9 thoughts on “Podcast on econometrics, oil shocks, and monetary policy

  1. rtd

    Definitely an interesting edition of a good podcast.
    I’m wondering if you could expand on your thoughts regarding r* and implications for monetary policy. You mention the obvious issues of a Taylor Rule if, as you suggest, r* is non-existent. I’m interested in your thoughts of an optimal monetary policy regime (I was astonished Beckworth didn’t ask about his preferred regime of NGDP targeting).

    Also, great job emphasizing the differences in theory vs empirics (Woodford’s work, etc) in research and particularly as it relates to policy.

  2. Steven Kopits

    Well worth listening to the podcast.

    A few points I would add:

    – Since 2005, US shale production growth is responsible for 60% of global oil supply growth and is expected to maintain this contribution share through 2019. This represents incredible supply dependence on a single source.
    – Since 2005, Canada is responsible for about 15% of oil supply growth, thus the US and Canada together account for 75% of global oil supply growth in the last 14 years.
    – OPEC has added only 2.8 mbpd of supply since 2005 (on current consumption of 100 mbpd). I would note that almost all of this came in the collapse of oil prices after mid-2015. That is, collapsing oil prices forced OPEC to focus on increasing production, rather than increasing prices, to maximize revenues. Had oil prices not collapsed, OPEC’s production growth may very well have been less.
    – All other non-US, non-OPEC supply growth, net, amounted to a paltry 1.6 mbpd in the last 14 years, and is actually lower today than it was in 2015.

    Without the US contribution, supply growth would have averaged 0.5 mbpd / year over the last 14 years (to y.e. 2019). Typical organic oil demand growth might amount to, say, 1.3 mbpd / year. Thus, without the US, the conditions which pertained from 2011 to mid-2014 would have persisted, with oil prices routinely above $100 / barrel.

    I would also note that US supply growth in the last 12 months is an impressive 1.6 mbpd. US supply, over the last three months, is growing at a 2.5 mbpd / year pace. Despite this, oil prices are high and rising and the futures curve is in historically steep backwardation, which is drawing oil out of inventories and into current consumption. In other words, implied demand growth appears to be running at 2.0 mbpd / year, and perhaps a good bit more. As robust as US supply growth is, it is not enough.

    So, for those who brush off peak oil as a weird anomaly, keep in mind that the global oil system remains shockingly dependent on the single trick pony of US shale production growth. When that pony tires, oil prices will shoot right back up, and significant part because OPEC + Russia are revenue, not volumes, maximizers. They want a higher oil price and lower volume growth. It is nice to be taking a breather from high oil prices, and we may enjoy solid supply growth for the next several years. But the oil supply will see adequate growth only as long as US shales remain resilient. The peak oil story is otherwise not much changed over the last 14 years.

    1. Jeffrey J. Brown

      And depletion is never stops. Just to maintain current US Crude + Condensate (C+C) production, US producers probably need to put on line about 2 to 2.5 million bpd of new production every year, roughly the equivalent of replacing Mexico’s current production every year.

      And of course, US C+C production has a very high condensate and very light crude oil content. Based on EIA data, about 43% of US lower 48 C+C production exceeds the maximum API gravity for WTI crude oil, 42 API gravity, i.e., about 43% of US Lower 48 C+C production can’t be sold as WTI crude oil, because the API gravity is too high.

      Globally, I suspect that virtually all of the post-2005 increase in global C+C production consists of condensate (generally defined as “Crude oil” with an API gravity of 45 API or higher), and not actual crude oil. A sketch showing the post-2005 gap between the rates of increase in global gas production and global C+C production (note the inflection point in 2005):


      An April, 2018 article on the topic of very light crude and condensate:

      The wrong kind of oil is flooding the US market — but that could be great news for a handful of producers


      Super-light crude is flooding the US oil market, and there’s little demand to meet it.

      All of the industry’s growth in the US over the last year was thanks to crude with a gravity above 40 on the American Petroleum Institute’s scale, which measures the weight of a petroleum liquid compared to water, according to analysts at Morgan Stanley.

      That’s a problem for domestic shale explorers. Most refineries in the US are designed for heavier crude grades, around 32 API. And refiners are running out of room to process super-light shale without seeing losses.

      “Domestic refiners cannot take much more of this and are close to hitting the ‘shale wall,'” the analysts said.

      Options to export what US refiners don’t want are limited. Demand for superlight crude outside of the US is modest.

    2. Jeffrey J. Brown

      Incidentally, I think that the real limiting factor regarding condensate demand is not necessarily what refiners can process, but the percentage of the total Crude + Condensate (C+C) input that has to be actual crude oil, in order to meet the demand for the full spectrum of refined products.

      In any case, it’s important to remember that when we ask for the price of oil, we get the price of actual crude oil, generally Brent or WTI, but when we ask for the quantity of oil, we get the volume of actual crude oil plus some combination of partial substitutes–condensate, natural gas liquids (NGL) and biofuels.

      I suspect that actual global crude oil production (45 API gravity and lower crude oil) has been on an “Undulating plateau” since 2005, while the volume of partial substitutes–condensate, NGL and biofuels–has continued to increase. Note that condensate and NGL are of course byproducts of natural gas production.

      And note that based on the most recent four week running average EIA data, US refiners are still dependent on net crude oil imports for 38% of the C+C inputs into US refineries.

      1. Steven Kopits

        You need to adjust for rising product exports, Jeffrey.

        You can see to total net import number on line 46, tab4a of the STEO.

        This shows US net import dependence falling from around 25% in 2016 to around 9% by year end 2019. The US could conceivably be a net oil exporter by year-end 2020.


        1. Jeffrey J. Brown

          Of course, a lot of product exports are the result of exporting imported crude oil.

          1. Bellanson

            I have a question about API gravity above 45: does that mean that we can’t produce diesel (or less diesel), but we still can produce gasoline , or does it mean that we really only can do increasing proportions of propane and similar?
            How serious is the situation for gasoline supply?

  3. 2slugbaits

    The interview covered a lot of ground, but I was disappointed that David Beckworth didn’t ask JDH what I thought was the obvious question. Why haven’t oil prices followed a Hotelling rule? If you’re going to talk about oil prices, peak oil and interest rates, this seems like a natural question to ask.

Comments are closed.