EconoMonitor

Are Either Low Interest Rates or Speculation Raising Holdings of Oil and Other Minerals?

Everyone is looking for someone to blame for high prices of oil and other mineral and agricultural commodities. Speculators (among others) are high on the list, followed by the Federal Reserve. While I don’t think blame is necessarily the right concept here, I have been arguing that low real interest rates have worked to raise real commodity prices through a number of channels. Each of these channels could be called “speculation,” if speculation is defined as behavior based on expectations of future prices.A number of commentators, including Don Kohn and Paul Krugman, have argued that low interest rates and speculation cannot be the sources of the problem, because oil inventories are low. It is true that low interest rates, other things equal, should in theory increase firms’ desire to hold inventories.UScrudeoilInv1998-20080611.jpg

US crude oil inventories do not appear to be especially low in the graph above, showing June 1998-June 2008 (from Bloomberg). But it is true that they are not especially high either.

We are talking about relatively integrated world markets, however, so it is world inventories that should matter most. According to the International Energy Agency’s Oil Market Report, oil inventories held in developed countries have been above average during most of the last year, as the next graph shows.OilInventoriesOECD07-08.pdf They rose sharply in January 2008, which happens to be the month when the very aggressive cuts in US interest rates took place.OilStocksOECD07-08.jpg These numbers are far from conclusive, but still…

OilStocksOECD07-08vsLR.jpg

The theory is meant to explain the mystery why prices of virtually all mineral and agricultural prices are high, not just oil, and in some ways fits others better. Inventories of some commodities are indeed high now. The price of gold, the last graph shown, is a good example. Here the evidence supports the theory (1) that easy monetary policy has driven up the price, and (2) that one channel is low interest rates making it more attractive to stockpile the yellow metal. But, as with oil, the biggest inventory is the one underground.

USgoldstocksHi99-20080611.jpg

[Thanks to Pravin Chandrasekaran.]


Originally published at Jeff Frankel’s Weblog and reproduced here with the author’s permission.

20 Responses to “Are Either Low Interest Rates or Speculation Raising Holdings of Oil and Other Minerals?”

GuestJune 17th, 2008 at 4:22 pm

So if sharp interest rate cuts are contributing to high commodity prices, how would they respond to the Fed being on hold? or tightening. Would you expect a sharp correction?How does the increasing liquidity pumped into the system from the asset swaps and central bank injections exacerbate the effect of monetary easing – it seems like the commodity markets have captured a lot of these flows.

AnonymousJune 17th, 2008 at 8:21 pm

Very well argued analysis; convincing evidence that part of the commodity price increase is driven by low interest rate and easy liquidity. I am quite convinced.rn

akmiJune 17th, 2008 at 9:30 pm

Hotelling lives!…at least with respect to oil. Since oil exporters (namely middle east, especially swing producer Saudi Arabia) tend to invest (a decreasing but still large) share of their oil windfall in Treasuries, the low yield environment since 2004 gives a dis-incentive to produce oil now when Treasury investment yields are low.

Jeff FrankelJune 19th, 2008 at 5:11 am

I am pleased to see that my first few hours as a RGE analzst blogger has produced four comments! Yet further confirmation that Nouriel has built a site with maximum dissemination and exposure for international economics.Furthermore, these are good comments. The second one (from anonymous) is especially perspicacious. In an answer to the first question: yes, on a day when the news points to Fed tightening (which seemed to be the case over the last few days when Ben Bernanke emphasizes the danger of inflation), I would expect the decline in real interst rates to be accompanied by a decline in commodity prices. But of course monetary policy is not the only important determinant of commodity prices, especially if one takes the sectors, such as oil, one-by-one.In response to Akmi: yes, your are right, it is just Hotelling.And to answer the fourth guest: Yes, but I would phrase it slightly differently: To the extent that low US interest rates contribute to low interest rates worldwide, it puts upward pressure on commodity prices worldwide (e.g., expressed in a basket of currencies). In addition, low US real interest rates puts downward pressure on the real value of the dollar. If you put the two propositions together, then low US interest rates put especially strong upward pressure on commodity prices when expressed in dollars, resulting in an increase in BOTH components of the product: (dollars per world currency units ) x ( world currency units per barrel of oil).

GuestJune 19th, 2008 at 1:20 pm

So will commodity prices fall now that many central banks will have to raise interest rates? Hopefully so even if it is a mixed blessing.

AnonymousJune 20th, 2008 at 9:46 pm

As an economist working on energy, I have to say that I am more than a little bit frustrated by this post. When one says, “speculation is defined as behavior based on expectations of future prices” one is essentially defining almost all rational economic behavior as “speculation”. As Humpty Dumpty famously explained: "When I use a word.. ..it means just what I choose it to mean–neither more nor less." This definition of speculation is far broader than the traditional definition. The more traditional definition is that people believing that prices will rise, buy at the current, lower price, and store the commodity; hoping to sell later at a higher price. By competing with current users, they bid up the current price, store (hoard — to use the pejorative) and ultimately sell at a higher price (if they are correct). However, they depress the future price by adding more supply when it is most scarce. This is a useful function, chock full of value-added and vitamins: It smoothes and stabilizes both price and consumption. Without somebody “hoarding” we’d have to buy and eat the whole harvest by Thanksgiving. It is true that sometimes the process gets out of hand and speculators believe that the price response to their own purchases is a fundamental trend. Then they buy too much: A bubble is created, and the whole process eventually crashes when the inventory overhang eventually has to come back on the market. For the speculators this implies huge losses and natural punishment. But that is clearly not happening in oil and most other commodity markets. Agricultural inventories are mostly at multi-decade lows. Most LME metals inventories have plunged since “speculators” bought in during the last recession. The big price increases started when the “speculators” stopped selling. As for oil, the picture painted in this posting is a little naïve to oil market professionals. It’s the year-on-year and week-to-week trend that is just as important as levels and is watched by the markets at 10:30 every Wednesday morning when U.S. data is announced. If inventory is down year-on-year, you are using more than you are producing and importing. U.S. crude inventories are down 50 million barrels. If we do that one more year, we will be down to minimum operating inventories — that is not sustainable. And the trend is going the wrong way when seasonal inventories are supposed to be building. You have to set the burden of proof pretty low to tell a traditional speculation story. The inventory data doesn’t support it the way the DNA evidence didn’t work for Mike Nifong. Now, if there is another explanation (aside from the fundamental story that the world seems to be using more oil than it’s producing), let’s describe it using a different word. Say, for example, it’s “loose monetary policy”, “strawberry root fungus”, or “disrespectful teenage behavior.” Then provide proof of those theories. But don’t use the word “speculation” – that word already has a meaning. If you have a good and original theory, describe it with words that won’t be confused with something else.

JuanJune 22nd, 2008 at 12:37 am

Inventories are also influenced by the shape of forward curve, which does have to do with interest rates and expectations.

Aaron GoldzimerJune 24th, 2008 at 8:44 pm

Dear Prof. Frankel:I understand the claim that low interest rates induce lower production, but are you also claiming speculation as a channel? If so, speculators (in the futures markets) have to sell to someone when the contract expires – or store it. Given that we can’t make a case that there’s a lot of storage going on, the market is (we are) buying and burning the oil at these prices. So how can we say that speculation (apart from reducing production) may be a channel?Thank you.Aaron

Jeff FrankelJuly 8th, 2008 at 8:50 pm

Reply to Aaron: (1) My core argument assumes that speculators generally expect prices to move toward fundamental equilibrium in the future (so that speculation is stabilizing rather than destabilizing). I.e., I am not yet ready to sign on the proposition that we have gone into a bubble phase. (2) It is true that if low real interest rates are the cause of high real oil prices, someone has to end up holding the oil. (The same would be true if destabilizing speculation were the cause.) The theory predicts that inventories would be higher than usual, and this is not clearly the case with oil. So, this is indeed a problem for my theory. (It does work for gold currently — high inventories — but admittedly not for wheat.) But the oil held underground in exporting countries dwarfs the oil held in tanks. Thus it is here, underground, with Hotelling, that I make my stand.Reply to Anonymous: It was not my intent to use an idiosyncratic definition of speculation. But please explain why you are sure that your, narrower, definition, is the correct one. You say that the person must physically buy and store (or "hoard") the commodity, for it to qualify as speculation. I believe that most financial traders in Chicago, New York or London who buy and sell spot or futures contracts, have never taken physical or even legal possession of an actual barrel of oil or bushel of wheat in their lives. Am I wrong in this belief? Or do you think it doesn’t qualify as speculation if someone buys a futures contract in the hope that the price will rise? JF

Marcus AnonymousJuly 8th, 2008 at 11:30 pm

JFMy point was that speculation in commodities is not the same thing as speculation in “commodity futures”. I am addressing the topic raised in your introductory sentence: “..high prices of oil and other mineral and agricultural commodities.” In and of itself, no one would care about the price of oil futures unless it affects retail prices of oil products. I agree that speculating in oil futures is a form of speculation. I strongly agree “that most financial traders in Chicago, New York or London who buy and sell spot or futures contracts, have never taken physical or even legal possession of an actual barrel of oil or bushel of wheat in their lives.” That is true, but not relevant to the topic at hand. I do not agree that you can conflate “oil futures” and “oil”. You cannot make gasoline out of oil futures. They are paper. You can recycle them into cardboard or textbooks. Does that mean there is no link between the two? I strongly agree that the price of oil futures depends strongly on the price of oil. Can you make the argument that the relationship works the other way, i.e. that the price of oil futures causes the price of oil? I would argue only in the limited case that a steep contango (which has reappeared in the last couple of weeks) serves as an incentive to build inventory. However, in the absence of actual inventory building, that relationship is only theoretical. I would argue that in the absence of both a persistent contango and simultaneous inventory build, there is no evidence that speculation in either physical oil or oil futures has caused the increase in oil prices over the last year. So while I agree that you can speculate on oil futures, I just have not seen a quality argument that speculating on oil futures causes the price of petroleum (liquid fossilized hydrocarbons) to increase. Faithfully, but not actually,Marcus Anonymous

Jeff FrankelJuly 15th, 2008 at 9:03 am

To Marcus Anonymous: It sounds like the gap between our positions has narrowed quite a bit. You don’t disagree with my view of speculation in the oil futures markets, but you make a sharp distinction vis-a-vis speculation in oil itself, and you believe the financial markets have little or no effect on the actual oil market. (In this sense the word "derivative" supports you: it is only derivative of the primary market.) I assume you agree that the spot price traded in Chicago, New York or London, is very closely tied to the price at which oil companies actually buy and sell oil? But you see all the causality going from the latter to the former, and none the opposite direction? As I have often admitted, one would expect under my "blame it on interest rates" theory, or under others’ "blame it on the speculators" theory, that oil inventories would be very high now, which they are not. But that is not precisely the issue that remains at stake between us. Let’s imagine a market in which storage capacity was completely inelastic. You would have to agree that even if the increase in demand came from speculation regarding future prices (and I don’t see that it matters whether it is financial speculators or commodity corporations that are leading the way here; there is only one spot price), that by definition it would not in this case show up as an observable increase in inventory holdings. Now of course inventory holdings do vary, within a certain range. But, as Larry Summers has recently pointed out, the argument that inventories are not especially high now "proves rather too much." Not only would it reject the speculation and interest rate explanations for high oil prices, but it would reject a majority of other theories as well: the "peak oil" hypothesis, expectations of continued rapid growth in China and India, risk-aversion associated with fears of military conflictin the Gulf or instability among other oil producers, and so on. Perhaps there is something wrong with the inventory data. (We should really be looking at worldwide inventories, not just domestic.) Or perhaps, as I have been arguing, the above-ground inventories are less important than the inventories under the ground, i.e., the decisions by Gulf producers how fast to pump oil is the factor that matters.What do you think?

Marcus AnonymousJuly 16th, 2008 at 1:10 am

JF — I think that you’re a real smart guy and a good economist. So I offer you the following logical proposition:“If speculation is the cause of high oil prices THEN Jeff Frankel can explain it coherently.”If that proposition is true, then so is its logical contrapositive:“Jeff Frankel cannot explain it coherently THEN speculation is not the cause of high oil prices.”So I think that you’ve tried really hard and that you are just twisting yourself into knots. So now you should just say, “Hey, I tried and it’s just impossible to get around the inventory issue!” Maybe you just wrote too quickly without proofreading the second paragraph, but your meaning is extremely unclear to me, beginning with the phrase “storage capacity was completely inelastic.” I’ll assume that means that the storage tanks are all topped off: You cannot store any more. In that case speculation — by the commonly accepted economic definition — is not possible. So speculation cannot increase the demand for petroleum. If you have a different meaning for speculation, then please use another term. Let’s not cleverly mix definitions – such proofs are sophistry. Let’s not prove that Abraham Lincoln died of malaria by defining malaria as “a gunshot to the back of the head.” No redefinitions. Let’s not call “OPEC cartel strategy” “speculation.” Nor do I include expressing a verbal hypothesis: “oil prices might go up” as speculation. speculation is not having an opinion. If nobody buys and stores the oil, it’s just an opinion that does not affect the price of oil. Such a pure opinion is no more relevant to the price of oil than the opinion "I’d rather date Pamela Anderson than Angelina Jolie" (we’re geeks and will never get the chance to date either). And let’s not apply the correct definition of speculation to the wrong market. Speculation in cherries does increase demand – for cherries, not oil. Speculation in oil futures does not directly increase the demand for oil. When the February futures contract expires it is recycled and comes back as the New York Post. Futures are not distilled into gasoline. Different markets, different excess demands.Write down a proper excess demand function. The only way futures or futures prices get in there is via demand for inventory. Formally:Excess Demand for Oil = (Consumption + Desired Inventory Build – Wellhead Production)The only term on the right hand side where futures prices can plausibly go is the inventory term. And let me cut you off before you try to switch definitions again – the Hotelling Principle may affect wellhead production (not that I’ve seen evidence of that operating today on oil production), but that is not the same thing as the strict definition of speculation.As for Larry Summers’ argument, it’s technically wrong to begin with. Look at the excess demand function above. It’s the change (build) in inventory that is the mark of speculation, not the level. And I certainly don’t buy that the inventory data disproves all the other hypotheses including peak oil. So it’s getting late and I must be fresh for tomorrow’s 10:35 inventory data release.Warm regards,Marcus Anonymous

Jeff FrankelJuly 20th, 2008 at 9:21 am

Marcus, As I understand the semantic issue, you believe that the definition of speculation –at least "speculation in oil" as opposed to "speculation in the oil futures markets" — requires an increase in holding of inventories. I do not agree that your definition is the normal one, but let’s leave the semantics behind. The economics is more important. Let’s consider a case where, yes indeed, the supply of oil storage capacity is inelastic because every tanker, tank, and tin can is already full. Then something happens to cause people to expect an increase in price in the future, for whatever reason (security fears in the Gulf, forecasts of growth in China, news of diminishing underground reserves, or even non-fundamentals based speculation — sorry, I mean a non-fundamentals based increase in demand stemming from expectations of a future increase in price). My claim, as a matter of economics, is that both the futures price and the spot price will rise to clear the market, whether one calls it speculation or not. You will then ask: "but higher price will reduce demand somewhat; how does the market equilibrate if there is no possibility of higher inventories?" I will reply: "in response to the expectations of higher prices in the future, oil producers will pump a bit less today, and leave more underground." As I keep saying, the oil underground is the real inventory. Finally, to try to achieve intellectual closure, let’s assume you then ask me, hypopthetically: "what if BOTH inventory capacity and pumping rates were completely inelastic." My response is, "yes, in that case expectations of future prices have little or no effect on today’s price, as is the case in the market for perishable commodities like milk. Storability is a necessary condition for intertermporal arbitrage, and therefore for speculation." Have we now achieved full convergence?JF

Marcus AnonymousJuly 20th, 2008 at 1:53 pm

JF,I agree now that you have a consistent economic theory that logically closes upon itself. Whether this is just an interesting theoretical possibility or an apt description of current developments in the oil market is another question. I will get to that later.But I do want to emphasize that your use of the language is extremely sloppy – and dangerous. In his “Politics and the English Language” Orwell lamented, “It (the English language) becomes ugly and inaccurate because out thoughts are foolish, but the slovenliness of our language makes it easier for us to have foolish thoughts.” If you have multiple definitions, use distinct words so that your meaning is unambiguous.In its current incarnation, your theory states that the actors causing current prices to increase are the producers who are reducing wellhead production. First, if they are right, then it is good that they are saving oil for future consumption. Second, futures trading has little or nothing to do with this theory. Shutting futures markets down would not have any effect on this process. (Note parenthetically that this theory of speculation could not explain crop price increases because farmers can’t “store” their land by not producing.) Yet you insist on using the same word “speculation” to describe and confuse two distinct theories: one in which prices rise because producers decide to conserve oil in the ground and a second where futures traders are bad actors. The “anti-speculation” policy remedies that your theory will be used to justify will attack the wrong type of speculation. So even if your analysis is correct, the remedies will be useless. Such policies will do no good, but will allow Congress to extend the “Law of Unintended Consequences” for another 10 years and maybe eliminate a few loopholes.Up on Capitol Hill, Sen. Cotton Mather’s “Bring Back Prosperity by Pitchforking Witches and Speculators Act” is working its way through Congress. I doubt that they will find many witches, but they could burn or fork few futures traders (I won’t personally miss them) and it will politicize the CFTC (I do care). By continuing to talk about “speculation”, you feed the process and demote yourself from “eminent professor of economics” to “pitchfork in the hands of terrified peasant.”As for your theory that speculation is “producers deciding to keep it in the ground” it’s sort of inconsistent for you criticize other for doing it (or claim it’s bad behavior) when you criticize “the Republican policy of ‘drain America first’” on another blog.As for your theory itself, let’s call it Hotelling (a verb) rather than “speculating” so that your theory has a name of its own. Hotelling is actually a bit peripheral to the current supply constraints. There are more important factors. 1) Dermot Gately’s work on OPEC’s weak production incentives is the best description of why OPEC would choose to keep oil in the ground. 2) Oil may not have peaked globally, but it has peaked in the OECD (be afraid, be very afraid). 3) And most important in restraining oil production in AD 2008 is the competition for economic rents among international oil companies, national oil companies, sovereigns (Putin, Chavez), corrupt government officials, oligarchs and crony capitalists, oil workers’ unions, rebels and bandits (MEND), and other miscellaneous opportunists. The higher prices go, the greater the incentive to fight over rents at the cost of both current and long-term output. That may be related to, but is not the same thing as, a conscious Hotelling decision to keep oil in the ground.Yours faithfully, but not actually,Marcus Anonymous

Marcus AnonymousJuly 20th, 2008 at 1:53 pm

JF,I agree now that you have a consistent economic theory that logically closes upon itself. Whether this is just an interesting theoretical possibility or an apt description of current developments in the oil market is another question. I will get to that later.But I do want to emphasize that your use of the language is extremely sloppy – and dangerous. In his “Politics and the English Language” Orwell lamented, “It (the English language) becomes ugly and inaccurate because out thoughts are foolish, but the slovenliness of our language makes it easier for us to have foolish thoughts.” If you have multiple definitions, use distinct words so that your meaning is unambiguous.In its current incarnation, your theory states that the actors causing current prices to increase are the producers who are reducing wellhead production. First, if they are right, then it is good that they are saving oil for future consumption. Second, futures trading has little or nothing to do with this theory. Shutting futures markets down would not have any effect on this process. (Note parenthetically that this theory of speculation could not explain crop price increases because farmers can’t “store” their land by not producing.) Yet you insist on using the same word “speculation” to describe and confuse two distinct theories: one in which prices rise because producers decide to conserve oil in the ground and a second where futures traders are bad actors. The “anti-speculation” policy remedies that your theory will be used to justify will attack the wrong type of speculation. So even if your analysis is correct, the remedies will be useless. Such policies will do no good, but will allow Congress to extend the “Law of Unintended Consequences” for another 10 years and maybe eliminate a few loopholes.Up on Capitol Hill, Sen. Cotton Mather’s “Bring Back Prosperity by Pitchforking Witches and Speculators Act” is working its way through Congress. I doubt that they will find many witches, but they could burn or fork few futures traders (I won’t personally miss them) and it will politicize the CFTC (I do care). By continuing to talk about “speculation”, you feed the process and demote yourself from “eminent professor of economics” to “pitchfork in the hands of terrified peasant.”As for your theory that speculation is “producers deciding to keep it in the ground” it’s sort of inconsistent for you criticize other for doing it (or claim it’s bad behavior) when you criticize “the Republican policy of ‘drain America first’” on another blog.As for your theory itself, let’s call it Hotelling (a verb) rather than “speculating” so that your theory has a name of its own. Hotelling is actually a bit peripheral to the current supply constraints. There are more important factors. 1) Dermot Gately’s work on OPEC’s weak production incentives is the best description of why OPEC would choose to keep oil in the ground. 2) Oil may not have peaked globally, but it has peaked in the OECD (be afraid, be very afraid). 3) And most important in restraining oil production in AD 2008 is the competition for economic rents among international oil companies, national oil companies, sovereigns (Putin, Chavez), corrupt government officials, oligarchs and crony capitalists, oil workers’ unions, rebels and bandits (MEND), and other miscellaneous opportunists. The higher prices go, the greater the incentive to fight over rents at the cost of both current and long-term output. That may be related to, but is not the same thing as, a conscious Hotelling decision to keep oil in the ground.Yours faithfully, but not actually,Marcus Anonymous

AndréJuly 23rd, 2008 at 2:16 pm

If your "blame the real interest rates" theory is correct, I think we should observe upward forward curves nowadays and the opposite in high real interest rates times. Is that what really comes to happen?

devenMarch 1st, 2012 at 2:24 pm

Dubai offers tax free investment opportunities. One may buy free hold property with clear title on his/her name and earn profits via price increase or rental income which is non taxable. Find out more about Dubai's tax free investment opportunities at http://www.indogulfproperties.com/blog/

Most Read | Featured | Popular

Blogger Spotlight

Emre Deliveli The Kapali Carsi

Emre Deliveli is a freelance consultant, part-time lecturer in economics and columnist. Previously, Emre worked as economist for Citi Istanbul, covering Turkey and the Balkans. He was previously Director of Economic Studies at the Economic Policy Research Foundation of Turkey in Ankara and has has also worked at the World Bank, OECD, McKinsey and the Central Bank of Turkey. Emre holds a B.A., summa cum laude, from Yale University and undertook his PhD studies at Harvard University, in Economics.

Economics Blog Aggregator

Our favorite economics blogs aggregated.