« What I've Been Reading | Main | Division of labor is limited by the extent of the market »
What will happen with commodity prices?
Megan McArdle gives one bottom line, referring to Paul Krugman's somewhat pessimistic column. I would say that China has been massively productive but not so much in producing commodities. That means the demand for commodities has gone up much more rapidly than the supply. You could imagine an alternative universe in which China grew by figuring out ways to produce oil, copper, and rice much more cheaply. Of course that's not what happened and it is relatively easy to see why not. Following some good policy changes, Chinese growth has been driven by a massive rural to urban migration and yes we are talking about hundreds of millions of people. It's plastic basketballs that have become cheaper, not the products of farms.
The mere addition of labor inputs to urban areas doesn't, in the short run, help you produce commodities more cheaply. Think of the Solow model where K and L have gone up lots but the rate of generating new ideas is only slightly higher.
When all those new Chinese engineers and scientists are at the peak of their creative powers, this relationship will reverse itself and commodities prices will plunge. But it's quicker to produce another toy than to bring about a new Green Revolution, so in the meantime commodity prices are very high. I give the current price trend another ten or fifteen years or so to run. Eventually high commodity prices will seem permanent and then the bottom will drop out.
We've never had a rapid and successful migration of hundreds of millions before, ever.
Posted by Tyler Cowen on April 22, 2008 at 06:11 AM in Economics | Permalink
Comments
We should also remember that the price elasticity of many commodities is fairly low on both the supply and demand sides. I won't predict when, but I will predict that when they do fall commodity prices will fall fairly far and fairly fast.
Posted by: MichaelB at Apr 22, 2008 7:45:46 AM
I think I've read that when metallurgy began, workable copper could be picked up off the ground on the island of Crete.
We tend to think that mines and production must have been similar in the past, and will be in the future.
The problem is that for many commodities we go for the easy stuff first.
The optimism in cheaper commodity prices is in thinking not just that we'll continue to find the stuff but that we'll never reach a limit in making the "harder" stuff "cheaper."
It is a battle between technology and resources - and at a minimum we should agree that the outcome cannot be preordained across the board. Some commodity prices will fare better than others.
Posted by: odograph at Apr 22, 2008 8:29:08 AM
Tyler you are spot on but one follow up question (you can consider this a late addition to yesterday's call for topics):
What will happen to the US$ along with this? Commodity prices will remain high on a global basis for the medium term (ok for the effectively long haul for those of your and my age). Prices in USD could change rather more to the extent the Yuan decouples.
Posted by: misplaced trust co. at Apr 22, 2008 8:57:44 AM
Looking at the Toyota Prius one could conclude that human effort is good substitute for petroleum. I believe that a hybrid car uses petroleum about 30 more efficiently than a regular gasoline car. That is a big leap and there is more to come.
Posted by: Floccina at Apr 22, 2008 9:19:02 AM
BTW IMO the rising commodity prices are mostly due to inflation of the money supply with China keep manufactured goods prices down. Now if the fed did not inflate we might have had more unemployment as China drove down anufactured goods prices. BTW Agriculture can increase output by increasin all sorts of inputs including human effort. I see no long term problem.
Posted by: Floccina at Apr 22, 2008 9:26:40 AM
Huh,
I could of sworn it was this blog that talked about most raw materials were under the contorl of third-world government-run companies that weren't reinvesting their profits into improving their production capabilites. Oil in the Middle East and Central and South America, for example.
Posted by: Xmas at Apr 22, 2008 9:43:39 AM
I think this run up will continue, but not for 10-15 years. I think it will run off the cliff in about 5-7 years.
Floccina, the hybrid doesn't substitute human effort, it simply uses a battery - which uses expensive chemicals. Unless we have advances that reduce the price of driving a mile, we don't have real progress. Although, the hybrid could represent real progress if the price of gas went up enough.
Posted by: SkepMod at Apr 22, 2008 10:09:05 AM
I'd bet a lot of Chinese manufacturing capacity is rather inefficient in terms of commodity inputs. These are labor intensive industries that came online during a period of low commodity prices. Given investment, there's probably a lot of room for more efficient production processes.
As for agriculture in China, the labor exodus is as at least as much a consequence of abysmal labor productivity in the countryside as it is demand in the cities. Rural China is still constitutionally collectivized, as far as I know, and that's hasn't been demonstrated a recipe for growth in production since the famines of the Great Leap Forward. See this February Economist article.
Posted by: Garrett Schmitt at Apr 22, 2008 10:11:36 AM
Furthermore, to add to Xmas's comment, food, especially rice, is absurdly regulated. On the one hand there's places like China and Ethiopia where agricultural land is leased from the State. Countries that have an advantage in rice production flirt with export tariffs and quotas (like Thailand). Countries that are less competitive have subsidies and import tariffs and quotas (like Japan).
Posted by: Garrett Schmitt at Apr 22, 2008 10:22:18 AM
FWIW, my price per mile in my Prius today is equal to my price per mile in my Camaro Z-28 in 1995. For me at least, the hybrid advantage has already been run.
Posted by: odograph at Apr 22, 2008 10:28:51 AM
Just remember whenever you hear a libertarian ideologue claim that commodity prices always go down over a reasonably medium time frame that our society is the product of a long experiment in survivorship bias (i.e. we're only here because all _our_ resource transitions worked out swell; the socities' whose didn't work out as well aren't here to serve as examples).
http://en.wikipedia.org/wiki/Survivorship_bias
Physics and geology trump economics any day of the week.
Posted by: M1EK at Apr 22, 2008 11:11:28 AM
"What Will Happen to Commodity Prices?"
A touch of fundamentalism is in order. Forget the trimmings for now.
Commodities have low short term price elasticities of supply, and of demand. The hundreds of millions getting richer is jacking up demand beyond longer term expectations.
Commodities tend to display reasonably high longer term price elasticities of both demand and supply. The longer term here has been measured in years - not months and not decades. To give the present price trend 10 to 15 further years is bucking history. To expect commodity prices to go over a cliff when they decline is in line with history - and more to the point, in line with current expectations of many larger commodity suppliers. A Predictify feature on when commodity prices will break should attract a wide, and attentive, audience.
Supply of some commodities - the agricultural ones - is much more affected by the weather. Current expectations of supply have the historical variation of the weather factored in. Unfortunately, the trend towards global warming carries with it a systematic increase in the variability of the weather. Nowadays that is just about taken into account in insurers' expectations of extreme weather; but not yet in lower expectations for average crop yields. The supply adjustment that will take agricultural commodity prices over their cliff is therefore likely to be longer delayed than history would lead us to expect. We may even have enough time to so improve information that the cliff turns into a gentler slope; though the liklihood of our collectively using that time for the purpose remains low.
Posted by: David Heigham at Apr 22, 2008 11:37:19 AM
Copper, at least, seems likely to fall in the near term. The housing construction slump in the US is likely to reduce US demand for copper by as much as 20%, while Chinese production is catching up to their demand so that their net imports are projected to drop year on year as well. All while global production is supposed to increase by about 5%...
It does look like oil and all of the commodities whose prices are influenced by it are going to be high for a while, though (and that includes agricultural products, uranium, maybe even aluminum...). It makes me wonder about the way that some governments with inefficient programs (Mexico, perhaps Saudi Arabia) have no great incentive to increase their production. On paper of course the Mexican government would make a lot more money if they could improve extraction from the Cantarell field - but with prices so high, they don't really *suffer* by not maximizing production, and if they believe prices will remain high then the unexploited oil still represents future wealth...
Posted by: bbartlog at Apr 22, 2008 11:50:04 AM
"On paper of course the Mexican government would make a lot more money if they could improve extraction from the Cantarell field - but with prices so high, they don't really *suffer* by not maximizing production, and if they believe prices will remain high then the unexploited oil still represents future wealth..."
Interesting. You could make a case for oil CEOs and stockholders making billions with this example. An oil company would try to get every drop it could right now to reap the profits where the state oil company is fine with the status quo.
And on a more general scale, the huge profits incentivize risk-taking for future production. Who knows how much more production Mexico would have with private companies owning the fields.
Posted by: Matt at Apr 22, 2008 1:14:50 PM
10-15 years Tyler? You must be counting on Bush's 2025 policy predictions.
David Heigham makes good points, by ag prices CAN fall of a cliff IF governments stop the distortions. Step 1, end ethanol programs. Step 2, approve Doha.
Posted by: David Zetland at Apr 22, 2008 1:39:17 PM
I wonder if at some point high extraction prices will lead to a prisoner's dilemma type scenario wherein oil producers withhold production in expectation of higher prices at some point in the future; the dilemma being that supply has to continue at a level high enough to keep the transitional costs to alternative sources less palatable but no producer wanting to act as sustainer of the habit for consumers, thus forgoing larger future profits.
Posted by: meter at Apr 22, 2008 2:09:05 PM
There has to be a more elegant way of articulating that...
Posted by: meter at Apr 22, 2008 2:10:38 PM
meter,
Elegance? I don't know. Von Neuman and John Nash had ways of describing such games with elegance; most of us are less gifted.
But for practical articluation of the issue, look to Saudi and Opec policy in recent years. The basis of their witholding supply is that they have the Russians et al. with their high discounts of the future (also known as greed) keeping prices below the technical transition thresholds the baseload producers fear.
Posted by: David Heigham at Apr 22, 2008 3:03:52 PM
The historical comparison would be the seventies and it took a decade. Since oil is largely in the hands of producers that see no reason to increase it or are incapable of doing so, it may well take longer this time. Some real breakthrough could shorten this, but since when have we seen one? If it were easy, it would already be done. Producers realize this and have no fear of being displaced for a long while.
Posted by: Lord at Apr 22, 2008 4:18:59 PM
The last time oil crashed it took BOTH a supplier who had ample excess capacity to spare (the Saudis) AND a practically-depressionesque downturn in some major growing oil-thirsty economies (Indonesia et al).
Anybody see a depression happening for China and India? Because no other supplier has the ability today to suddenly open up the taps a lot wider than they are now - and there's strong evidence of decreasing flow coming soon for many of them.
Posted by: M1EK at Apr 22, 2008 4:57:30 PM
Agricultural commodities are more highly regulated than almost any other commodity (with the exception of drugs which are illegal in the US :).
In addition to the US and EU agribusiness price supports, even developing countries have high barriers:
http://www.reuters.com/article/latestCrisis/idUSISL265987
"...wheat producer Pakistan and rice grower Vietnam, have banned exports..."
Philippines charges five over rice hoarding:
http://www.abc.net.au/ra/news/stories/200804/s2221217.htm?tab=latest
Traders won't buy more rice, fear hoarding raps:
http://tinyurl.com/5gzqp6
Posted by: Mr. Econotarian at Apr 22, 2008 5:04:40 PM
If you think Tyler is right, as I do, there are two good funds in which to invest. First, for a pure commodities play, there is the Pimco Commodity Real Return Fund. This tracks the Dow Jones - AIG Commodities Index -- a broad index not so heavily dominated by oil. Second, there is the Permanent Portfolio -- a very diversified fund that includes silver, goal and Swiss francs as well as stocks and T-bills. It was founded by Harry Browne and has an excellent long-term performance record.
Posted by: Mario Rizzo at Apr 22, 2008 6:06:52 PM
It is a gross violation of Occam's Razor to attribute the recent very broad-based run-up in dollar commodity prices primarily to the plethora of disparate causes to which they have been attributed: "peak oil", the war in Iraq, ethanol subsidies displacing food, and so on. Rises in industrial demand, increases in the costs of transporting commodities due to high oil prices, and so on explain only a small fraction of the rise in other commodity prices, and do not explain at all why precious metal prices have increased alongside those of other commodities. Occam's Razor points us, as it did to wise investors and economists in the 1970s, to the one kind of commodity all these other commodities have in common: the government currencies they are priced in.
In the 1980s, there were just as many of these disparate bullish factors for commodities as in the 1970s and as today. In the 1980s there was increasing industrial demand in Asia and the developing world, and the Iran/Iraq war drove insurance rates for oil tankers in the Persian Gulf into the stratosphere, yet commodity prices in dollars and dollar-linked currencies broadly fell, due to the emphasis of the Federal Reserve on fighting inflation throughout the 1980s.
Commodity prices now reflect more the value of commodities as stores of value and hedges or media of exchange, i.e. their values as money substitutes or hedges, than they reflect demand for their industrial consumption. The trend lines for global industrial demand have not really changed that much -- the increases in Asia are largely offset by slower demand growth in Europe and the U.S., and at today's high prices we will see industrial demand slowly fall in the U.S. and Europe and level off in Asia.
But demand for the oft-dreaded but ill-understood "hoarding" and "speculation", that is storing extra commodities (often off-the-books, or at least not in the officially measured warehouses) and the purchase of extra commodity futures and other commodity derivatives to hedge transactions based on government currencies, will remain strong as long as the Federal Reserve continues to inflate the dollar supply, and as long as many developing countries continue to link their currencies to this dollar. Commodity prices in dollars will level off, and then move back down close to historical trends based largely on just industrial consumption, if or when the Fed stops increasing the supply of dollars faster than the demand for dollars. If the Fed continues trying to inflate its way out of the latest bubble-following slump, commodities will not be the next bubble: instead they will form the basis of the new de facto currencies of high finance. See
The monetary value of liquid commodities
and
Commodity derivatives: the new currencies
Obviously the euro also plays a big role, but it is important to observe that the euro too has inflated, just to a lesser degree than the dollar. The euro has not developed the track record that would make many international investors think euro-denominated debt does not need to be hedged with commodity derivatives. Some wise investors still remember the hyper-inflations and outright defaults of the late 1920s and early 1930s that made many government bonds almost worthless, the destruction in the value of the U.S. debt in the 1970s from inflation, and many similar episodes. In these circumstances commodities are usually safer than government debt denominated in the fiat currencies of the same governments that owe that debt.
That's why the current flight to safety involves commodities as much or more than it involves government debt, and indeed the debts of less robust governments (ranging from municipalities in developed countries to the national debts of less developed countries) are no longer considered safe. It is well known by now that U.S. Treasuries are "safe deposit boxes" that pay negative real rates of interest -- i.e. one must pay for the privilege of storing one's wealth in a form backed by the vast future revenues of the IRS. They are a way losing one's wealth with less rapidity and volatility than with many alternative investments, not a way of actually building wealth. Commodities also store rather than build wealth. By contrast to Treasuries, commodities are more volatile, but are not subject to unknown future amounts of fiat currency inflation. A mix of U.S. Treasuries, European debt, and commodity derivatives now forms the ideal safe "store the cash under your bed" portfolio. Government debt alone is far too risky during periods when central banks are not strongly committed to reigning in money supplies.
Posted by: nick at Apr 22, 2008 7:05:19 PM
nick, again, in the 1980s the Saudis had capacity they could bring online at any moment (and eventually did, of course).
Occam's Razor is getting pretty close to suggesting that the Saudis are now essentially pumping as much as they possibly can. (How high does oil have to get for them to stop promising to pump more and actually do it, in other words? At what price level do we confirm it's an empty promise?)
Posted by: M1EK at Apr 23, 2008 8:41:11 AM
nick, again, in the 1980s the Saudis had capacity they could bring online at any moment (and eventually did, of course).
The same was true in the 1970s and is true today, and people in the oil business are well aware of this and it is properly reflected in the current price. Nothing has magically changed about the geology of oil in the ten years since oil was $20 a barrel. And not only Saudis, but Russia, Iraq, Venezuala, and Mexico are all pumping well below their potential (mostly for monetary reasons which I explain below), and will still have plenty of oil around decades into the future. And in that time most of the following -- Canadian tar sands, U.S. shale oil, deep sea oil, coal gasification, biofuels, nuclear, wind, solar, and conservation -- will make a big dent in the consumption demand for oil. As purely economics 101 matter of supply and demand for consumption, the value of oil for consumption will likely decrease in future decades. That of course doesn't say much about what the dollar price of oil will do, but I can say with confidence that if oil keeps rising it will be due to monetary factors, and in particular continued growth of the dollar and euro supply beyond the demand for dollars and euros. Oil will not rise due to traditionally recognized supply and consumption demand factors, in fact due to the current overinvestments in oil and alternative energies the percentage of the oil price attributable to such "fundamentals" will continue to decrease.
It's an incredible delusion that most of us are under, that the dollar is a stable standard of value and thus the dollar prices for commodities provide an accurate measure of supply and consumption demand for those commodities. They don't, and it's not even close. The supply and demand for the dollars is the far more dominant factor, and on top of that the demand for commodities as money rather than for consumption becomes, in times of inflation, much more important than "fundamentals." Those who just look at dollar prices have a wildly distorted view of economic reality.
Oil prices today reflect the best consensus estimate of what oil prices will be in the future. Regardless of how high oil prices get, they don't provide substantially more incentive to pump, because how high they are today doesn't say anything more than low prices about whether prices will be higher or lower five years from now. These commodities behave very differently from a widget where you either make it now or lose the sale. Furthermore, oil prices in terms of other commodities have not risen substantially, and the oil companies and countries understand that their increased wealth as measured in dollars or euros is partly an illusion (because its is the value of the dollar and to a lesser extent the euro that has dropped, not value for consumption of oil that has increased) and partly a matter of extra demand for commodities as money that would disappear if the Fed got its act together, as it did in the 1980s.
The monetary function of commodities can be seen from all the dreaded hoarding that is going on. With the failure of the dollar people need extra commodities in the "vault", even if these vaults are just the pantries of third-world families or the warehouses of governments.
Oil is mostly not hoarded directly above ground, because it is much more expensive to store above ground than to just keep it in the ground to pump later. When oil futures are used as money, this discourages pumping, despite the high oil prices, and that's exactly what we're seeing. If oil had no monetary function we'd see a huge increase in pumping, but instead we're seeing oil hoarding in the form keeping the oil in the ground and selling more oil futures backed by that unpumped oil. Practically all major oil countries are currently pumping below their potential output, and it is again a gross violation of Occam's Razor to explain this as a matter of a wide variety of political problems that just happen to all be occuring at the same time. These countries on average have no more political problems than they did during the 1980s and 1990s. They are under-pumping because the demand for oil as money, that is as backing for oil futures and other oil derivatives, dictates that more oil be stored relative to consumption, and that this oil be stored in the lowest cost way, i.e. that it be kept in the ground.
There are about $10 trillion worth of commodity derivatives outstanding, a vast increase over less than $2 trillion five years ago. It's much like fractional reserve banking -- commodity derivatives are a money substitute or hedge based on a much smaller value of commodities that have to be "stored in the vault" for use at "redemption windows", especially in cases of "runs on the bank." During such runs we will see big spikes in commodity prices completely unrelated to commodity fundamentals. Indeed, we've likely seen something like that during the recent and still ongoing credit crunch. (The comparison to fractional reserve banking is, of course, just analogical -- modern commodity derivatives markets are their own beast, and perhaps not well understood by anybody, but I think this analogy gives as good an insight as anybody has into what is going on).
Oil prices have also been driven up by the above-ground hoarding of more storable commodities, again because these commodities are increasingly being used as a substitute for the dollar. This increases the demand for such commodities, which increases the demand for oil used to produce the commodities. But this, like the direct excess demand for oil futures, goes away if or when the dollar resumes stability or another stable currency takes over from the dollar.
In short, we are increasingly seeing commodities used as a substitute for dollars, and to a lesser extent as a substitute for euros, by a very wide range of people from third-world families and governments to the most sophisticated international traders. The recent large run-up in commodity prices is explained by this factor. Commodity prices will continue to reflect a strong demand for their use as a monetary substitute until the time when monetary policy becomes far less inflationary than it is now.
Posted by: nick at Apr 23, 2008 6:13:15 PM
"The same was true in the 1970s and is true today"
No, it is most definitely not true today. The consensus even among the conservative end of the oil industry is that the Saudis can't increase production in the short-term. The fact that they've been saying they'll do so right away while oil's gone from 80 to 120 should tell you something.
Your theory does a very poor job of explaining the facts on the ground even compared to the admittedly imprecise Peak Oil version.
Posted by: M1EK at Apr 23, 2008 7:30:03 PM
The fact that they've been saying they'll do so [increase oil pumping] right away while oil's gone from 80 to 120 should tell you something.
It tells me just what I explained above: the Saudis, along with the other oil producing countries, make more money from directly or indirectly selling oil futures, backed by oil kept in the ground, than they do from pumping and selling oil today. These countries and oil companies are essentially printing money backed by their oil reserves. In the last five years demand for this kind of money, i.e. commodity derivatives, has grown radically, from less than $2 trillion outstanding to probably more than $10 trillion today. That is because it has during this time been a far more secure form of money than dollars, and has even been more secure than the euro.
As demand for this money substitute has increased, less of these monetary reserves are being let out of their vaults, i.e. oil producing countries are pumping less oil than oil prices suggest they should. If or when the monetary environment changes, i.e. if or when the Fed returns to noninflationary monetary policy, or a noninflationary euro is substituted for the dollar in far more places than today, and thus demand for oil futures as a monetary substitute decreases relative to demand for oil consumption, the oil countries will start pumping more oil.
As for their statements about increasing production, that is the politically correct thing for them to say, but means nothing. The Saudis can't predict or control the monetary policies of other countries, so they can't credibly predict what their optimal production strategy will be in the future, even assuming they have the incentive to be straightforward about their plans in the first place.
Posted by: nick at Apr 23, 2008 9:02:07 PM
nick, Occam's Razor suggests that the Saudis simply can't pump enough - they are now easily above the point at which their customers are realizing they need to invest in alternative fuels and alternative transportation, which will inevitably hurt the Saudis, no matter what they do with their oil.
As for them simply lying about increasing production - we both agree that they are; but the more obvious answer is that they can't, given all the other evidence that they are pumping as much as they can (declines in some of their fields; inability rather than unwillingness to deploy more drilling infrastructure; etc)
Again, the Peak Oil theory, with all its flaws, still does a better job of matching the observed facts than does yours.
Posted by: M1EK at Apr 24, 2008 10:18:14 AM
Occam's Razor suggests that the Saudis simply can't pump enough
How does Occam's Razor suggest anything about one data point? It's not a synonym for "what seems like common sense to me", you know.
given all the other evidence that they are pumping as much as they can
What evidence is this? I haven't seen any credible evidence of it. I've just seen such claims flying around the press and blogs; they may have all originated from the press releases of Aramco or associated companies. Even if this was true Saudi Arabia would be an outlier -- practically all the other major oil countries are producing below their capacity.
Posted by: nick at Apr 25, 2008 6:12:24 PM
If the current run up in prices of fuel and other commodities is fueled by speculation,then perhaps some of the margin players will bolt for the door causing a faster crash than would seem likely.We just witnessed this on a smaller scale with housing.
Posted by: mike butler at Apr 26, 2008 7:15:19 PM
If the current run up in prices of fuel and other commodities is fueled by speculation,then perhaps some of the margin players will bolt for the door causing a faster crash than would seem likely.We just witnessed this on a smaller scale with housing.
Posted by: mike butler at Apr 26, 2008 7:15:52 PM
nick, your claim that "practically all other oil producers are producing below capacity" is ALSO unsupported by the facts, and yet another outlier opinion.
In general, even most rational disbelievers in Peak Oil rest on the contention that there's big finds which will be producing Real Soon Now, NOT that there's a lot of capacity being held offline.
Posted by: M1EK at Apr 27, 2008 10:35:26 AM
