We believe, however, the same factors that drove MLP outperformance in the past eight years could continue to drive outperformance in the next eight years.
First and foremost, domestic energy production continues to rise. Recent technological advances have allowed the domestic energy production of natural gas, natural gas liquids and crude oil to increase in the past few years. Natural gas and natural gas liquids production is at an all-time high while on the crude oil side, domestic production is above 8 million barrels per day, or mmbpd, a level not seen in 25 years. Many experts are calling for domestic production by the end of this decade to exceed the record 10 mmbpd set back in late 1970.
Secondly, much of the incremental production is expected to come from areas that have never before produced commercial amounts of fossil fuels. This means the local energy infrastructure likely will need to be expanded, if not built entirely. For example, the Bakken Shale formation in North Dakota and the Eagle Ford Shale in south Texas have both seen an exponential growth in crude oil production in the past five years. Bakken production increased to more than 1 mmbpd now from roughly 100 thousand barrels per day (or mbpd) in 2006. The Eagle Ford's growth has been even more impressive, increasing to more than 1.4 mmbpd today from less than 1 mbpd in 2008. In both locations, the magnitude of the incremental production quickly overwhelmed the existing energy infrastructure, creating potential opportunities for midstream operators (read: MLPs).
The developed world's consumption is expected to stay largely flat over the next 20 years, and industry forecasts suggest that the incremental US energy production will not be consumed domestically. In January of this year, BP released its closely followed global energy outlook survey. It predicted that global energy consumption would rise 41% by 2035, with 95% of the growth expected to come from emerging markets. The US Energy Information Agency, in its 2014 Annual Energy Outlook, predicted the United States will be exporting 9 billion cubic feet of liquefied natural gas by 2035.
We have already seen MLPs propose building natural gas pipelines into Mexico, while natural gas liquids and refined products exports have steadily increased in the past few years. Several liquefied natural gas projects have been approved by the U.S. Department of Energy and are in various stages of development. Sources suggest the first of these should begin exporting natural gas by the middle of next year, with several more projects expected to come on-line over the next several years.
As for crude oil, exports from the United States are effectively banned. While we do not expect to see the crude oil export ban lifted anytime soon, the spike in domestic production from the Bakken and the Eagle Ford formations has created a crude oil supply/demand mismatch. The crude oil produced in these two plays is lighter and sweeter than the oil most refineries are calibrated to process. In fact, many refiners spent the last decade adjusting their plants to process the heavier, sour crude oil produced in Canada, Mexico, Venezuela and the Middle East because conventional wisdom was that the world had run out of light, sweet crude.
Conventional wisdom appears to have been wrong, as demonstrated by the roughly 2 mmbpd of incremental production from Bakken and Eagle Ford. There are now roughly $45 billion in announced crude oil expansion projects, the overwhelming majority of which are being built by MLPs.
Whether the goal is to bring natural gas, natural gas liquids or crude oil to market, all of these projects will require capital, and a fair amount of it. The Interstate Natural Gas Association of America recently released an update to its 2011 midstream energy infrastructure spending forecast through 2035. Using constant 2012 dollars, INGAA now projects aggregate infrastructure spending to reach $641 billion, or $29 billion annually. This is in stark contrast to the association's 2011 estimate of $261 billion, or $10.5 billion annually. Most of the forecasted increase in spending is related to crude oil infrastructure, which climbed from $1.4 billion per year to $12.4 billion, a figure we believe is far more realistic than the laughably low previous estimate.