Links for 2014-01-19

Quick links to a few items I found of interest.

Scott Irwin explains the facts to the New York Times.

Greg Mankiw summarizes why Stanley Fischer is a great choice for vice chair of the Federal Reserve.

Tyler Cowen finds it “amazing how the single largest extension in the responsibilities and fiscal obligations of the Irish state– ever– has been turned into a PR defeat for the idea of fiscal conservatism.”

Shell Oil’s profits are declining as it copes with lower production from old fields and increasingly expensive and risky sources for producing new oil.

And here are some useful evaluations of the implications of shale gas development for property values and global warming.

StumbleUponLinkedInReddit

17 thoughts on “Links for 2014-01-19

  1. jonathan

    I hope that Scott Irwin does well out of the (unfortunate) publicity brought on, I believe, by the arrogance of the other guy – won’t mention name – in blowing off the NYT’s request for information. Some notoriety is bad, some is good so this may work out for him.

  2. Steven Kopits

    Re: Irwin. The weakness of Irwin’s, or your or Lutz’s work is that the result is essentially negative: price spikes are not the result of speculation. But none of these answer the question, well then, what is the cause of such price surges?
    It seems to me, as I’ve written elsewhere, that if there is a differentiated demand shock in the presence of price inelasticity of demand, then you could see an endogenous price overshoot. To put it another way, if the Chinese want to buy more oil, the Americans are unwilling to yield their incumbent consumption, and the supply is unresponsive–as was the case in the period leading to 2008–then you could see a de facto bidding war for that oil. Thus, in this view, a price spike can develop purely from demand-side dynamics, without the need for financial speculation as a factor.
    And such a spike is resolved with the defeat of one of the parties, that is, in the bidding war, one side eventually concedes defeat, with demand toggling from inelastic to elastic mode, from “fight” (“animal spirits”) to “flight” (“depression”). And if this is true, then all the work that you and Lutz have done on elasticity is only applicable to periods of price inelastic demand. Your work is good, but it reflects a special case, not the general case.
    And if that’s true, then effect of oil on the economy during periods of price elasticity will occur through volumes, not prices. And the Fed does not see this at all, because they are mistaking the specific case for the general case.
    In any event, the hypothesis of an endogenous price spike should be amenable to empirical testing and would be a nice assignment for a PhD student.

  3. Steven Kopits

    Re Shell: You’ll recall I used the word “catastrophic” to describe Shell’s plans to dispose of 17% of its assets. Well, I think we’re right on track for catastrophic now. Of course, Shell’s numbers are adversely affected by temporary outages in its LNG production (which is not oil), but I think the underlying concern continues to be cost pressures.
    We’ll have to see whether these results generalize to the industry more broadly. In the capex compression model, it was clear that the oil companies would be pushed off their projects. What was not clear was whether the oil majors’ production decline would be gradual or sharp. With Shell’s results, sharp seems to be more likely, and consistent with slide 50 of my Woodrow Wilson School presentation.
    For those interested, I’ll be presenting an updated version of the deck at Columbia’s SIPA, Room 1512, Tuesday, Feb. 11, at 12:30 pm. It’s open to the public.

  4. Steven Kopits

    Re: Greenhouse gases. The article speaks of the role of methane. Methane is thought to decay into CO2 over a ten year period. Thus, it has only a transient effect.

  5. Nick G

    Steven,
    such a spike is resolved with the defeat of one of the parties, that is, in the bidding war, one side eventually concedes defeat, with demand toggling from inelastic to elastic mode, from “fight” (“animal spirits”) to “flight” (“depression”).
    This is an unnecessarily negative view. $100 per barrel is simply more expensive than the alternatives, for the majority of uses for oil in the US.
    But, it takes time for people to decide that high prices are permanent, and then to implement the alternatives. It takes time for automakers to ramp up hybrids, PHEVs and EVs and gain economies of scale. It takes time for manufacturers to redesign bottles to use less plastic, or to switch to different feedstocks entirely. It takes time for shippers to convert trucks to NG, or to switch from trucks to rail.

  6. Steven Kopits

    Nick -
    The regime change to price elastic demand is a short term phenomenon, occurring in a periods of weeks to a few months.
    We are talking about the impact of pessimism (“flight”) or optimism (“fight”) during a kind of auction process.

  7. Nick G

    Steven,
    .The regime change to price elastic demand is a short term phenomenon, occurring in a periods of weeks to a few months.
    Do you have any evidence for this (links to analysis/articles)? It doesn’t make sense to me.
    First, oil prices have been low for generations with relatively short spikes. The secular rise since 2004 is new in our historical memory. Further, there’s enormous misinformation being disseminated about this price rise being temporary due to the Arab Spring, or shale/tight oil.
    Second, many adaptations require large investments of managerial time and capital expense. UPS and FedEx aren’t going to change their trucks overnight – The first thing they went to were low-cost operational changes, like improved routing. Even now, they’re still in the pilot phase of a number of alternatives: CNG, hybrids, hydraulic hybrids, etc.
    It took 10 years for Prius sales to be really large. Tesla, Leaf and Volt sales are growing much faster, but change still takes a while: customer education, supplier development, etc., etc.
    We are talking about the impact of pessimism (“flight”) or optimism (“fight”) during a kind of auction process.
    Doesn’t it make more sense to talk about normal market dynamics, where better and cheaper alternatives replace market incumbents?

  8. ivars

    I have been thinking lately about negative nominal interest rates and how their compounding would cause money supply contraction as interest will grow exponentially negative in nominal terms.
    1) And of course, as QE is a continuation FURTHER of ZIRP policy, it has to be lending with negative nominal rates. How-easy:
    QE creates excess reserves of banks at FED = FED purchases of securities, so 85 billion/month
    FED pays 0,25% interest to banks on these reserves;
    So FED is lending to banks 85 billion per month at negative NOMINAL interest rate of -0,25% /year as its paying to banks to take the money
    That is DEFLATIONARY action unless the speed of printing exceeds the depreciation of money supply due to interest rates; so far it exceeds it;
    But in general, its revolutionary; negative nominal interest rates mean that money can not be invested with profit, so money is not needed. As a result of this development, sooner or later REAL money supply will start to contract as negative nominal rates become common.
    2) Bail- in is also negative nominal interest rate phenomena, as, in fact, it is tax on deposits. That means the depositors are PAYING the bank to have their money. It is the same phenomena in principle, difference being that FED uses taxpayer money to lend at negative interest rates via USG treasury purchases, EU will use depositor money .
    Negative nominal interest rates are here to stay, and that means deflationary spiral worldwide and very long one.. Negative nominal interest rates mean that history is moving backwards like positive nominal interest rates mean bringing future demand and population into today, and fixing the future.
    3) Government printing of money which we still have to see ( Modern Monetary Theory) is a short -relatively short excursion in negative nominal rates territory where government must print money faster then negative rates (hidden in govt) eat real money supply up to keep real money supply constant. Such printing will follow exponential increase as that is how interest rates compound, in order to compensate for exponentially decreasing real money supply due to negative interest rates. It will continue until the cost of transaction with this money per minimal or average time of transaction will exceed the value of money transacted. The fact no one wants to hold this money is exactly the idea of negative interest rates in action. With negative nominal interest rates ( implied) You will be trying to get rid of money as fast as possible, driving inflation up= but that would be a consequence of negative nominal interest rates in govt printed money.
    I like to place things on same basis so that one can use one concept- that of interest rates- both positive and negative- to explain many differently worded concepts. Since usury is what drives development, negative usury drives decline.

  9. Steven Kopits

    If you read a study by Jim or Lutz Killian, you will find that they calculate the price elasticity of demand for oil at something like -0.2 to -0.3. Price elasticity is dQ/Q/dP/P, or the percentage change in volume demand for the percentage change in price. Thus, if price goes up by 10%, then demand will only fall by 2-3%. If that’s true, then the consumer must have accepted a 7-8% price increase!
    Why would a consumer do this? For one of four reasons:
    - the consumer believes the price increase is transient
    - the consumer believes that his income will rise to cover the increased cost of oil
    - the consumer is willing to consume less of another good
    - the consumer is willing to borrow or save less in order to maintain his oil consumption
    If these conditions are met, then the consumer will fight to maintain consumption, that is, they will accept a price increase rather than reduce consumption pro rata to maintain a constant budget constraint. And that’s the normal behavior in the historical record.
    However, at some point, consumers will accept a given price as permanent, lose faith in income growth, run out of debt capacity, and be unwilling or unable to further reduce other consumption expenditures. At that point, the consumer will not fight to hold on to oil consumption. Rather, when faced with a price increase, the consumer will simply roll over and let the competition take it. This is an attitude of depression, of “flight” rather than “fight”.
    In such an event, we would not see a price spike. Rather, we might expect to see declining oil consumption even in the face of flat to declining oil prices–which exactly describes what’s happened to the OECD economies since the price “lock-out” of late 2010. European and US oil consumption fell from April 2011 to April 2013 (in the case of the US) or to the present (in the case of Europe). That’s not an elasticity of -0.2, as Jim or Lutz would have it from the historical record. It’s an elasticity of -2% volume/-2% price, or 1.0–which of course is an unintelligible number, but as a practical matter, it’s more than -1.0. This would suggest Jim’s and Lutz’s numbers are wrong, but of course they’re not–they’re literally drawn from the historical record. Thus, something different is going on now, and I’m suggesting that oil demand has been evolving in a different regime, one of price elasticity, which accounts for the stark deviation from the historical record.
    Of course, “flight” is adaptation. It’s learning to do without. Doing without means finding substitutes, reorganizing your life to need less of the constrained commodity, or simply doing less. And we see all those kinds of adaptations. However, the only which is available immediately is “doing less”. Thus, by this approach, “flight” is associated with depression and passivity in the short to medium term, and more radical adaption (eg, electric cars) over the long term–which is the time frame you’re referencing, Nick. When you’ve done enough adapting and begin to have faith in the future again, behavior toggles back into “fight” mode.
    I believe, but do not know, that the US has begun toggling back to price inelastic demand mode in Q4 2013. That’s what the product supplied data suggests, but I think I need to see more data to have a firm opinion.
    To the best of my knowledge, I am the only person to have suggested this framework of analysis. Jim has certainly discussed elements of it, related to his inoculation theory (as I call it). He has not, to the best of my knowledge, tied the inoculation theory to price elastic or inelastic behavior, although it follows from his work by implication, I think.
    Now, my hypothesis is essentially speculative, but it’s consistent with the data. It would be nice to see if such findings could be replicated (or refuted, as the case may be) through some empirical testing.
    There is more here in a gated article: http://www.europeanenergyreview.eu/site/pagina.php?id=3761

  10. Nick G

    Steven,
    Again, I’m baffled by the negativity of this approach. How is moving from a 23 MPG vehicle to a similarly priced 46 MPG vehicle a depressing retreat??
    A Tesla is simply a better car: cheaper, faster, with better handling. There is no sacrifice involved. A Volt also has better handling and acceleration than it’s price peers, with 10% of the fuel consumption. All of the taxi drivers I’ve interviewed report that they won’t give up their hybrids without a fight – they’re simply better than their previous cars, even without the efficiency.
    Now, of course, sometimes lower fuel costs will be partly offset with higher capital costs, as the alternatives improve, the overall lifecycle cost will fall. Already, those costs are substantially lower with hybrids and PHEVs.
    Now, I think you need to clarify your analysis of elasticity: you’re mixing up short-term and long-term. Just as important, historical analysis from before 2008 doesn’t apply: there really isn’t any data on long-term high prices before, well, now. Further, elasticity isn’t linear: the effect of moving from $25 to $50 is very, very different from moving from $50 to $100: you’ve reached a tipping point, where prices are higher than the alternative.
    So, we’re currently seeing falling oil consumption combined falling prices, which makes perfect sense: oil prices are higher than the cost of alternatives, so substitution continues to take place even as oil prices fall modestly.
    So…this meme of “doing without” appears to me to be public relations for oil companies. The idea that nothing is as good as oil, and reduced oil consumption means suffering is as silly as the statement from Charlie “Engine” Wilson that “What was good for the country was good for GM, and vice versa.”.

  11. Steven Kopits

    “How is moving from a 23 MPG vehicle to a similarly priced 46 MPG vehicle a depressing retreat??”
    If the change is purely through efficiency gains, then there is no retreat. But in today’s world, you’re giving up something: size, performance or safety. It’s not a Pareto optimal trade-off. It is a defeat, in the sense that the consumer was not able to maintain prior, preferred levels of consumption.
    “Depression” here is not used as a clinical word (although a lot of people tend to be depressed during a depression). Rather, depression here means specifically that the consumer is unwilling or unable to maintain a preferred consumption bundle and will not fight to keep it.
    So, imagine your girlfriend wants to walk out on you. You could say, “No, you have to stay!”, or you could say, “Well, if that’s what you want.” In the first case, there could be a huge argument, screaming and nasty. In the second case, the whole thing could end without a sound. So, to take the oil analogy, in the first case we see a price spike followed by the defeat of one of the parties (she stays or she doesn’t for the example above). In the latter case, there is no fight, but the relationship is broken, and the guy is the clear loser. (She walked out.) This is the equivalent of the damage being done by volumes, not prices. The fact that there was no argument does not mean that the relationship did not end. The Fed, to continue the analogy, is the neighbor listening for the yelling through the wall. If there’s yelling, then there’s trouble in the relationship. I am arguing, by contrast, that if the guy is unwilling to defend his claim, then the relationship will end without a sound–and the neighbor will miss it if yelling is what they’re focusing on!
    That’s one of the points I’m trying to make.
    So, we’re discussing the pre-conditions for each response, in our case, related to increasing oil prices. Jim and Lutz and Irwin and others have consistently noted that financial speculation is not a cause of price spikes. But no one, to the best of my knowledge, has proposed a mechanism of how you produce price spikes well above the carrying capacities of the respective economies. I am proposing one.
    That’s what I trying to get across.

  12. kharris

    Cohen is right – framing matters. That makes his own use of framing all the more curious.
    His point is that, once bank debt has been guaranteed, we need to figure out whether austerity in the near term is better overall than austerity over a longer time in paying off the debt. He has skipped an important question – is paying off the debt over any period of time the best option? Can we even assume that austerity over any reasonable period of time will allow servicing of the debt taken on by the sovereign to guarantee bank creditors? If the best answer would have been to repudiate banks’ debt, but that answer is no longer available, why isn’t repudiation of sovereign debt an option? The domestic welfare-maximizing option may be default. Transferring the debt from banks to the sovereign is bad business, but once the deed is done, we still need to determine the welfare maximizing option. Options are not limited, as Cohen’s framing suggests, to paying quickly (austerity now) or paying slowly (austerity over the long term). Once debt grows sufficiently large, the welfare maximizing option is default.

  13. Nick G

    Steven,
    I certainly agree that there’s a difference between short-term elasticity and long-term elasticity: people will wait a little to see if fuel prices will stay high, before taking the trouble to address the problem. For a little while, they’ll just put the extra fuel bills on the credit card, or eat out a little less often.
    But, if prices persist, people will start to find solutions to the problem. Cars in the US are traded, on average, every 3.5 years: that means that on average it will take a few years for almost everybody to change vehicles. At that time, they can easily get something that uses less fuel.
    The average US vehicle gets 22MPG. It’s easy to trade to something that gets 44MPG. For the same price it may be a little smaller: maybe a sedan instead of an SUV. On the other hand, the average new car goes for more than $30, while a Prius-C (50MPG) costs $20k, so you may well pay less for your new vehicle, while also reducing operating costs.
    Let’s explore a business case: a sedan driver who goes hybrid. They might go from 25MPG and 16 cents per mile (at $4/gallon) for fuel cost, to 45MPG and 9 cents per mile for fuel costs (equal to paying $2.22 per gallon with the 25MPG car). The hybrid might cost $5k more, have $1k less in lifetime maintenance costs (mostly brakes), and reduce fuel costs by about $10k (9 cents per mile over a lifetime of about 140k miles). In effect, we go from 16 cents per mile to 12 cents per mile, while reducing time spent filling up and doing repairs. In every way, this is an improvement, not a “defeat” or retreat.
    Now, it’s true that costs haven’t been reduced to the status quo ante, when gas was dirt cheap. On the other hand, if all of the external costs of gas are included (at least $1 per litre), the total cost of the new vehicle is indeed lower than for oil, even compared to when the out of pocket cost was $25/bbl.
    Finally, hybrids, PHEVs and EVs will continue to get cheaper, and society will continue to acknowledge the reality of external costs, until it becomes blindingly obvious to all that elimination of oil as a fuel will in fact make us more prosperous, rather than constitute a net cost.
    So, the girlfriend analogy only works if the girlfriend is the evil mistress in “Fatal Attraction”. The hero is far better off with his loyal wife…

  14. Nick G

    Let me sharpen that point about elasticity: it makes perfect sense for prices (and volumes) to “overshoot” their long-term equilibrium level: it takes time for consumers to respond, and suppliers of alternatives to ramp up production.
    It is, in fact, little different from the classic “boom and bust” cycle that dominates commodity markets, it will just take longer, and the new supply will be a little less familiar.

  15. Steven Kopits

    Or, maybe, Nick, if you can’t get enough of a critical commodity, your society just fades away. Historically, that commodity was more likely water. Sometimes it was trees and top soil, like Easter Island (if I recall correctly). Sometimes it was temperature, like Greenland. It doesn’t always get better.
    In any event, a society has any number of means of adapting. But it takes time. In the meanwhile, you’ll have under-employed factors. One of those adaptations, by the way, could be having fewer children.
    But you’re right, for any given budget constraint of oil, society will learn to use it better over time, and growth will resume. This historical record is very clear on this matter. However, if you have to reduce consumption continuously, then it’s a race between declining oil consumption and efficiency gains, with GDP growth likely to be relatively poor during that period.
    And then we’re back to, say, dozens of comments I’ve made over the last few years. It’s the pace of efficiency gains that matter. But that’s all long run stuff.
    In the short run, it’s flight or fight that matters.

  16. Nick G

    Steven,
    if you can’t get enough of a critical commodity, your society just fades away
    Well, I’m trying to educate your intuition about this – oil just isn’t critical in the same way as water, climate or topsoil. The US grew faster on wood and coal from 1800-1850; growth slowed on coal from 1850-1900; it slowed further as oil grew, from 1900-1950; and it’s been even slower since WWII!
    We’ve built a transportation infrastructure than uses oil, but it won’t be hard to convert: PHEVs use similar factories, they drive the same (or better), they use the same roads and garages, etc., etc. Interior illumination in 1890 was built around kerosene, but electric lights
    were growing fast, and 15 years later kerosene was obsolete for lighting (just as we’re now seeing in Africa!).
    I’m not sure where this “Peak Oil as The End Of The World As We Know It” meme came from. Perhaps from the fatally flawed Club of Rome LTG model. In any case, it’s deeply unrealistic.
    adapting…takes time. In the meanwhile, you’ll have under-employed factors. One of those adaptations, by the way, could be having fewer children.
    Only if you’re blindingly misinformed (perhaps, if you watch Fox News…). Eliminating expensive oil is easy. Buy a Prius-C, for 65% of the average new car price. Buy a plug-n Prius, for 75%. Or, a Leaf, for 80%, or a Volt, for 90% of the average. Or, buy a used hybrid. These solutions are cheap, fast and effective.
    Why would anyone choose not to have children, when they can eliminate the drain on their finances so incredibly easily??
    if you have to reduce consumption continuously, then it’s a race between declining oil consumption and efficiency gains, with GDP growth likely to be relatively poor during that period.
    Well, the US reduced it’s oil consumption from 1978 to 2009, while growing by 150% – growth of the GDP:oil ratio of 3% per year. And that’s without really trying: CAFE regulations were frozen in 1990, and US oil prices are far below their real cost.
    it’s flight or fight that matters.
    No, it’s just choosing a cheaper and better alternative. Right now, the primary barrier is political: the various institutions built by the oil industry (in the form of the Kochs) has brainwashed many people into rejecting Volts as “Obamacars”, and Priuses as a threat to…something, I’m not sure what.
    Further, the real cost of oil is hidden, and necessary fuel taxes have been demonized: again, wouldn’t you agree that external costs are real??

Comments are closed.