January 25, 2006
Soaring commodity prices
Is U.S. monetary policy behind the surge in commodity prices?
Impressive run-ups in the price of oil and natural gas over the last 5 years have kept energy markets in the headlines. But it is difficult not to see this also as related in part to a broader movement in commodity prices generally. As these graphs from Kitco Base Metals reveal, the dollar price of copper, aluminum, zinc, and nickel have all basically doubled over the last three years.
Jeffrey Frankel, Professor of Economics at Harvard University, argues that U.S. monetary policy may be part of the explanation. Low real interest rates, he notes, lower the cost of carrying physical inventories and increase the attractiveness of speculating in commodities relative to holding Treasury bills. He documents an impressive historical correlation as revealed by the scatter diagram at the left: those years in which the real interest rate was lowest tended also to be years in which commodity prices were high relative to other prices.
The graph below plots the difference between the 3-month Treasury bill rate and the change in the median CPI over the preceding year. It does indeed seem likely that the negative real rates of 2000-2003 contributed to the surging commodity prices of the last few years, just as they surely contributed to the high commodity prices of the 1970's.
One puzzling aspect of this theory, however, is the fact that the most dramatic surge in commodity prices seems to have materialized during the past few months, when real rates no longer appear to be low by historical standards. Frankel predicts that "As the Fed funds rate goes back up over the coming year, one can expect commodity prices eventually to come back down."
A bold prediction, but I am tempted also to call attention to the very dramatic emergence of the Asian economies, particularly China and India. At least as far as energy is concerned, the historical record suggests rapid growth in the use of raw materials in the early stages of economic development. Given the sheer magnitude of the economic growth in Asia, it is hard to dismiss this as a big factor in the price not just of oil, but perhaps commodities across the board.
Posted by James Hamilton at January 25, 2006 04:13 PMdigg this | reddit
Listed below are links to weblogs that reference Soaring commodity prices:
» Interest Rates and Commodity Prices from The Stalwart
It's been a while since we talked about oil, but with prices back up the debates must begin anew. The aforementioned James Hamilton links to some interesting stuff from Jeffrey Frankel, an economist at Harvard University. Frankel explores the link [Read More]
Tracked on January 26, 2006 06:32 AM
» Carnival of the Capitalists #121 from PHOSITA
Welcome to the 121st edition of Carnival of the Capitalists! We here are known as PHOSITA (pho – see – tah) : an arcane bit of patentese that refers to the mythical person of ordinary art. If an invention is... [Read More]
Tracked on January 30, 2006 06:42 AM
» This Week's Carnival Of The Capitalists from Financial Rounds
This Week's Carnival Of The Capitalists is up at Phosita. As always, there's a wide assortment of posts up in a number of categories. My picks of the week are: [Read More]
Tracked on January 30, 2006 10:11 AM
You pointed out in a post a few days ago that it shouldn't matter to market participants what currency the commodity is priced in (your post about oil and dollars). Couldn't low real rates in *any* currency lower the carrying cost for anyone willing to take the exchange rate risk? Maybe the relevant low rates aren't ours, but some other major currency.
Posted by: sestina at January 25, 2006 04:55 PM
Would it be too much to ask for?:
GDP per Capita vs Growth in Petroleum Demand per Capita
Posted by: Robert Cote at January 25, 2006 04:58 PM
There was an article on the Scientific American web site last week saying the world is running out of copper:
I don't know about zinc and such, though.
It's strange how we have to try to puzzle out what the markets are up to. Why can't we just call up the traders and ask them? Or just read their blogs? Maybe in a few years we'll have a lot more transparency in these matters.
Posted by: Hal at January 25, 2006 05:59 PM
Good post, Jim.
Is it not beneficial to also display charts reflecting actual demand growth for each of the base metals during the same period? What about projected demand? Same issue for inventory build (balance) history.
I haven't put a lot of time into researching projected demand growth, but I will take a shot at this.
Quote: Frankel predicts that "As the Fed funds rate goes back up over the coming year, one can expect commodity prices eventually to come back down."
I expect that interest rates will have to rise much further, say 3-5 points, to force a sustained and significant retreat on base metals prices.
Some industries requiring such base metals are involved in the production of durable goods which, in part, are presently running behind demand requirements. Hydrid vehicle drivetrain components, for example. And this type of demand will not shrink considerably, even if interest rates rise another few points.
Commodity prices levels last Fall served notice that conventional thinking about interest rate increases may not come into play on some commodities unless interest rates are pushed up considerably higher.
I recommend studying the demand data.
Posted by: Movie Guy at January 25, 2006 10:22 PM
Hhhm. Not wholly convinced. If low interest rates meant that carrying costs were reduced (sorry, of course they are, but if it were important that this were so), then one would expect to see larger stocks being carried. Is this in fact so? As far as I recall, LME copper stocks are extremely low by historical standards.
Posted by: Tim Worstall at January 26, 2006 03:45 AM
A bold prediction, but I think that Science need bold predictions, i. e., falsiable predictions.
And the chinese afirm and show data that they aren't causing the oil prices spike....
Sorry my bad english, my native language is portuguese.
Posted by: João Carlos at January 26, 2006 05:16 AM
A different approach is that in the 1990s when we were overinvesting in technology we were under investing in basics -- on a world wide basis -- so now we need higher prices in commodities, including oil to balance suppply and demand.
Historically commodity prices have always been a good leading indicator of oil prices. For example, over the last 20 years the correlation between copper prices this month and oil prices next month is 0.93 -- its 0.97 for the same month. Over 50 years the correlation is not that strong, but it is still very good.
the CRB industrial material price index is always a good indicator of world economic growth
and it is just saying the world economy is unusually strong. consequently demand for commodities and oil is outpacing supply.
Of course you can argue that the strong world economy is just a consequence of easy money and low real rates.
Posted by: spencer at January 26, 2006 06:39 AM
Let me preface my comment by noting I'm not an economist, but an engineer who is tired of the cheerleading by the investing establishment, and hence trying to get more educated about macroeconomics. I'm a novice here, so sorry if my question sounds outright silly.
OK, with that out of the way, I'm a little confused with the argument. It goes: the reason commodity prices have historically been high is because real rates have been low. But real rates are simply nominal rates minus inflation, which in turn is affected by commodity prices. (The y-axis is real commodity price index, but I assume the commodity inflation would be higher than a broad-based inflation measure) Chicken and egg, you say?
To argue the point that low real rates cause commodity price surges, shouldn't you look at rates and the consequent forward price changes(rather than for the trailing period)?
I guess my own amateurish interpretation would have been that high commodity prices are sustained (rather than caused) by a failure to aggressively raise rates at faster than inflation. Would you agree? (The consequences of such a high interest rate are besides the point here)
Posted by: Karteek at January 26, 2006 08:59 AM
It's also important to remember commercial metals production is very energy intensive. This has two effects. First, of course, the energy cost component of production has been approximately 15% (copper) to 40% (aluminum). Clearly in a demand driven market those costs are being passed on so it isn't really the commodity but the energy that is driving some of the price increases. Delivery costs are likewise higher. Second and less obvious, is marginal production capacity. The high cost of electricty in the Pacific Northwest has kept aluminum production off line constraining supply.
Rather than assert energy and base metals are both subject to commodity price inflation it is likely that the base metals are higher in some part due to higher energy prices. Hopefully a real economist can do the hard math as I've obviously strained my amateur reasoning skills to the breaking point.
Posted by: Robert Cote at January 26, 2006 11:53 AM
Does there need to be direct connection between overly stimulative monetary policy and rising commodity prices?
First, lax monetary policy encourages overinvestment and malinvestment typically by corporations, but over the past couple years by home renters masquerading as home owners as well. In the case of corporations, low interest rates permit expansion of manufacturing capacity. All this extra manufacturing capacity consumes more input materials, namely commodities. As this leveraged demand increases, then commodity prices increase.
Second, lax monetary policy in the U.S. forces China to increase its money supply accordingly. That increased money supply is used to finance manufacturing capacity. As above, the extra manufacturing capacity consumes more commodities, and prices therefor increase. (I use China as an example. But, clearly, there is no shortage of offshore manufacturing platforms servicing the American consumer -- was that too cynical?).
In view of the above, I wonder whether comparing current Federal Funds rates with current commodity prices is even helpul because corporations would have financed additional manufacturing capacity with loans made a year+ ago.
Posted by: Suresh at January 26, 2006 12:05 PM
I used the words "asymmetric inflation" a while back, to try to capture my end-user's (and not economist's) view of housing and fuel price increases. I don't even know if you call it "inflation" when it is driven by things like expanding asian demand. But on a gut level it kind of walks and quacks like inflation.
Posted by: odograph at January 26, 2006 02:18 PM
One of the more useful things to do would be to talk to people that actually produce copper. Former and current managers at Phelps Dodge (second largest copper producer in the world) believe that it has been increased demand from China and India that has driven up the price. The nominal price will remain high as long as those countries continue to develop and invest in infrastructure, and worldwide production remains at current levels.
Prices will fall as new production come online. We've got a couple of new mines opening here in Arizona in the next couple of years. Several more are being developed around the world.
On the cost side, new heap leach technologies will drive production costs down and turn waste rock into ore.
Also note that the current price is high only in nominal terms. For some perspective, have a look at http://minerals.usgs.gov/minerals/pubs/ds05-140/copper.pdf
Posted by: Scott at January 26, 2006 02:53 PM
I see that Suresh at IncomeTrap has a link to "Raw sugar price is at a 25-year high." On the one hand the article says it's due to sugar being redirected to ethanol production for transportation, but on the other hand they say "U.S. foodmakers that have had to swallow higher costs for everything from energy to vanilla in the past two years."
How widespread are these price increases? Please help me understand the difference between "a commodity bull market" and "inflation."
Posted by: odograph at January 26, 2006 04:47 PM
I think the lax monetary policy story has some merit. In the textbook version lax monetary policy raises aggregate demand, and causes prises to rise. This time there is a huge supply increase of cheap unskilled labor from asia, and this prevents price increases for labor incentive consumer goods. In a sense the agregate supply curve also shifted out. For commodities on the other hand the short run supply curve is almost vertical,because of capacity constrains, as a result, prices go up. Prediction: people will invest in new capacity all around the place. Yes it is true, that a lot of the commidity demand ended up in asia, so in a sense the commodity boom was created by asia. But the big question is how much of the asian boom was generated by US consumption and expectations of future consumption(exports and investment to export oriented manufacturing). In China's case probably a lot. And the US consumption story is very dependent on low interest rates
Posted by: Laszlo Varro at January 27, 2006 12:41 AM
This seems to me to be a classic case of confusion between correlation and causation.
In China, about 20 million people per year are moving from rural areas to urban ones. Imagine buidling urban infrastructure for 20 million people every year - say 10 Clevelands or Detroits. The consumption of cement for concrete, copper for pipes and wires, stainless steel for appliances, steel for rebar, etc. is staggering. The commodity price boom is a pretty straightforward supply squeeze in the face of rising demand. You cannot turn on the supply of nickel or copper overnight. Mine development is time and capital intensive. Look at Caterpillar's results yesterday - the demand for heavy equipment is directly related to increased mineral exploration and exploitation. The supply will come, probably in 2007 for copper and nickel, earlier for steel and zinc. Coal has probably already peaked out.
Sometimes the easy obvious explanation is also the right one.
Posted by: David Baskin at January 27, 2006 08:10 AM
Kitco, eh? I love that site, especially the contributions of the Mogambo Guru Richard Doughty, "The Angriest Man in Economics."
Posted by: Emmanuel at January 27, 2006 09:28 AM
In response to Odograph’s question regarding the article on sugar prices over at IncomeTrap, it is my opinion that we are seeing the effects of loose monetary and fiscal policy reflected in prices of energy and sugar.
Nobel Prize Laureate and brilliant economist Milton Friedman uttered the often-quoted lines, “Inflation is always and everywhere a monetary phenomenon.” Dr. Friedman meant that inflation is always an increase in the money supply—and deflation is always a decrease in the money supply. Indeed, given the ever expanding goods and services of the global economy, if money supply did not increase, prices would fall. This would happen because the ratio of consumables to units of currency would increase over time, making the currency relatively more valuable. What we’re seeing in prices of raw goods, finished goods, and services is price increases due to an increase in money and credit, or what pundits these days call “excess liquidity.”
“Commodity bull market” is the term used to describe constantly rising prices of commodities—but is really just the flip side of falling currency values. Currency values are falling because of “excess liquidity” or inflation. Lazzlo Varro pointed out that low labor costs of Asia have helped keep the price of finished goods down, so that we Americans don’t experience the effects of inflation at Wal-Mart as much as we do at the gas pump. David Baskin makes a good point: “What about the huge demand for concrete (China consumes 70% of the world’s supply) [and other raw materials?]” If currency supplies were fixed, commodities would follow their classic boom & bust cycles of falling supply, rising prices, increased production, supply glut, price bust, and falling supply again. What is different is an input to all commodities is energy—and input prices are rising. Commodities prices are, in my opinion, on a journey to a new, higher level.
Which brings us back to the cost of sugar. The US dollar has the enviable position of world reserve currency—and oil is sold only for dollars on the global market. When Asia buys oil, they do so with dollars. The rapidly rising price of oil has relieved Asia of many of their excess dollars, and helped support the dollar in the process. Without high oil prices, the dollar would have fallen, in my opinion, farther and faster than it has. That same high oil price, however, has made economical other sources of energy earlier thought too expensive—including biofuels from sugar.
Posted by: Dave at January 27, 2006 09:55 AM
Thank you very much. I've read all of that once, some of it twice, and will give it a third pass as I think about it some more ;-)
Posted by: odograph at January 27, 2006 11:28 AM
Hmm, the varience is pretty wide and the data set is small, so I am still with Hamilton - as I have posted previously I am of the opinion the current increase is due to the BRICS demand shock.
In principle I suppose Frankel's argument has some merit - the opportunity cost of holding inventory is certain lower during periods of low real interest rate, so that should have some effect on commodity prices by a simple increase in demand. But I would suppose this would be a one-off effect and probably small. It is also a very US centric view - why should real interest rates in one part of the world set prices globally?
Posted by: ChrisA at January 28, 2006 05:03 PM
It seems to me that the commodity bubble has many dimensions, many of which have been addressed here. Most of these arguments assume that the markets are always adjusting toward a stable equilibrium based on supply and demand.
However, like the NASDAQ bubble of 1999-2000, the current commodity market is mired in the irrational world of human greed and fear, market manipulators, hoarders, mercantilist self interest, and other non-textbook, real world agents.
Bubbles grow from a base of sound, equilibrium principles. Demand is stimulated by monetary and fiscal stimulus squeezing supply and raising prices which in turn stimulates an increase in supply which reduces price, etc. As M Freedman points out, in a world of excess money supply, the net result is not a net balance of supply and demand back at the equilibrium price but a permanent increase in price at the market clearing equilibrium, ie inflation.
At some point, in all this equilibrium push-pull, the markets tend to leave the rational world. Beanie babies, cabage patch dolls, NASDAQ, tulip mania, Tokyo real estate, railroads and canals, etc. The end result is a crash.
Commodities are not nearly in deficit as one would assume from the pricing structure. Look out below!
Posted by: STARK at January 29, 2006 09:24 AM
STARK, I don't know where you get the idea that such factors as "human greed and fear, market manipulators, hoarders, mercantilist self interest," etc. are not part of the textbook description of markets. My reading of textbooks is that it is exactly this kind of self-interested, profit-driven motivation which is assumed to drive market behavior. Markets in fact work best when people compete most vigorously to make as much money as possible. This causes the price discovery process to work at maximum efficiency and provides the greatest information and insight into the likely course of future events.
Posted by: Hal at January 29, 2006 10:17 AM
You make some very good points, most of which I agree with. My reference was to a competitive market structure in which there are no dominant buyers or sellers and each participant acts rationally. Some of the agent types I cited always exist but do not have excessive market power in a competitive market at equilibrium.
The concept of equilibrium itself is a useful educational construct that works quite well in most real world situations. I have interpreted the comments that are being made to "assume" that an underlying competitive market is in place and the supply and demand functions are being defined by multitudes of "rational" buyers and sellers with no particular individual market power, who make "rational" decisions on bid/ask pricing. That is not necessarily the case. OPEC for instance has a degree of monoploy power. Sumitomo in the 1990's cornered the market by "buying up" (hoarding) copper in hidden warehouses and ended up with a virtual corner for over six years, maintaining the price well above the competive market clearing price.This is an example of people competing most vigorously to make as much money as possible that resulted in a disfunctional market.
The point I really was making is that a bubble by definition is an irrational market. The concept of equilibrium is undefined. The character types I cited may actually gain a certain degree of market power. The pricing structure of the underlying asset is determined by "irrational" fear and greed, etc. Bubbles, by definition, involve a degree of mispricing that is not compatible with market equilibrium and other rational concepts. Some analysts, including myself, believe that the metals market is currently in a "bubble" environment. No amount of production/consumption analysis can describe the near hysterical amount of money flowing into the metals commodity markets.
Thanks for the comments.
Posted by: STARK at January 29, 2006 04:24 PM
The level and direction of commodity prices, more than any other good or service, is determined by the supply and demand for that particular commodity. To understand commodity price behavior we must remember:
These are global goods - so the global supply and demand determine the international price.
The local price relative to the global price will move in the opposite direction of the value of the local currency. The loss of 25% of the value of the dollar since 2001 is an enormous part of the rise in the dollar price of commodities (not to mention the main factor in the dollar increase in imports.)
Business relationships cause lags in price adjustments. Consumers and producers of commodities often operate with multi year contracts. This is especially true with metals.
Impact of Fed policy:
Anyone who suspects that increasing inventories lead to rising prices has never been out the ivory tower. Even if low rates resulted in rising inventories always result in falling prices.
However, the Fed certainly played a role. The question is what portion of the drop in the value of the dollar can be attributed to the Fed's easy money policy, versus tax cuts ...
Posted by: Rich Truxel at February 19, 2006 06:21 PM
When I was a real idiot I took a course in college on China; just your average intro-type class. This was 1985. At that time, I recall our professor explaining to us that because of China's then "wildly expanding" economy, the prices for all commodities would be "greatly affected". Well, about one year after that class, I recall a barrel of oil fetching about 8 dollars. Hmmm.
Posted by: MIke The Ignorant at April 17, 2006 11:52 AM
I am looking for a source of electrical generating coal price from 1986 to present. Novice at research. Am trying to make case that my Lodge is discounting dues from 1986. Dues doubled. Utilities quadroupled. Any help for a chart or graph that can be downloaded? Sorry to intrude on learned discourse.
Posted by: Madbob at July 6, 2006 12:28 PM
Madbob, DOE has average electricity prices per state going back to 1990.
Posted by: JDH at July 6, 2006 02:59 PM
STARK - Yep, you are dead on correct.
More than anything, the multi-year run up in commodity prices has had to do with easy money + more than a ten-fold increase in quantity of such directed towards commodity price speculation, and with these an expanded divorce between real economy fundamentals and commodities' prices, most evident in the oil market.
Which is not to say that there's been no demand increase since clearly that's the norm during any recovery. If the China story was correct we should have seen prices pull back last year, as both import and total consumption of crude oil came in lower than 2004. More, there's been a decent build in total OECD stocks.
At least as early as 1998, Matt Simmons made the case that price of WTI had ceased to be dependent on fundamentals but, rather, become a function of speculative trade. Others have noted same but this tends to be dismissed in the struggle to perpetuate what's been a quite profitable operation.
I'm afraid that too many neoclassical, i.e. mainstream, economists have fully accepted the questionable assumptions of that 'science', which in the last instance sees them fight to force reality to fit theory. So we 'end up' with bubbles that, by definition, are not bubbles until, of course, they deflate as they always do.
Posted by: juan at July 31, 2006 11:01 PM