and possibly harder to trace in the short run, but quantitatively more important in the medium run.
To better characterise these differences we compute for each shock a summary measure of the overall ‘pass-through’ to core prices, defined as the ratio between the cumulative responses of core and energy prices over a given time horizon.
The results, reported in Figure 3, show that the pass-through is almost twice as large for gas shocks. The results are consistent with multifaceted role of natural gas: in Europe, gas is employed for industrial uses, heating, and electricity generation, and it plays the role of marginal source of energy used to generate electricity (thus determining European electricity prices).
Figure 3 The pass-through of oil and gas shocks
Notes: For each shock, the figure shows the ratio between the response of core prices energy prices over horizons ranging between one and 50 months.
Conclusions and policy implications
After the Russian invasion of Ukraine in February 2022, energy has become a crucial issue for many countries around the globe, and natural gas – a previously somewhat neglected commodity – has gained centre stage in policy debates, particularly in Europe. However, little is known about the implications of gas supplies for economic activity and inflation. We fill this gap by suggesting a new strategy to isolate supply shocks in the European gas market and by quantifying the macroeconomic effects of those shocks in a rigorous econometric setup. We find that negative shocks to gas supplies are stagflationary, leading to a drop in economic activity and a significant rise in both energy and core consumer prices. We also find that gas and oil shocks work though the economy in different ways. Oil shocks generate sharp responses in economic activity and energy prices in the short run, but these abate quickly leaving core inflation relatively unaffected. Gas shocks have little impact in the short run, but significantly increase energy and core prices over longer horizons. After a year, the pass-through to core inflation is almost twice as large for gas shocks as for oil shocks. The analysis confirms that gas supply shocks were a key driver behind the surge in inflation in Europe in 2022 and suggest that their influence is likely to be felt well into 2023. It also highlights differences between oil and gas that could be relevant for future research: if oil and gas play a structurally different role in the economy, treating ‘energy’ as a single homogeneous good in empirical or theoretical macroeconomic models may be very misleading.
Authors’ note: This column does not necessarily reflect the view of the Bank of Italy or the European System of Central Banks (ESCB).
References
Alessandri, P and A Gazzani (2023), “Natural gas and the macroeconomy: not all energy shocks are alike”, Mimeo.
Auclert, A, H Monnery, M Rognlie and L Straub (2023), “Managing an Energy Shock: Fiscal an Monetary Policy”, Working Paper.
Bachmann, R, D Baqaee, C Bayer, M Kuhn, A Lo¨ schel, B Moll, A Peichl, K Pittel and M Schularick (2022), “What if? The economic effects for Germany of a stop of energy imports from Russia”, Wirtschaftsdienst 102: 251–255.
Caldara, D and M Iacoviello (2022), “Measuring Geopolitical Risk”, American Economic Review 112(4): 1194–1225.
Cook, T R, A Lusompa, and J Nie (2022), “Disruptions to Russian Energy Supply Likely to Weigh on European Output”, Economic Bulletin 1–4, 4 November.
Garicano, L, D Rohner and B Weder di Mauro (2022), Global Economic Consequences of the War in Ukraine: Sanctions, Supply Chains and Sustainability, CEPR Press.
Ka¨nzig, D R (2021), “The macroeconomic effects of oil supply news: Evidence from OPEC announcements”, American Economic Review 111(4): 1092–1125.
Lan, T, G Sher and J Zhou (2022), “The economic impacts on Germany of a potential Russian gas shutoff”, IMF Working Papers 2022/144.
Lenza, M and G E Primiceri (2022), “How to estimate a vector autoregression after March 2020”, Journal of Applied Econometrics 37(4): 688–699.
Mertens, K and M O Ravn (2013), “The Dynamic Effects of Personal and Corporate Income Tax Changes in the United States”, American Economic Review 103(4): 1212–47.
Pieroni, V (2023), “Energy shortages and aggregate demand: Output loss and unequal burden from HANK”, European Economic Review 154, 104428.
Stock, J H and M W Watson (2012), “Disentangling the Channels of the 2007–09 Recession”, Brookings Papers on Economic Activity: 81–135.
