Chapter 3. Structural Changes in Natural Gas Markets

Rabah Arezki, and Akito Matsumoto
Published Date:
December 2017
  • ShareShare
Show Summary Details

Natural gas markets are much less integrated than oil markets, a reflection of the cost and logistical difficulty of trading gas across borders. This results in substantial price differences across regions despite increasing liquefied natural gas trade. Global natural gas production and consumption have increased steadily over time and are projected to increase even more rapidly in the medium term. Three major historical developments have had particularly important implications for gas and energy markets: the shale gas revolution in the United States starting in the 2000s, the reduction in nuclear power supply after the Fukushima disaster in Japan in 2011, and the geopolitical tensions between Russia and Ukraine from the mid-2010s. These developments not only had profound effects on regional prices but also revealed specificities about the structure of natural gas markets. Natural gas could constitute a bridge from coal and oil to renewables during the so-called energy transition.

Natural gas is a cleaner fossil fuel than either petroleum products or coal and does not present the potential environmental liabilities associated with nuclear energy generation. Despite these advantages, the cost and logistical difficulty of trading gas across borders leave natural gas markets much less integrated than oil markets. Shipping or transporting natural gas requires either costly pipeline networks or infrastructure and equipment for liquefying (compressing) the gas, dedicated vessels for transport, and then facilities for regasification at the destination. The lack of integration of gas markets leads to substantial price differences across regions. These have been exacerbated by the U.S. shale gas boom in the 2000s and the Fukushima nuclear disaster in Japan in 2011, despite the growth of trade in liquefied natural gas (Figure 3.1).1

Figure 3.1.Natural Gas Prices

(US dollars a million British thermal units)

Source: IMF, Primary Commodity Price System.

Note: EU = European Union; LNG = liquified natural gas; US = United States.

Technological improvements in exploration and drilling activities have enabled both new discoveries and exploitation of previously identified reserves of natural gas, and there are many more prominent producers of natural gas today than there were in the 1990s.2 Iran, Russia, Qatar, Turkmenistan, and the United States had the largest reserves of natural gas in 2015. The largest producers of natural gas in 2015 were the United States and Russia, followed by Iran, Qatar, and Canada (Table 3.1).

Table 3.1.Production and Consumption of Fossil Fuels and Natural Gas by Country, 2007 and 2015
Fossil Fuels
Proven Reserves20072015
Oil (billion barrels)1,4191,696
Natural Gas (trillion cubic meters)162186
Coal (million tons)n.a.891,531
Oil (thousand barrels a day)82,27791,670
Natural Gas (billion cubic meters)2,9653,539
Coal (million tons)6,6887,861
Oil (thousand barrels a day)87,08795,008
Natural Gas (billion cubic meters)2,9693,469
Coal (million tons of oil equivalent)3,4763,840
Natural Gas
Proven Reserves (percent of world reserves)20072015
United States4.175.59
Production (percent of world production)20072015
United States18.4021.68
Consumption (percent of world consumption)20072015
United States22.0322.43
European Union16.3311.59
Source: British Petroleum, Statistical Review of World Energy, 2016.Note: n.a. = not available.
Source: British Petroleum, Statistical Review of World Energy, 2016.Note: n.a. = not available.

Natural gas consumption has risen steadily over time and in 2015 accounted for nearly 25 percent of global primary energy consumption, whereas the share of oil has declined rapidly, from 50 percent in 1970 to about 30 percent in 2015. Global natural gas demand is projected to increase strongly in the medium term (IEA 2014), with countries outside the Organisation for Economic Co-operation and Development (OECD) accounting for the bulk of the growth. Natural gas usage faces competition from substitutes for natural gas in all sectors, particularly from renewables and coal in power generation, in part because of subsidies and gas-pricing regimes. In addition, the implementation of widespread carbon taxation would tilt demand from coal toward natural gas and eventually from natural gas toward renewables. On the other hand, natural gas can complement the use of renewables, particularly to compensate for intermittency—at least while battery technology remains insufficient. Natural gas is also expected to make further inroads as a transportation fuel, including the use of liquefied natural gas for commercial trucks, passenger vehicle, and marine vessels.

The pattern of global trade in natural gas has evolved rapidly. Because natural gas has mainly been transported to consumers via pipeline, only one-third of the natural gas that is consumed is traded internationally. Europe and North America are by far the largest markets integrated by pipelines, but their net imports have declined since 2005 on account of weaker economic activity and higher gas production in the United States. One-third of internationally traded natural gas is shipped as liquefied natural gas, and that share has been expanding rapidly, mainly to Asia (Figure 3.2). There were nearly 20 countries producing liquefied natural gas in 2013. Qatar has rapidly developed liquefied natural gas export capacity in the past decade and is to date the largest exporter, accounting for about one-third of global natural gas trade. Australia has been investing massively to export liquefied natural gas to Asian markets and may exceed Qatar as the world’s largest exporter in coming years.

Figure 3.2.Liquefied Natural Gas Imports and Exports, 2013

(Millions of tons)

Source: Argus Media (

Note: UK = United Kingdom; US = United States.

Global Implications of the U.S. Shale Boom

The surge in the production of shale gas made the United States the largest natural gas producer in the world as of 2011,3 and the United States started exporting liquefied natural gas in the spring of 2016 and became a net exporter of natural gas in the fall of 2016. With surging supply, natural gas prices in the United States fell sharply since the global financial crisis in 2008 and have not recovered their precrisis levels. Moreover, the structural shift represented by the United States becoming the world largest producer of natural gas has left U.S. prices effectively decoupled from those in the rest of the world. In particular, prices in Asia and the European Union doubled after the financial crisis, partly because the price of imported natural gas was indexed to oil prices until oil prices collapsed, and while natural gas prices there fell in line with oil prices after 2014, U.S. gas prices remained much lower.

Energy users in the United States and Mexico have been the main beneficiaries of the energy price declines that resulted from the U.S. shale revolution. However, U.S. shale production has helped to stabilize international energy prices, including by freeing global natural gas supplies for European and Asian markets and thus offsetting some of the shortages caused by geopolitical disruptions.4 Also, in Europe imports of U.S. shale oil displaced imports of U.S. coal and lowered overall energy costs.

The shale gas boom in the United States has had a significant impact on the geography of global energy trade.5 U.S. fossil fuel imports decreased to $97 billion (0.5 percent of GDP) in 2016 from $425 billion (2.9 percent of GDP) in 2008. Both U.S. demand for coal and U.S. coal prices also declined. This in turn encouraged increased exports of coal to Europe, which, together with weak activity there following the recession, reduced Europe’s demand for natural gas.6 The shale gas boom also drastically reduced U.S. liquefied natural gas imports from Africa, the Middle East, and Trinidad and Tobago (Figure 3.3) and also substantially reduced natural gas imports from Canada, triggering a sharp decline in prices. Exporters have shifted energy exports to other locations such as China, Europe, and India in response to the U.S. reduction in energy imports, but Trinidad and Tobago has seen its exports of liquefied natural gas plummet since the start of the U.S. shale gas boom, and the country is actively seeking to reorient its liquefied natural gas exports toward Asian markets.7 In the United States, the shale gas boom has made redundant much of the liquefied natural gas import infrastructure. The infrastructure cannot easily be converted to export capacity because liquefaction capacity is different from import regasification capacity. In addition, U.S. firms are required to obtain authorization to export natural gas (except to Canada and Mexico), although there are signs that these regulatory obstacles are loosening.8 In the medium term, the removal of U.S. gas export restrictions will trigger the build-up and reconversion of liquefied natural gas facilities for export purposes and in turn could help reduce energy price differences worldwide and further affect other natural gas exporters.

Figure 3.3.United States: Liquified Natural Gas Imports

(Billion cubic feet)

Source: US Energy Information Administration.

The U.S. advantage in natural gas has also led to an increase in U.S. competitiveness in nonenergy products. Results of a bivariate vector autoregression, including the difference in industrial production and the difference in the price of natural gas between the United States and Europe, suggest that natural gas prices can have a substantial independent impact on economic activity (Figure 3.4). This specification controls for global shocks such as the global financial crisis, an issue that has been overlooked in other studies.9 A 10 percent reduction in the relative price of natural gas in the United States is found to lead to an improvement in U.S. industrial production relative to that of the euro area of roughly 0.7 percent after 1.5 years. Box 3.1 provides estimates of the gain in international competitiveness of U.S. manufacturing exports due to cheaper natural gas.

Figure 3.4.Impulse Response of Relative Industrial Production to a One Unit Relative Natural Gas Price Shock

(Months forward on x-axis)

Source: IMF staff calculations.

Note: The estimated vector autoregression model includes two variables: the relative industrial production in the United States and the euro area and the relative natural gas price in the United States and Germany, using monthly data for 2005–13. The impulse-response functions correspond to the response of relative industrial production to a one unit shock in relative natural gas prices. Red lines indicate 80 percent confidence intervals, and the shaded areas correspond to 95 percent confidence intervals.

Cheaper natural gas prices benefited energy-intensive sectors in general and the natural-gas-intensive petrochemical sector in particular. Indeed, the petrochemical industry has made very sizable investments in new plants in the United States, and this is likely to continue as shale gas supplies will likely continue to expand for the foreseeable future.

These phenomena also suggest that when considering the effect on the U.S. economy of the oil price decline that began in 2012, the positive effects should be somewhat discounted given that natural gas prices declined before oil prices, unlike in the past when oil and gas prices moved in tandem.

Aftermath of the Fukushima Disaster in Japan

The Fukushima Daiichi nuclear disaster in March 2011 highlighted the environmental liabilities associated with nuclear power generation and induced a sharp increase in natural gas usage. Before the disaster, about one-quarter of Japan’s energy was generated by nuclear reactors. Following the disaster, the Japanese government decided to halt production at all nuclear power plants. To compensate for the resulting loss in electricity generation, electric power companies enhanced their use of fossil-fuel power stations and appended natural gas turbines to existing plants. As a result, Japan’s liquefied natural gas imports increased dramatically—by about 40 percent (Figure 3.5).

Figure 3.5.Japan: Liquefied Natural Gas Imports

(Thousands of metric tons)

Source: Thomson Reuters Datastream.

Japan became the world’s largest importer of liquefied natural gas. In 2013, its imports of liquefied natural gas amounted to 119 billion cubic meters, more than a third of the world total. Increased natural gas demand from Japan has benefited producers in Asia, the Middle East, and Oceania at a time when global natural gas demand has slowed (Figure 3.6). Japan’s imports have helped offset some of the negative effects of the reduction in U.S. liquefied natural gas imports. Exports to Japan of liquefied natural gas from Australia, Brunei Darussalam, Indonesia, Malaysia, and Qatar rose rapidly. The sharp increase in natural gas demand led to higher prices in Asia, and in Japan in particular, with prices in Asia reaching twice European prices and four times U.S. prices at one point. However, after Japan began to reactivate its nuclear power plants and increase the use of ultra-high efficiency coal plants, the price difference between Asia and Europe narrowed substantially, although Europe continues to rely primarily on pipeline gas. Natural gas prices in Europe and Asia declined substantially due to the oil price collapse, as they are typically indexed to oil prices.

Figure 3.6.Japan: Liquefied Natural Gas Imports by Region

(Trillions of Japanese yen)

Source: Thomson Reuters Datastream.

Note: ASEAN = Association of Southeast Asian Nations.

Box 3.1.The Trade Implications of the U.S. Shale Boom

The shale boom has led to a debate in the United States about whether relaxing restrictions on exports of natural gas will diminish the gains in external competitiveness resulting from lower domestic natural gas prices. The shale gas boom led to a decoupling between U.S. natural gas prices and those in Europe and Asia since 2005, and these price differentials are expected to persist. At the same time, the share of energy-intensive manufacturing exports in total U.S. manufacturing exports has been rising steadily, whereas the share of non-energy-intensive exports has been declining (Figure 3.1.1).

Figure 3.1.1.Manufacturing Sector Exports

(Percent of total US manufacturing exports, unless indicated otherwise)

Source: IMF staff calculations.

This box examines the global trade implications of international differences in natural gas prices using the U.S. shale gas boom as a natural experiment. The main finding, based on sector-level data, is that the current price gap between the United States and the rest of the world has led on average to an increase in U.S. manufactured product exports by 6 percent since the start of the shale gas boom. Even though natural gas and energy costs in general represent relatively small shares of total manufacturing input costs, the lower natural gas price in the United States, which is expected to persist in the future, has had a noticeable effect on U.S. energy-intensive manufacturing exports.10

Energy Intensity and Manufacturing Exports

For the period 2000–12, which covers the shale gas boom in the United States, the logarithm of manufactured product exports is regressed on the interaction between differentials in energy intensity and in price between the United States and the rest of the world. The specification is a classical equation suggested by trade models. The coefficient associated with the interaction term is expected to be positive; that is, the more energy intensive a product is, the more likely it is to be exported. The equation estimated is

in which αi,j,k are origin, destination, and sector-specific joint fixed effects capturing sector-specific distance, and γt are time fixed effects capturing common shocks. The product export is equal to the value exported of a specific manufacturing sector at the five-digit level for which information is available, from Schott 2008, on the customs district of origin i and the country of destination j and sector k. The direct energy intensity is the share of energy cost obtained using input-output tables from the Bureau of Economic Analysis, as described by Fetzer (2014). The price differential is taken to be the ratio between the United Kingdom and the United States prices obtained from the OECD.11 The baseline sample consists of more than 940,000 observations corresponding to an unbalanced panel of manufacturing product exports from origin to destination pairs.

What Is Learned from the Results?

The coefficient associated with the interaction between energy intensity and price differential is large, positive, and statistically significant (Table 3.1.1). The baseline point estimate is 0.42 with a standard error of 0.09. The direct energy cost share for manufacturing products slightly more than 5 percent, and the total energy cost share is about 8 percent. In comparison, the direct labor cost share for manufacturing goods is 20 percent. The measure of the price differential between the rest of the world and the United States is of a factor of three, on average.12 This suggests that for the average manufacturing product, U.S. exports have risen by at least 6 percent (0.42 × 3 × 0.05).

Table 3.1.1.
Energy Cost ShareNatural Gas Cost Share
(1) Total(2) Direct(3) Total(4) Direct
Total Utility Share × price Difference0.415***

Direct Utility Share × price Difference0.432***

Total Nat Gas Share × price Difference0.423***

Direct Nat Gas Share × Price Difference0.402***

Adjusted R20.2770.2770.2770.277
Note: Dependent variable is logarithm of the value of product exports at the five-digit level. The specification is a classical equation suggested by trade models and also controls for year, product, and location (destination and origin) fixed effects. The regressions include product level. Standard errors are in parentheses. Nat. = natural.* p < .10; ** p < .05; *** p < .01.
Note: Dependent variable is logarithm of the value of product exports at the five-digit level. The specification is a classical equation suggested by trade models and also controls for year, product, and location (destination and origin) fixed effects. The regressions include product level. Standard errors are in parentheses. Nat. = natural.* p < .10; ** p < .05; *** p < .01.

The results are checked to determine their robustness to using the natural gas cost share as opposed to the energy share and also to the use of year dummies instead of natural gas price differentials; further, oil and petroleum manufacturing products, which have a direct energy cost share above 60 percent, are dropped. The direct natural gas cost share is on average 2 percent for manufacturing products. This measure does not account for the fact that gas could be indirectly consumed through electricity. The baseline results are robust to using those alternative specifications, and broadly similar figures are obtained.

Further evidence suggests that the channels through which cheaper domestic natural gas prices in the United States might have an impact on manufacturing exports are operating at both the intensive margin (expansion by existing firms) and the extensive margin (new firm entry). As more countries exploit new sources of natural gas, it is likely that not only will the geography of trade in energy products continue to change, but that the geography of manufacturing exports will change as well.

Risks From Geopolitical Tensions Between Russia and Ukraine

The ongoing crisis in Ukraine highlights European energy markets’ dependence on natural gas. In January 2009, Gazprom, the Russian energy utility, shut off all supply to Europe through Ukraine. In 2009, the spot gas price increased by 50 percent, but the one-month-forward contract price moved up slowly—by 20 percent—during the three-week shutoff, and crude oil prices did not react noticeably. Europe’s dependence on natural gas transiting through Ukraine subsequently decreased from 80 percent to roughly 50 percent. On June 16, 2014, Gazprom stopped providing natural gas to Ukraine but left the transit and supply to Europe unaffected.

Ukraine and countries in southeast Europe appear particularly vulnerable to potential disruptions of Russian gas supplies. Should the gas cutoffs persist and be extended to other countries, the greatest impact will be on Ukraine and countries in southeast Europe that receive Russian gas transiting through Ukraine—in particular Bulgaria and countries of former Yugoslavia, which rely on Russian gas for virtually all of their import requirements and have only limited access to gas from alternative sources. Other countries, however, will be affected through rising spot prices, which may spread from natural gas to other fuels. Such risks can be mitigated through the accumulation of reserves, purchasing pipeline gas from Algeria and Norway, importing liquefied natural gas,13 or buying Russian gas transported via other pipelines. Other fuels, notably coal and oil products, could also be substituted for gas.

Continental Europe imports a substantial portion of the gas it needs from Russia. In 2013, roughly 152 billion cubic meters of Russian gas were exported to Europe via pipeline, which amounts to 36 percent of European gas consumption. On average, Russia has supplied about 30 percent of Europe’s natural gas needs. Roughly half of the gas supply from Russia is transported via pipeline through Ukraine (down from 80 percent before the Nord Stream pipeline was built). The share of natural gas in primary energy consumption ranges widely among European nations, from less than 2 percent in Sweden to 42 percent in the Netherlands.

The geopolitical tensions in the region have barely affected natural gas and crude oil prices so far. This is less surprising in the case of crude oil than for natural gas because there are significantly fewer concerns about the consequences of a potential Russian oil supply disruption than for a natural gas supply disruption. In May 2014, Russia signed a $400 billion deal to transport 38 billion cubic meters a year of gas from Eastern Siberia to China beginning in 2018. Pricing on the deal has not been disclosed, but the price is thought to be somewhat below what Europeans are paying for pipeline gas from Russia. This gives Russia greater export flexibility should European gas demand continue to fall.


Overall, the pattern of global trade in liquefied natural gas—and in energy more generally—is expected to evolve rapidly. In particular, the United States is now a net exporter of natural gas; Japan is likely to remain the world’s largest importer of liquefied natural gas; and Europe is likely to continue to face uncertainty in its supply of natural gas as a result of the geopolitical tensions between Russia and Ukraine. Energy policy plays a key role in shaping the energy mix, including for coal and renewables, which in turn affects global trade in energy. Specifically, Europe and Japan are at a crossroads, facing a difficult balance between energy security, environmental concerns, and economic efficiency goals. In the medium term, natural gas prices in Asia are expected to be lower, assuming the return of nuclear power in Japan and lower oil prices. European gas prices could edge lower as countries in the region move further toward spot-priced gas imports and index long-term contracts to the spot price, but again the tensions between Russia and Ukraine create uncertainty. Russia has been actively exporting natural gas to Europe in an attempt to prevent U.S. imports from penetrating deeply into the European market. Domestic natural gas prices in the United States are expected to rise with growing liquefied natural gas exports but should remain lower than those in Europe and Asia, given the costs of liquefaction. Natural gas consumers in Mexico also benefit from this situation as they receive low-priced pipeline natural gas from the United States.


    Arezki, Rabah, ThiemoFetzer, and FrankPisch. 2017. “On the Comparative Advantage of U.S. Manufacturing: Evidence from the Shale Gas Revolution.” Journal of International Economics107 (C): 34–59.

    • Search Google Scholar
    • Export Citation

    Celasun, Oya, GabrielDi Bella, TimMahedy, and ChrisPapageorgiou. 2014. “The US Manufacturing Recovery: Uptick or Renaissance?” IMF Working Paper 14/28, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation

    Cohen, Gail, FrederickJoutz, and PrakashLoungani. 2011. “Measuring Energy Security: Trends in the Diversification of Oil and Natural Gas Supplies.” IMF Working Paper 11/39, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation

    Fetzer, Thiemo. 2014. “Fracking Growth.” CEP Discussion Paper 1278, Centre for Economic Performance, London School of Economics and Political Science, London.

    • Search Google Scholar
    • Export Citation

    IHS. 2013. America’s New Energy Future: The Unconventional Oil and Gas Revolution and the Economy – Volume 3: A Manufacturing Renaissance. Englewood, Colorado.

    • Search Google Scholar
    • Export Citation

    International Energy Agency (IEA). 2014. Gas: Medium-Term Market Report 2014—Market Analysis and Forecasts to 2019. Paris. IEA.

    Loungani, Prakash, and AkitoMatsumoto. Forthcoming. “Oil and Natural Gas Prices: Together Again.” IMF Working Paper, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation

    Melick, WilliamR. 2014. “The Energy Boom and Manufacturing in the United States.” International Finance Discussion Papers 1108, Board of Governors of the Federal Reserve System, Washington, DC.

    • Search Google Scholar
    • Export Citation

    Schott, PeterK. 2008. “The Relative Sophistication of Chinese Exports.” Economic Policy23 (53): 5–49.

    U.S. Energy Information Administration. 2013. Cost of Natural Gas Used in Manufacturing Sector Has Fallen. Manufacturing Energy Consumption Survey. Washington, DC.

    • Search Google Scholar
    • Export Citation

    U.S. Energy Information Administration. 2014. Annual Energy Outlook 2014 with Projections to 2040. Washington, DC.

Prepared by Rabah Arezki (team leader), Rachel Yuting Fan, Prakash Loungani, Akito Matsumoto, Marina Rousset, and Shane Streifel, with contributions from Thiemo Fetzer (a visiting scholar to the IMF) and research assistance from Daniel Rivera Greenwood and Vanessa Diaz Montelongo.


Because of the sector’s high capital intensity, natural gas suppliers tend to enter long-term contracts with customers. Prices are indexed to crude oil prices, which introduces rigidities on the price side.


An index of diversification in global gas supplies shows a steady increase in the extent of diversification (Cohen, Joutz, and Loungani 2011).


Natural gas production from shale deposits in the United States began in the 1980s, but the combination of hydraulic fracturing and horizontal drilling allowed gas production to increase sharply in the 2000s (with higher natural gas prices supplying additional motivation). Shale gas production accounts for about half of total U.S. natural gas production. The drilling technology was applied to extract oil from shale deposits in part as a response to high oil prices, and the number of rigs drilling for shale oil has risen sharply.


Both the shale oil and shale gas booms led to lower world average energy prices, the shale gas boom in particular increased the dispersion of regional prices.


Shale gas development has significant potential in many parts of the world, notably in Argentina, Australia, China, Poland, and Russia, where shale gas developments are underway, but also in many other locales. Development of this potential could further shift the patterns of global energy and nonenergy trade. However, shale gas production is expected to rise at a slower pace elsewhere than in the United States, because many of the conditions that facilitated the U.S. shale gas boom are not in place or are in place at an insufficient scale.


With regard to trade, this shift has affected primarily Algeria, Norway, and Russia, the largest gas exporters to Europe.


The fall in liquefied natural gas exports from Trinidad and Tobago also coincided with supply constraints due to maintenance activities on liquefied natural gas facilities.


It is estimated that if the United States were to export at its potential, the U.S. trade deficit would be reduced by more than $164 billion, approximately 1 percent of GDP, in 2020 (IHS 2013).


Using industry-level data, Melick (2014) estimates that the fall in the price of natural gas since 2006 is associated with a 2–3 percent increase in activity for the entire manufacturing sector, with much larger effects of 30 percent or more for the most energy-intensive industries. Celasun and others (2014) find that a doubling of the natural gas price differential in favor of the home country would increase manufacturing industrial production by 1.5 percent.


These results are also robust to an array of checks including additional controls such as country differences in labor costs and GDP. Arezki, Fetzer, and Pisch (2017) present extensive technical details and robustness checks. There are a multitude of factors driving U.S. manufacturing exports that go beyond scope of this box. The interpretation of the present results is, of course, all else equal.


Using benchmarks other than the United Kingdom yields similar results, because the variation in the relative price is coming mostly from U.S. prices.


The price differential is measured as the ratio between rest of the world’s natural gas prices and those of the United States.


Limited imports of liquefied natural gas from the United States began in 2017.

    Other Resources Citing This Publication