Today, we’re fortunate to have Willem Thorbecke, Senior Fellow at Japan’s Research Institute of Economy, Trade and Industry (RIETI) as a guest contributor. The views expressed represent those of the author himself, and do not necessarily represent those of RIETI, or any other institutions the author is affiliated with.
Last year Donald Trump tweeted “Oil prices are too high, OPEC is at it again. Not good!” President Trump’s tweet reflects the conventional wisdom that higher oil prices reduce growth and lower stock prices in oil-importing nations like the U.S. The IMF (2014) similarly forecasted that a 20 percent increase in oil prices would raise inflation in advanced oil-importing countries by between 0.5 and 0.8 percentage points, lower GDP by between 0.4 and 1.9 percent, and decrease aggregate equity prices by between 3 and 8 percent.
However, American shale oil output has exploded since 2010. As Figure 1 shows, this has caused total U.S. production to soar. Oil prices may affect the U.S. stock market differently after the shale revolution (SR) than they did before.
Figure 1. U.S. Crude Oil Production. Source: U.S. Energy Information Agency.
Demand- and supply-driven changes in oil prices
When investigating how oil prices affect the stock market and the economy, researchers confront an identification problem. Oil prices not only impact the economy, but weakness in the global economy reduces oil demand and thus oil prices. To control for this, Hamilton (2014) employed several variables whose short run changes are correlated with changes in global growth and thus global oil demand but uncorrelated with changes in global oil supply. One such variable is the price of copper. It tends to fall when the global economy slows. However, over a week or less changes in oil supply do not affect copper prices. Two other variables that Hamilton used that are also correlated with global growth but not with oil supply over the short run are the ten-year Treasury interest rate and the trade-weighted dollar exchange rate.
Bernanke (2016) used Hamilton’s method to investigate the relationship between demand- and supply-driven changes in the daily price of West Texas Intermediate crude oil (WTI) and daily changes in the Standard & Poor’s 500 (S&P 500) stock price index over the June 2011 to December 2015 period. He found a correlation of 0.68 between oil price changes driven by demand factors and stock prices. He also reported a correlation of 0.05 between oil price changes driven by supply factors and stock prices. He questioned why supply-driven oil price changes are positively correlated with stock prices, given the conventional view that oil price increases harm the U.S. economy.
Why oil price increases raise stock prices after the shale revolution
To answer this question I investigated how WTI oil prices affected U.S. aggregate and industry stock returns before and after the SR (Thorbecke, 2019). For the pre-SR period I employed data from 3 January 1990 to 1 June 2007. For the post-SR period I employed data from 1 June 2010 to 28 September 2018. I excluded the Global Financial Crisis period between June 2007 and May 2010 because it contained wild swings in both oil prices and stock returns that could cloud inference.
The results indicate that after 2010 oil price increases driven by both demand and supply factors increase aggregate stock prices in the U.S. Before the SR, price increases driven by demand reduce aggregate stock prices while increases driven by supply have no effect. To shed light on why these responses differ before and after the SR I examined how oil shocks affect industry stock returns. Supply-driven price increases reduce industrial machinery and industrial engineering stock returns before the SR and increase them after. As domestic oil production has increased, spending by oil producers and other firms on industrial machinery and other capital goods has increased. Melek (2018) reported that higher capital expenditures by oil producers triggered increased capital expenditures by non-oil producers after the shale boom but not before. The coefficient on the chemical industry also changed from negative and significant to positive and significant, reflecting the growing importance of the petrochemical industry within the chemical sector and of chemicals as inputs to shale oil production. The coefficients on commercial vehicles (e.g., buses and energy saving devices) and marine transport changed from negative and significant before the SR to positive and significant after. The coefficients on oil industry stocks, while positive and significant in both periods, increased after 2010.
For demand-driven oil price shocks, consumer-oriented stocks such as personal goods, household goods, and food and beverage were much less harmed by demand-driven increases in price after the SR than before. This may indicate that consumers’ marginal propensity to spend windfall gains from lower oil prices has dropped in recent years.
Investigating oil price shocks using Kilian and Park’s and Ready’s approaches
Kilian and Park (2009) employed data on world crude oil production, dry cargo bulk freight rates, and oil price changes and used Cholesky decompositions to disentangle supply and demand effects on oil prices. I also used this approach and monthly data over the pre-SR (January 1990 – June 2007) and post-SR (June 2010 – June 2018) periods. The results indicate that positive oil price shocks decrease aggregate stock returns before the SR but do not affect them afterwards. Many of the sectors whose coefficients on supply-driven oil price increases change from negative to positive using Hamilton (2014) and Bernanke’s (2016) approaches also have coefficients on oil price shocks that change from negative to positive using Killian and Park’s approach. These include industrial machinery, industrial engineering, chemicals, commercial vehicles and marine transportation. In addition, many consumer-oriented stocks such as leisure goods, hotels, food & beverage, and travel & tourism are much less exposed to positive oil price shocks after the SR than before.
Ready (2018) reasoned that oil producers would beneﬁt from price increases due to oil demand, but have a natural hedge against supply difficulties. If oil becomes more diﬃcult to produce then producers would sell less but sell at higher prices. These two effects tend to offset each other. He used this logic to identify oil demand shocks. He then took oil supply shocks to be the portion of oil price changes that could not be explained by oil demand shocks and by unexpected changes in the Chicago Board Options Exchange volatility index (VIX).
Using Ready’s approach and daily data, the results indicate that before the SR the coefficients on oil shocks driven by both supply and demand factors are negative and insignificant for the aggregate stock market and after the SR these coefficients are both positive and are statistically significant for demand shocks.
Examining industry stock responses, before the SR 26 sectors are harmed by supply-driven oil price increases and eight sectors benefit. After the SR six are harmed and six benefit. Among the industries that are harmed before the SR but not after are industrial machinery, industrial engineering, commercial vehicles, and many consumer-oriented stocks such as retailers, food & beverage, and restaurants & bars
This work investigates how oil price increases driven by supply and demand factors affect U.S. stock returns. Results from several identification strategies indicate that supply-driven increases in oil prices reduce stock returns over the 1990-2007 period, but have attenuated effects or even raise returns over the 2010-2018 period. Industries that provide inputs or services to the energy sector such as industrial machinery and marine transport, industries in the oil supply chain such as petrochemicals, and industries producing energy-saving devices such as buses gain from higher oil prices after the shale revolution. In addition, in many specifications consumer-oriented stocks are harmed less by oil price increases after the SR than they were before.
These findings imply that the conventional view that oil price increases harm the overall U.S. stock market no longer holds. President Trump should be careful about seeking lower oil prices. Otherwise, he may reduce stock prices and damage large swathes of the U.S. economy.
Bernanke, B., 2016. The relationship between stocks and oil prices. Web blog post. Ben Bernanke’s Blog, February 19.
Hamilton, J., 2014. Oil prices as an indicator of global economic conditions. Web blog post. Econbrowser.com, 14 December.
IMF. 2014. World Economic Outlook. Legacies, Clouds, Uncertainties. International Monetary Fund, Washington.
Kilian, L., Park, C. 2009. The impact of oil price shocks on the U.S. stock market. International Economic Review, 50(4), 1267-1287.
Melek, N.C., 2018. The response of U.S. investment to oil price shocks: Does the shale boom matter? Economic Review, Federal Reserve Bank of Kansas City forthcoming.
Ready, R.C. 2018. Oil prices and the stock market. Review of Finance, 22(1), 155–176.
Thorbecke, W. 2019. Oil Prices and the U.S. Economy: Evidence from the Stock Market. RIETI Discussion Paper 19-E-003, available at: https://www.rieti.go.jp/jp/publications/dp/19e003.pdf . forthcoming in the Journal of Macroeconomics.
This post written by Willem Thorbecke.