Michael Masters' testimony regarding the role of speculation in commodity prices has drawn a lot of comment since last week. [See, for example, Cassandra, James Hamilton, Tim Iacono, John Mauldin, Michael Shedlock, Yves Smith.] According to Masters', portfolio investors' increasing participation in commodities via futures markets has been driving a speculative price boom, over a period of years.

I have to say, I am very skeptical of Masters' view. Perhaps I have drunk too deep of the Kool-Aid of orthodox finance, but, as the saying goes, "for every long there's a short", and Masters does very little to explain who is taking the other side of what he presents as a one-way bet, a virtual cornering of the commodity markets. We'll come back to this, because the shorts are the most interesting characters in our story. But before we go much further, we might as well opine a bit on the debate du jour, is "speculation" driving commodity (and especially oil) prices?

This question annoys me, because people rarely define what they mean by "speculation". Are you concerned about...

  1. Traditional speculators, making active predictions about future supply and demand, and determining that commodity are underpriced relative to other goods and services.

  2. Nervous hedgers, who respond to recent price volatility by taking larger-than-usual precautionary positions in order to manage operational risk.

  3. Portfolio diversifiers, who allocate some fraction of their portfolios to commodities in a price-insensitive way, as it becomes ever more convenient to do so, and the investment profession comes to view commodities as an attractively uncorrelated "asset class".

  4. Momentum investors, chasing recent price rises into a classic speculative bubble.

  5. Inflation hedgers and monetary skeptics, who view the purchasing power of financial assets as increasingly volatile, or who expect a decline in the purchasing power of financial assets, but who do not view commodities as undervalued relative to other goods and services.

  6. Corporatist governments, who seek to shed market risk by obtaining non-market access to commodities (vertical integration), or whose policies amount to speculation on future market conditions. Examples include countries that restrict food or commodity exports in response to high prices; China, whose state-affiliated firms purchase stakes in suppliers of essential commodities; Saudi Arabia whose purchase of GE Plastics looks to capitalize on preferred access to petrochemicals; oil producers generally, when they produce below capacity; and the United States with its strategic petroleum reserve. All of these practices have the potential to reduce supply to unaffiliated commodity users who rely on public markets.

It takes all kinds to make a market, and I think that we've got the whole menagerie. Also, we shouldn't forget this story, from Jeff Matthews (ht WSJ):

...the fact that a) world oil demand is up 12 million barrels a day since 2000, and non-OPEC oil supply is up only 4 million barrels a day since 2000, and b) America decided to convert food into ethanol at the very moment that c) China's demand exploded.

See James Hamilton for a fuller exposition of the case that oil price fundamentals are driving prices.

Masters fingers as the villain "index speculators", a Frankenstein combination of Types 3 and 4 above. There outta be a law agin' them, he suggests to Congress. Pension funds should be barred from commodity investing, loopholes that have undermined speculator position limits should be closed, and the increasingly meaningless distinction between commercial and noncommercial traders should be resurrected in CFTC reports. Okay.

But what if the price-setting speculators are not momentum-driven index funds, but "traditional speculators", correctly predicting that prices are below long-term fundamentals? Then limiting commodity speculation would prolong the mispricing, and cause us to waste resources that are kept artificially cheap. Alternatively, what if (as I suggested in the previous post) commodity prices are being driven by monetary fears? Then banning pension funds from commodities would amount to barring the exits, forcing workers to watch helplessly as their retirements are devalued away. If "fundamentals" are driving prices, or a flight by official actors from market to non-market means of resource allocation, limiting speculation would do no good, but would obscure the news by interfering with price transparency. The only circumstance under which limiting "speculation" might be a good idea is if the dominant tale is a momentum-driven speculative bubble. Which could, of course, be the case. Or not.

Which brings us back to the shorts. "Irrational exuberance" isn't enough to cause a speculative bubble. There needs to be something else that discourages rational traders from taking irrational traders' money when they buy overpriced assets, "limits to arbitrage" in the lingo.

Now, this is an old conversation in academic finance, especially with respect to the stock market. Heck, go chat with Brad DeLong and Robert Waldmann about noise traders, they're right here in the blogosphere. We'll dispense with the details here, and recite the pithy Keynes quote...

The market can stay irrational longer than you can stay solvent.

If a stock is overvalued, to correct the mispricing, you must sell it short. Even if you are right that it is overpriced, if the speculative mania continues, red ink on your short position might drive you out of the market and into poverty long before your foresight is vindicated. On the stock market, unleveraged "longs" can safely buy and hold, but "shorts" are forever at the mercy of the lunatics, hoping and praying that starry-eyed optimists don't go even more batshit insane. Sane people sit on the sidelines, allowing enthusiasm to run unchecked, for a while.

But there's a problem with applying this story to commodities. At least in theory, shorts in commodity futures needn't face the same risks as stock short-sellers. Commodity futures are time-bound and perfectly hedgeable. If you are a commodity producer, and know that futures prices are way too high, you can sell your own product forward into the market. If prices move irrationally against you, your only cost is the foregone opportunity of a speculative gain. If cash prices are out of sync with inflated futures markets, then anyone (in theory) should be able to get into the act, purchasing physical commodities and storing them for future delivery, thereby locking in a certain gain, a perfect arbitrage. If you think that the commodity boom is a speculative bubble, then you have to explain not only who is buying, but why all that speculative interest doesn't attract knowledgeable sellers who hold the price to "fundamentals".

A while back, Yves Smith pointed out the possibility that...

the volume of futures contracts is so large relative to the actual deliverable commodity that arbitrage (via taking physical delivery) won't force convergence of futures prices to cash prices at contract maturity.

In other words, in this messy real world, speculative interest could overwhelm the arbitrage mechanism designed to tether futures prices to fundamentals, for a while. But that begs another question. If you buy Masters' story, then we are in the midst of a speculative bubble that has been building over a period of years, not a sudden spike. So why haven't arbitrageurs increased their capacity to store and deliver goods, as speculative demand has slowly ramped up? The opportunity to profit is tremendous, especially if there are hordes of paper speculators who have no choice but to liquidate or roll their positions every few months. People with access to the physical commodity could profit from more than the ordinary arbitrage. At every roll, they have the entire community of "index speculators" over a barrel. Shorts are under no obligation to let speculators close out their positions at inflated "market prices", or even estimated "fundamental values". They can force longs to accept prices that overshoot downward, exacting a price for release from obligations that paper speculators are incapable of fulfilling, the obligation to accept delivery. If you think Masters is right, you have to explain why, year after year, those taking the short side have been willing close their positions at a loss rather than forcing more deliveries. Why haven't shorts entered the market who are capable of calling index speculators' bluff?

Hmm. Let's turn once again to Smith:

Remember, you can arbitrage futures to physical only if you are permitted to do so (only certain traders, known to have access to the storage and transport, are allowed to take or make physical delivery) and can actually obtain the relevant commodity.

So, there are potentially barriers to entry for bluff-callers. Who are these "certain traders" permitted to make delivery? I don't know, but one would imagine that commodity producers would be prominent among them. So, for the conspiracy-minded among you, here's a theory: Producers' core asset is not the stock of goods they have for sale today, but their potential to produce and sell a stream of commodity out into the indefinite future. It might be worth it for producers to bear an opportunity cost by not exploiting futures trades aggressively — that is by letting specs close positions at artificially bid-up prices — in order to inflate the apparent value of their enterprises, especially when producers intend to borrow funds, sell equity, or make stock-based acquisitions. Managers whose compensation is equity-linked might be particularly enthusiastic. Depending on how numerous and competitive the community of enterprises capable of physical delivery on prominent contracts, there might be a tacit cartel on the producer side, accommodating speculative futures prices, while managing spot supply so that cash market prices (which are less consistent and transparent than futures prices) are not outrageously out of line with futures market benchmarks.

Is this really going on? I have no idea. As I said initially, I can see all kinds of reasons why commodity prices might be rising, besides "irrational speculative bubble". But I do know this. If it is the "index speculators", if it is a speculative bubble, then those who blame workaday money managers asset-allocating into commodities are buying the con and blaming the patsies. If there's a speculative bubble, the mystery — and the target of any reasonable policy interventions — lies on the short side. Sooner or later, the lemmings going long will take care of themselves.

Update History:
  • 29-May-2008, 3:30 p.m. EDT: Eliminated a superfluous "so" (only one of many).
Steve Randy Waldman — Wednesday May 28, 2008 at 7:01pm permalink
Nemo (mail) (www):
If a large amount of money is always long the longer-term futures contract, doesn't that create an incentive for the producers to curtail production, thus causing the "artificially high" futures price to become a self-fulfilling prophecy in the spot price?

I mean, suppose the Big Money is always long, as Masters claims. And suppose that Money is Big enough to overwhelm the "smart" money that actually understands the underlying supply and demand and tries to short. Then it is only the producers who really have the ability to take them to the cleaners.

True, the producers would make a killing every time they sold at the spot price. But as long as the Big Money rolled into the long-term futures contracts and drove them up, the producers would have no incentive to drive the spot price down very far. Their incentive, in fact, would be to drive it no lower than the discounted present value of the (always inflated) futures price. As long as such a bubble persisted, the long-only funds would only lose money at the prevailing discount rate, I think... Which is pretty light punishment for participating in a speculative bubble.

It seems to me that this could go on for quite a long time. But perhaps I am missing something.
5.28.2008 8:33pm
Melancholy Korean (www):
Ah, yes. M. Masters. One of the biggest tools in the market. Specialist in picking off trading desks. The question is what's the angle? How does he benefit from giving this testimony, because it's not from some noble desire to help people.

Let's see. I can think of the obvious publicity for himself and his fund and the goodwill such testimony generates in Congress, especially among Democrats, as we will most likely get a Democratic Administration next January.

Oil is prob a speculative bubble, but the discussions about what determines "speculation" and all the studies and recent blog posts and comments (like mine now) are beside the point. They are academic (academic means, of course, useless). What's the trade here? That's the bottom line. I would bet we go higher before the awesome, cliff-diving crash in prices, as demand contracts from the coming worldwide recession.
5.28.2008 8:35pm
Anon:
I think the dynamic that your analysis is missing is that there's just plain too much liquidity out there. Too much investable money looking for a home is raising the natural level of asset prices. I wouldn't be surprised if this crisis drags on for years -- and we never have a 20% stock market decline, the new normal for oil becomes something well north of $100, etc.

Where are the shorts? Trying to figure out how to factor asset price inflation into their analysis. The speculative aspect is the over shooting that goes on as the shorts move cautiously before they feel confident about where the new fundamental level lies.
5.29.2008 1:03am
Hubert:
I think the main part of the story is relatively simple.
Rising prices bring in trend followers and index speculators. On the short side, look f.e. on oil: Which Gulf or Canadian producer with high cost would not like to lock in at least a big part of 2009 or 2010 production ? One could look at the Q´s and K´s of the bigger producers to find out. The seven sisters are more complicated because higher prices bring their reserves down via sharing agreements. As long as these two forces are more or less in balance they might even out.
But they won´t stay in balance.
If index speculators panic first then producers can always lock in trading profits by buying their hedges back under cash prices (exhibit A: NatGas squeeze in Autumn 2006). But if Producers stop selling forward to the extent that index money or trend folowing money wants to be long, then what? Are we there already or is this only the start?

Also totally missing is the political dimension of this story. If the Obama/McCain might stop deploying these annoying missiles in Putins backyard he might be willing to cut a deal for a reasonable oil price range. Reasonable is anybody´s guess for sure.
5.29.2008 2:08am
Alessandro (www):
As far as oil is concerned I don't see how people expect GCC countries not to cut supply.

Right now they have a huge trade surplus that end up tied in US dollar denominated debt. So they are depleting their reserve of real, non-renewable, universally usefull resource exchanging that for monetary reserves hold in doubtful debt denominated in a devaluating currency.

If I had oil reserves and I felt the need to keep reserves, I'd keep oil. Why pump it out at all once you can't exchange it for anything useful.
5.29.2008 6:45am
fernando (mail):
Rick Santelli demolished masters argument yesterday, what he says makes perfect sense. on the expiration day of the contract you have the PURE market price of the commodity because only the real economic players are left, there is no speculative bid
http://www.cnbc.com/id/15840232?video=753754816&play=1
5.29.2008 6:49am
Independent Accountant (mail) (www):
I read Masters article and thought it preposterous.
5.29.2008 7:28am
Benign Brodwicz (mail):
Bubble bubble toil and trouble.... liquidity (debt) everywhere (in the US, anyway) and not a stable numeraire (commodity) anywhere... we are thrown back on diversification among liquid asset classes to ride this monetary plasma into the foreseeable future... trying feverishly to skirt the random debt-deflation (e.g., US real estate).... hoo! One only hopes that the central bankers of the world, having reflated pretty much in unison, don't all tighten up at the same time....
5.29.2008 10:50am
David Merkel (mail) (www):
I like your blog. I came to the same conclusions in a less eloquent and less rigorous manner, as part of a more general discussion of bubbles.

Toiling Over Bubble Troubles, and
Blowing the Bubble Bigger

For what it is worth. Keep up the good work.

David
5.29.2008 12:18pm
gaius marius:
the truth is that commodities DO boom and bust -- they always have. it may well be important to argue about the mechanism, as you here do and cogently, but not as important as simply realizing that -- because it has happened -- there IS a mechanism, regardless of whether this is it or not.

moreover, that one is in a speculative bubble is NEVER simply as obvious as looking for plainly-visible inventory buildups, as so many now are, for if it were so easy there would never be a speculative boom in the first place.

i happen to think that your analysis is a fine one and very likely holds at least some water. but the more pragmatic point is that it is not nearly so important to know that you are right on the mechanism as it is to acknowledge that commodity bubbles do occur without what all would see in real time as inventory stockpiling.
5.29.2008 3:37pm
Richard H. Serlin (mail) (www):
Steve,

I think your missing something simple and very important: Speculation of commodities pushes up the spot price by taking a lot of the commodity off the market and putting it in storage, like oil or diamonds. But the inventories of most commodities of interest are reported, and they haven't been unusually high, which points to it not being speculation that's pushing up oil, etc., but rather fundamentals.

Here's Krugman on this from his May 13th blog post, "More on oil and speculation":


If the price is above the level at which the demand from end-users is equal to production, there's an excess supply — and that supply has to be going into inventories. End of story. If oil isn't building up in inventories, there can't be a bubble in the spot price.

So my challenge to people who say there's an oil bubble is this: let's get physical. Tell me where you think the excess supply of crude is going.


And from his April 20th post, "Commodities and speculation: metallic (and other) evidence":


We've had a huge runup in commodity prices — fuels, food, metals. But why? Broadly, the debate is between those who see it as a speculative phenomenon, driven by some combination of low interest rates and irrational exuberance, and those who see it as a collision of rapidly growing demand with constrained supply.

My problem with the speculative stories is that they all depend on something that holds production — or at least potential production — off the market. The key point is that the spot price equalizes the demand and supply of a commodity; speculation can drive up the futures price, but the spot price will only follow if the higher futures prices somehow reduces the quantity available for final consumers. The usual channel for this is an increase in inventories, as investors hoard the stuff in expectation of a higher price down the road. If this doesn't happen — if the spot price doesn't follow the futures price — then futures will presumably come down, as it turns out that buying futures produces losses.

As far as I can see, this creates real problems for any claim that high metal prices are speculatively driven. Food inventories are also historically low. I just don't see how a low-interest-rate or bubble story works here.


Richard
Blog: http://richardhserlin.blogspot.com/
5.29.2008 7:33pm
Steve Randy Waldman (mail) (www):
Nemo — I think the story you're describing is quite similar to the one I'm suggesting is the only one consistent with the "index spec bubble" story — trendy big money goes long, potential to arbitrage is restricted to producers, who find it in their interest to accommodate a bubble and enjoy it while it lasts. I don't know that this is actually happening, but it's the only one I buy that calls this an ordinary speculative bubble, rather than something fundamental, or a flight from money. If it is the story, it lasts until it doesn't, or until barriers to arbitrage are reduced, more on which later.

Melancholy — I'm not at all certain that we're going to see an oil price crash (in dollar terms), though I think we very well might. Re Masters' angle, I've no comment, but you might enjoy this.

Anon — What's harder to explain is where the shorts are who don't have much to fear, that is the commodity producers themselves. They can lock in certain profit by shorting. If they think doing so will incur a large opportunity cost, as prices continue to rise, then either they 1) perceive a bubble dynamic which they are intentionally failing to interfere with; or 2) are genuinely willing to gamble that either monetary depreciation or fundamental supply and demand will push up prices. If (1), we should address the limits to arbitrage, so the strategy becomes untenable. If (2), then this is not likely a momentum-driven bubble, as producers speculation is likely to be well-informed.
5.29.2008 10:36pm
Steve Randy Waldman (mail) (www):
Hubert — If producers are so willing to lock in prices, why do they seem so willing to roll rather than deliver? And (as Richard would ask), where is the storage, as primary storage by traditional producers would be measured as inventory (unless they are intentionally hiding).

Alessandro — I certainly don't disagree. Your story is speculation on a coming depreciation of financial assets, which I think is a big part of the story. (And you largely answer Richard re storage).

Rick — I don't think it's a sufficient answer to point out that there are no specs near the delivery date. If the specs have rolled into the next contract, calendar arbitrage might keep the spec-evacuated near contract price consistent with the spec heavy next, so you can't call the near contract "pure supply and demand". That said, I too am skeptical of Masters.

IA — Yup.

B9 — It's a good point that when monetary factors are introducing price volatility, that can cut both ways. If monetary ease / CB-provided "liquidity" has contributed to high commodity prices, an reverse monetary policy shock could trigger a gut-wrenching drop, with or without "index speculators". Driving away inflation-hedgers and central-bank-skeptics could be enough.
5.29.2008 10:45pm
Steve Randy Waldman (mail) (www):
David — You are hardly less eloquent. Au contraire. Thanks for the compliment, and I look forward to reading more of yours.

GM — I don't disagree with you, but the case for historical speculative commodity bubbles seems much weaker to me than the case for stock bubbles. I suppose tulip bulbs were a commodity, but that's too alien for me to think about. The Hunt Bros silver corner and the late 80s gold spike are bubbly, but they were both atypical. The silver corner was clearly intentional market manipulation, and it seems to me that the gold spike was a vote of monetary no confidence as well as mere momentum. Both were amenable to clear policy action to pop (tho' killing gold took the guts and pain of a Volcker), which isn't necessarily as clear with stocks (I s'pose 20% interest rates would do the trick there, too, but it's less clearly appropriate when the bubbly asset isn't a money substitute.) While I do think commodities can have bubbles to, I think fundamentally they are less bubbly, and that some pretty simple tweaks to market structure could make them less frothy still. I also think a flight from financial to physical assets is not necessarily a bubble — that can only be decided ex post, after monetary authorities have succeeded or failed to restore confidence without devaluing.

Richard — I'm not making the case for speculation here, though I do think the recent run-up (not the years long boom Masters describes) has speculative aspects, and monetary aspects. That said, I think the lack of visible storage is not disposative. As Alessandro points out, not producing is for some commodities equivalent to storage. Also, from Yves Smith:
1. Given the speed of the run-up, there may be a delay in hoarding taking place (real world buyers and sellers may have thought prices would fall back, as they did for a bit earlier this year).

2. Tankers full of Iranian crude are floating around the Gulf. Admittedly, this is nasty, less preferred crude, but it is still in surplus

3. The Chinese are very secretive, and known to be stockpiling diesel, and possibly crude as well to prevent any embarrassing outages before and during the Olympics. According to Xinhua, China's oil and oil derivative products growth 1Q 2008 versus 2007 is well ahead of GDP growth of 10.6%.:
China's net imports of crude oil was 44.95 million tonnes in the first quarter, up 14.9 percent, and net imports of oil products rose by 31.8 percent from a year ago to 5.47 million tonnes, according to General Administration of Customs. China's imports of diesel in the first quarter surged over 600 percent to 1.66 million tonnes and the imports of gasoline, rose by nearly twice to 76,654 tonnes.

The article also noted:
Deng Yusong, a researcher with the Development Research Center of the State Council, said that abnormal needs boosted by below-cost prices of refined oil products controlled by the central government over concerns of the country's rising CPI is another major reason contributing to the country's surging oil consumption.

In other words, domestic players suspect that the government will have to raise oil prices and are moving their purchases forward.

4. The IEA only counts primary inventory as inventory. Anything else is demand:
Demand is total inland deliveries plus refinery fuels and bunkers minus backflows from the petro-chemicals sector. It is thus equivalent to oil consumption plus any secondary and tertiary stock increases.

Further note how narrow the definition of primary stock is:
Unless stated otherwise, all stocks included in the report are primary. They include stocks held in refineries, natural gas processing plants, oil terminals and entrepôts (where these are known), pipelines and stocks held on board incoming ocean vessels in port or at mooring.

Thus any end user inventory, which is one place you might see hoarding, would not be included as inventories.

5. Finally, some suppliers may simply be choosing not to pump. Some readers and commentors have provided anecdotal evidence, and James Hamilton in his new paper also provided some quotes from the Saudis and Kuwatis along those lines. A reader gave the logic:
It is being left in the ground.

For over a hundred years the oil industry has been building out a huge infrastructure for mining, transporting, refining and distributing oil and its associated products.

It doesn't make much sense to add to this infrastructure, if we've reached the peak of physical throughput. Going forward, refining capacity will probably move closer to the source of crude, and more of the transport will be devoted to moving of product. But, that prospective shift is a detail.

There's no accumulation of speculative inventories in the refining and distribution chain, because there's no slack in that chain, and won't be any, because, looking forward, creating such slack makes no sense.

The only way to “hoard” oil effectively in these circumstances, is to delay oil production from fields with low marginal unit cost, while shifting production to fields or sources with high marginal unit costs. This conserves the expected rents, in a balanced way. That is, those with low marginal cost oil to produce see their wealth rising, while those with quasi-rents on infrastructure are protected.
5.29.2008 11:03pm
Independent Accountant (mail) (www):
SW:
I read an article today in the WSJ about a CFTC investigation of trading in oil markets. It got me to thinking. There was another WSJ article today about pension funds investing in commodities. Then it hit me, as to what is happening here. In about 1912, I'll have to check the date, Irving Fisher, economist proposed creating a "compensated dollar". The commodities index funds ARE Irving Fisher's "compensated dollar". We can't use TIPS to hedge. Why? The CPI is hopelessly manipulated to understate inflation. This is all so stupid. If Congress wants to discourage pension funds from commodities investments, I have a better idea: make Helicopter Ben stop the presses. I'll get you a cite on the Irving Fisher proposal.
5.30.2008 9:24am
Benign Brodwicz (mail):
Even the Economist seems confused.
Benign

Economist.com




Energy

Double, double, oil and trouble
May 29th 2008
From The Economist print edition


AFP
AFP


Is it “peak oil” or a speculative bubble? Neither, really

AFTER oil hit its recent record of $135 a barrel, consumers and politicians started to lash out in every direction. Fishermen in France have been blockading ports and pouring oil on the roads in protest. British lorry drivers have paraded coffins through London as a token of the imminent demise of the haulage industry. In response, Gordon Brown, Britain's prime minister, is badgering oil bosses to increase production from the North Sea, while Nicolas Sarkozy, the president of France, wants the European Union to suspend taxes on fuel.

In America, too, politicians are haranguing oil bosses and calling for tax cuts. Congress has approved a bill to prevent the government from adding to America's strategic stocks of oil, and is contemplating another to enable American prosecutors to sue the governments of the Organisation of the Petroleum Exporting Countries (OPEC) for market manipulation.

But the most popular scapegoats are “speculators” of the more traditional sort. OPEC itself routinely blames them for high prices. The government of India is so sure that speculation makes commodities dearer that it has banned the trading of futures contracts for some of them (although not oil). Germany's Social Democratic Party proposes an international ban on borrowing to buy oil futures, on the same grounds. Joe Lieberman, chairman of the Senate's Homeland Security Committee, is also mulling regulation of some sort, having concluded that “speculators are responsible for a big part of the commodity price increases”. The assumption underlying such ideas is that a bubble is forming, and that if it were popped, the price of oil would be much lower.

Others assume the reverse: that the price is bound to keep rising indefinitely, since supplies of oil are running short. The majority of the world's crude, according to believers in “peak oil”, has been discovered and is already being exploited. At any rate, the size of new fields is diminishing. So production will soon reach a pinnacle, if it has not done so already, and then quickly decline, no matter what governments do.

As different as these theories are, they share a conviction that something has gone badly wrong with the market for oil. High prices are seen as proof of some sort of breakdown. Yet the evidence suggests that, to the contrary, the rising price is beginning to curb demand and increase supply, just as the textbooks say it should.

Stocks, bonds and barrels

Those who see speculators as the culprits point to the emergence of oil and other commodities as a popular asset class, alongside stocks, bonds and property. Ever more investors are piling into the oil markets, the argument runs, pushing up the price as they do so. The number of transactions involving oil futures on the New York Mercantile Exchange (NYMEX), the biggest market for oil, has almost tripled since 2004. That neatly mirrors a tripling of the price of oil over the same period.

But Jeffrey Harris, the chief economist of the Commodity Futures Trading Commission (CFTC), which regulates NYMEX and other American commodities exchanges, does not see any evidence that the growth of speculation in oil has caused the price to rise. Rising prices, after all, might have been stimulating the growing investment, rather than the other way around. There is no clear correlation between increased speculation and higher prices in commodities markets in general. Despite a continuing flow of investment in nickel, for example, its price has fallen by half over the past year.

By the same token, the prices of several commodities that are not traded on any exchange, and are therefore much harder for speculators to invest in, have risen even faster than that of oil. Deutsche Bank calculates that cadmium, a rare metal, has appreciated twice as much as oil since 2001, for example, and the price of rice has risen fractionally more.

Investment can flood into the oil market without driving up prices because speculators are not buying any actual crude. Instead, they buy contracts for future delivery. When those contracts mature, they either settle them with a cash payment or sell them on to genuine consumers. Either way, no oil is hoarded or somehow kept off the market. The contracts are really a bet about which way the price will go and the number of bets does not affect the amount of oil available. As Mr Harris puts it, there is no limit to the number of “paper barrels” that can be bought and sold.

That makes it harder for a bubble to develop in oil than in the shares of internet firms, say, or in housing, where the supply of the asset is finite. Ultimately, says David Kirsch of PFC Energy, a consultancy, there is only one type of customer for crude: refineries. If speculators on the futures markets get carried away, pushing prices so high that refineries run at a loss, they will simply shut down, causing the price to fall again. Moreover, speculators do not always assume that prices will rise. As recently as last year, the speculative bears on NYMEX outweighed the bulls.

There is, admittedly, a growing category of inherently bullish investment funds that seek to track commodity-price indices, in which oil is usually the biggest component. Politicians have begun to denounce these “index funds”, since they make money for their investors only if prices rise. According to Mr Lieberman, they have grown in value from $13 billion to $260 billion over the past five years. This surge of investors betting on rising prices, many observers contend, has become a self-fulfilling prophecy, helping to push prices ever higher and thus attract yet more investment.

But Bob Greer, of PIMCO, an asset-management firm, argues that even index funds make unlikely suspects. For one thing, they too invest in futures, rather than in physical supplies of oil. So every month, they must trade contracts that are about to fall due for ones that will not mature for several months. That makes them big sellers of oil for prompt delivery.

What is more, their growth is not as impressive as it first appears. Paul Horsnell of Barclays Capital, an investment bank, puts the total value of index funds and other similar investments at $225 billion. That is less than half the market capitalisation of Exxon Mobil, he points out, and a tiny fraction of the $50 trillion-odd of transactions in the oil markets each year. Although index funds have grown quickly, that growth stems in large part from the rise in value of the futures they hold, rather than from fresh investment flows. He estimates that index funds swelled by $13 billion in the first quarter of this year, for example, of which all but $2 billion derives from the rise in commodity prices.

Back to basics

Mr Harris of the CFTC, for one, believes that the oil price is still a function of supply and demand. For the past few years, the world's production capacity has grown only sluggishly. Meanwhile, demand, especially from the developing world, has been growing faster. So there is hardly any slack in the system. Only Saudi Arabia and the United Arab Emirates are thought to be able to increase their output from today's levels, and even then, there are doubts, since Saudi Arabia, in particular, is secretive about the state of its oil industry.

That leaves the oil market at the mercy of even small disruptions to supply. Prices tend to jump each time militants sabotage an oil pipeline in Nigeria, bad weather threatens production in the Gulf of Mexico, or political clouds gather over the Persian Gulf.

The problem is exacerbated by a growing mismatch between the type of oil being produced and the refineries that must process it. The most common benchmark prices, including the one used in this article, refer to “light” crude, the least viscous sort, which produces the most petrol and diesel when refined. “Heavy” oil, by contrast, yields more fuel oil, which is used mainly for heating.

At the moment, diesel is in short supply and there is a glut of fuel oil. That makes processing heavy oil unprofitable for some refineries, since the gains from diesel are outweighed by losses on fuel oil. As refineries turn instead to lighter grades, it pushes their prices yet higher. The discount on heavier crudes has risen to record levels. But even then, points out Ed Morse, of Lehman Brothers, another investment bank, Iran is having trouble selling the stuff. It is storing huge quantities of unsold oil on tankers moored off its coast.

Presumably, Iran and other heavy-oil producers will eventually be obliged to drop prices far enough to make processing the stuff worth refiners' while. In the longer run, more refineries will invest in the equipment needed to crack more diesel out of heavy oil. Both steps will, in effect, increase the world's oil supply, and so help to ease prices.

But improving an existing refinery or building a new one is a slow and capital-intensive business. Firms tend to be very conservative in their investments, since refineries have decades-long life-spans, during which prices and profits can fluctuate wildly. It can also be difficult to find a site and obtain the right permits—one of the reasons why no new refineries have been built in America for over 30 years. Worse, new kit is becoming ever more expensive. Cambridge Energy Research Associates (CERA), a consultancy, calculates that capital costs for refineries and petrochemical plants have risen by 76% since 2000.

Much the same applies to the development of new oilfields. CERA reckons that the cost of developing them has risen even faster—by 110%. At the same time, oilmen remain scarred by the rapid expansion of output in the late 1970s, in response to previous spikes in prices, that led to a glut and so to a prolonged slump. Exxon Mobil claims that it still assesses the profitability of potential investments using the same assumptions about the long-term oil price as it did at the beginning of the decade, for fear that prices might tumble again. Environmental concerns are also an obstacle: America, for one, has banned oil production off most of its coastline.

Increasing nationalism on the part of oil-rich countries is adding to the difficulties. Geologists are convinced that there is still a lot of oil to be discovered in the Middle East and the former Soviet Union, but governments in both regions are reluctant to give outsiders access. Elsewhere, the most promising areas for exploration are also the most technically challenging: in deep water, or in the Arctic, or both. Although there have been big recent discoveries in such places, they will take longer to develop, and costs will be higher. The most expensive projects of all involve the extraction of oil from bitumen, shale and even coal, through elaborate processing. The potential for these is more or less unlimited, although analysts put the costs as high as $70 a barrel—more than the oil price this time last year.

Nonetheless, PFC Energy has examined projects that are already under way, and concluded that global oil production will grow by over 3m barrels a day (b/d) over the course of this year and next. In particular, it expects production outside OPEC to grow by about 500,000 b/d both years—a marked increase from the near stagnation of recent years.

Meanwhile, the high price is clearly beginning to crimp demand. The growth in global consumption last year was barely a quarter what it was in 2004 (see chart); this year, it is likely be even lower. In rich countries (or at least among the members of the Organisation for Economic Co-operation and Development (OECD), a rough proxy), the effect is even more pronounced. Consumption has been falling for the past two and a half years.

Poorer countries' demand for oil is still rising, albeit at a slowing pace. That is partly because their economies are growing faster, and partly because their consumers are shielded from the rising price through subsidies. But the increasing expense of such measures is forcing governments to water them down or scrap them altogether (see article). That, in turn, should further sap consumption.

Oil pique

China's growing thirst for oil is often put forward as one of the main factors behind today's higher oil prices. Demand for diesel there, for example, rose by over 9% in the year to April. But Mr Morse argues that such growth might not last. The government has ordered oil firms to increase their stocks of fuel by 50% to be sure there are no embarrassing shortages during the Olympics. It is also planning to run some power plants near Beijing on diesel rather than coal, in an attempt to reduce pollution during the games. These measures are helping to boost China's demand for diesel, but the effect will be transitory.

In the short run, neither demand for nor supply of oil is very elastic. It takes time for people to replace their old guzzlers with more fuel-efficient cars, or to switch to jobs with shorter commutes, or to move closer to public transport. By the same token, it can take ten years or more to develop an oilfield after its discovery—and that does not include the time firms need to bolster their exploration units.

Gary Becker, an economist at the University of Chicago, has calculated that in the past, over periods of less than five years, oil consumption in the OECD dropped by only 2-9% when the price doubled. Likewise, oil production in countries outside OPEC grew by only 4% every time the price doubled. But over longer periods, consumption dropped by 60% and supply rose by 35%. The precise numbers may be slightly different this time round, but the pattern will be the same.


Copyright © 2008 The Economist Newspaper and The Economist Group. All rights reserved.
5.30.2008 12:33pm
Francis (mail):
GM/Steve,
I am intrigued by GM's posit that you can have a speculative commodity bubble without a noticeable increase in inventories/storage. Especially for oil.

Clearly many commodities are different than stocks, real estate, bonds, gold, art, etc. in that they are not repositories of value upon consumption by an 'end user'.

Even leaving aside commodities such as gold and silver, I would also distinguish between commodities such as corn/copper and oil/coal because the former is replenishable/recyclable and the latter is depletes.

When I take the above into account, I am not clear how one can have a oil bubble especially when you layer in the highly efficient pricing mechanism that Steve has laid out between the physical and paper world of oil.

And it's tricky to also loosely interchange the term boom with bubble. I'm not saying that oil will not go down, but just take the view that if it does, it is mainly a function of demand destruction or supply increases.

And finally, I think the phrase 'speculators' is somewhat negatively loaded which could also skew the discussion.

I thank you in advance for you comments.
5.30.2008 12:39pm
burrite (mail):
Twenty plus years ago in intermediate micro-economics, our class spent an hour and a half on non-renewable resource economics. I got almost nothing from the lesson, except that low real interest rates meant that you should extract less of your oil now and more later. When rates are high, you should drill more now, and sell it to invest the money; when rates are low, you should leave it in the ground because you can't make much on the money you get for selling.

But it struck me even then that talking about interest rates was just a dodge to avoid talking about the more important question (namely, the shape of the demand curve now vs. the probable shape of the demand curve in the future), which was hopelessly complex.

Even in the simplest case, where there are NO possible substitutes, now or ever, for oil, and the slope of the demand curve is unchanged, doesn't the extraction decision have more to do with game theory than it does with real rates? If there are 20 producers, each with a finite amount of oil, and the demand curve is normal, wouldn't you ideally want to sell your oil after everyone else has exhausted theirs (especially if, as now, real rates in much of the world are near zero)? Couldn't you then take advantage of monopoly pricing, and cherry-pick your sales to only the users with the highest marginal value for the oil, rather than selling it today for MR=MC? But, of course, if all the producers perform the same calculus, then NO ONE will be selling oil today, and you can earn monopoly rents by being the ONLY one to sell today. But then others will sell today, so you'll want to wait until later. And so on.

If that's the SIMPLEST case, imagine the case where the slope of the future demand curve is unknown. What substitutes for oil will there be in 20 years? How about in 50 years? How much will they cost? How much will developing nations be willing to subsidize consumption? How effectively will the US have been in the interim at raising fuel economy standards, and taxing itself to build better mass transit infrastructure, and deploying wind, solar and other renewables? How about if those renewables themselves REQUIRE petroleum to deploy (wind turbines are made from carbon fibers, which come from propylene; solar grade polysilicon is creating by supplying an insane amount of heat, among other things, to purified sand)? Furthermore, is “peak oil” here now (or soon) or isn't it? Will the timing of diminished production capacity interfere with efforts to move to a less oil-intense infrastructure? And what about geo-politics? Is there a price of oil above which I run an unacceptable risk that a declining imperialist power will decide to invade my country? Am I willing to bet that someone else's country, rather than mine, will be invaded first?

How you answer these questions goes much further to determining how much of your oil reserves to drill now vs. later than does the real rate of interest, or the value of the dollar, or the volume of outstanding futures contracts.

For a very brief period in the late 80's, I was the world's worst real estate agent. To get licensed, I had to take an appraisal class, from which I remember nothing, except this: In real estate (as in everything) there are three ways of determining price…1) Replacement cost (ie, the cost to buy the land and build the house); 2) Market Comparables (ie, what are similar houses selling for?); and 3) Discounted Cash Flow (ie, what's the NPV of the income that can be produced from the property). The first two methods should be used if the market is competitive (if there are other existing or potential properties similar to the one you're trying to sell), while the third should be used if the market is not competitive (if there are no ready substitutes for the property in question).

Now, obviously, some variant on these three methods determines how prices are set for EVERYTHING, though in the case of commodities such as oil (where it's not exactly clear HOW competitive the market is, or HOW available/costly are the substitutes, or HOW MUCH is available to sell), the difficulty of determining which combination of the three methods to use becomes overwhelming.

But notwithstanding the complexity, isn't it quite likely that DIRECTIONALLY, the oil market has started to move away from competitive (replacement cost) pricing, and toward oligopolistic (discounted cash flow) pricing? There are several reasons why this could be true:

1. Increased plausibility of “peak oil” speculation. Whether ultimately true or not, disappointing supply response from major producers in the face of a significant price signal has added plausibility to the Hubbert's Peak argument. Maybe they want to, but they can't. Or (and this comes to the same thing), maybe they can but they won't, because they've found “peak oil” to be a convenient anchor on which to base more effective collusion.

2. Decreased fear of an invasion by the United States in response to limited supply, due to expense and political fallout from the Iraqi adventure.

3. Increased trust, as time has gone on and price has remained high, that other producers won't exploit the situation and ruin it for everyone.

4. Decreased fear that too high a price will lead to more investment into bringing down the cost of substitutes. Partly, this may be due to the declining balance of trade positions of the consuming nations, which now have less resources available to implement “energy independence” strategies. Partly it may be that consuming nations have shown less than expected political will in this area. Partly it may be that, with “peak oil” here sooner than expected, it's now more rational to believe that consuming nations won't have TIME to implement these strategies before scarcity becomes acute, even once they acquire the will (remember, most of the renewable resource strategies, other than nuclear, require lots of petroleum).

5. Increased optimism that rapid growth from developing nations will keep the demand curve shifting outward even in the face of increased substitution by OECD nations.

You could easily add 10 more items to this list if you chose. My point is, for decades, oil acted more or less like a competitive market, with price bearing some relationship to the production costs of the incremental supplier. OPEC's activities, though not completely unsuccessful, were never sufficient to inhibit supply to such a degree that price became unrelated to extraction cost. And as a result, oil producers could make some intelligent guesses about how much to invest in new technologies for extraction, because they could know with REASONABLE certainty, how much impact that the increased supply was apt to have on market price, and whether the money spent to exploit higher-cost locales would return at least the cost of capital.

Now, that relationship has become unmoored. Instead of being related to production cost, oil price now bears some relation to the price of substitutes and its value to consumers, but through equations that are hopelessly complex, and certainly far, far beyond the predictive capabilities of all industry observers (including the oil producers themselves). It's little wonder that, even with oil at $125, the major integrated oil companies still balk at exploiting projects that require oil at $70 or even $50 to be NPV positive. Even if they were blind to the sweetened possibilities of exploiting monopoly rents by delaying action (which I do not believe), they are paralyzed by a dawning recognition that their planning tools are worthless when it comes to predicting future prices.

So, what, then, do I think is the “right” price for oil now that producers have started to effectively act like oligopolists? Unfortunately, I have no idea. There IS an equilibrium price, reflecting the current realities of substitutes, subsidies, economic growth, and demand elasticity…I just don't have the hubris to try to quantify it. One thing I AM sure of, however: It will be uncovered the way ALL new equilibria are uncovered…by trial and error. This necessarily means that price will continue to be forced up until it is CLEAR to a majority of market participants that it has overshot the market clearing price. Given the very limited short term elasticity of demand, it strikes me that it will take a very strong price signal to make this happen. Today's price, which has only just recently surpassed the inflation-adjusted all-time high of the late 70's, is not yet close to that price. We may very well be close to the “equilibrium” price, or we may be above it, but I think it is very unlikely that we are yet close to the “overshoot price” sufficient to signal to market participants that we've gone too far.

Perhaps I'm wrong. Maybe the foregoing will prove to be nonsense, and oil will go back to $50 in a year. I hope it does. But if it does, I think the REASON it does will have nothing to do with banning evil speculators. It will be because there will be some change in the “story” of long term oil supply vs demand that will undercut the confidence of the market in peak oil/scarcity/etc, and cause a breakdown in the collusive behavior of the producers. If so, those who will say that the oil run-up was a bubble will be proven correct. Having lived now through an internet bubble, a housing bubble and a credit bubble in just the past decade, however, we've all become bubble connoisseurs. As a result, we know that the more plausible the phony story that launches the bubble, the bigger it will get before it bursts. And from my vantage point, if you compare this “bubble” (if it is a bubble) story to the other three, it's a pretty good one.

The “internet grows to the sky” story required you to believe that there was an inexhaustible supply of pimply-faced twenty-something managers who “got it” to such a degree that they would be able to exploit a new mass-medium rewrite the rules of competitive success in countless industries. The “house prices grow to the sky” story required you to believe that there was a perpetual-motion machine whereby a nation's citizens could increase their wealth endlessly by building houses and selling them to one another. By contrast, the “peak oil” story requires only that you believe we have extracted 50% of the total recoverable quantity of a non-renewable resource with a particularly low elasticity of demand, so that further growth from developing markets can overwhelm efforts by consumers to find effective substitutes. Again, time may well prove this story to be as foolish as the others, but if a good bubble requires a good story, you have to admit that oil has got one at least as sound as other recent examples. And given that folks were calling for the end of those other bubbles at least 2-3 years before they finally burst, it suggests that the bubble call in oil, even if right, could still be quite a bit early.
5.30.2008 1:11pm
Nemo (mail) (www):
Hi, Steve. Thank you for your reply.

I think the story you're describing is quite similar to the one I'm suggesting is the only one consistent with the "index spec bubble" story — trendy big money goes long, potential to arbitrage is restricted to producers, who find it in their interest to accommodate a bubble and enjoy it while it lasts

Well, that is not exactly what I mean. I am suggesting that a rise in long-term futures contracts creates a direct incentive for producers to curtail (or more precisely, delay) current production.

The observation underlying Hotelling's Principle is that for a finite resource, the relevant question for a producer is not "how much should I produce today?", but "when should I produce each barrel?" As Hotelling observed, if I can make more money by curtailing my production for (say) a year, then I will; and that decision depends on the cost of production today, the price of oil today, the prevailing rate of interest, the cost of production one year from now, and the price of oil one year from now. But futures contracts can directly influence the price of oil one year from now, and therefore my decision of how much to produce today.

Imagine a long-only oil fund that buys contracts one year out and rolls them forward again at expiry, never taking delivery. Since it never takes delivery, such a fund would have no influence on the spot price when they rolled. But if enough money were committed to such a fund, it would keep the one-year futures price elevated indefinitely. A rational producer would curtail current production to capture that elevated price. In this way, the fund's "irrational" futures bet would always become a self-fulfilling prophecy, almost. In the steady state (and absent any actual supply/demand shocks), this strategy would lose money, but slowly. Possibly very slowly if interest rates -- which ultimately govern the difference between spot and futures (Hotelling) -- are very low.

Why would anyone invest in a fund that is tied to oil prices but with a "slow bleed"? For the same reason anyone would invest in the GLD ETF...
5.30.2008 1:12pm
Richard H. Serlin (mail) (www):
Steve,

With regard to reading Hayak, who makes some good points if you keep in mind they are about only certain factors in the economy, not the economy as a whole -- so I would disregard the silly simple exageration, the assumption that the simple model behaves exactly like the actual economy, I hope you'll please take just a little time to read some Krugman commentary on the Austrians from his 1998 Slate article, "The Hangover Theory". A quick quote:


A few weeks ago, a journalist devoted a substantial part of a profile of yours truly to my failure to pay due attention to the "Austrian theory" of the business cycle—a theory that I regard as being about as worthy of serious study as the phlogiston theory of fire.
5.30.2008 2:14pm
Richard H. Serlin (mail) (www):
Steve,

I'm not expert on commodity inventory estimates, so I look to people like Krugman, a man who has obviously shown himself to be immensely intelligent, competent, well connected, and well intentioned.

Yves Smith, in his profile, says ideas should stand on their own merits and not on the messenger, and this is true to a large extent, but the messenger is still usually important for a number of reasons including:

First, ideas often depend on raw facts and data that the reader doesn't know, so the messenger could be lying or mistaken about them. If the messenger is highly credible and competent, this can greatly increase the odds that the facts and data are essentially correct.

Second, ideas are often complex. You can make a mistake in evaluating them at first blush. If you agree with the idea, and the messenger of the idea is one you know to be highly competent and trustworthy, then you know the odds are greatly decreased that you have made a mistake and there's actually something very wrong with the idea.

So the messenger matters, although many great ideas and insights come from those who have not yet established a reputation, so you should try to give them a hearing to the extent that it's reasonable.

In any case, for things like commodity inventories, I'm not an expert, so I demur to the ultra-intelligent Krugman, who over his vast career has become an expert on a great number of things, and is also super connected and resourced with Princeton, the New York Times, etc., etc., to find things out.

He doesn't see excess inventories and actually responds to Yves Smith in his May 12th post, "Why oil isn't gold":


Here's the key fact: as nearly as I can tell, private gold stocks — gold held as a store of value and/or for speculation — are equal to about 50 years' worth of production.

Meanwhile, private stocks of crude oil are equal to about 50 days' worth of production. Yes, we should also add stocks of refined products; and there may also, as Yves Smith likes to remind us, be some additional unrecorded stocks (in 1979, I remember, everyone was filling their gas tanks as often as possible, adding to the speculative demand); but the fact remains that oil prices are much more closely tied to the flow of production and consumption than gold, or any other good that is mainly held as an asset rather than actually consumed.


With regard to just keeping oil in the ground, I haven't researched this thoroughly, but I am an avid daily reader of the New York Times, Washington Post, etc., and over the last couple of years I don't remember reading anything about OPEC significantly cutting production. I actually seem to remember them somewhat increasing it.
5.30.2008 2:23pm
Independent Accountant (mail) (www):
The NYT article about Irving Fisher can be "googled" and found through "compensated dollar". I agree with Krugman and have so stated at my own blog.
5.30.2008 3:12pm
Steve Randy Waldman (mail) (www):
Francis — The question of storage is (almost by definition, if you accept a wide definition of storage) a question of the balance of speculation. For every long in a futures market there's a short. The short is either hedged or not. If speculators are balanced on both sides, there's needn't be storage. But if the longs are speculative and the shorts not, then the shorts must be hedged, that is, directly or indirectly they must have the capacity to deliver, which we can call storage, though it also includes production capacity. (Interestingly, if the preponderance of speculation is on the short side, that must be balanced by a capacity to nonwastefully consume.)

I agree with you both about the connotations of "speculator" (I consider myself proudly a speculator, BTW), and the difficulty of discriminating between a bubble and a boom, especially now.

Nemo — I agree with you that prices at any maturity force prices to move at all times by arbitrage, so certainly bidding up 1 year forward contracts would bid up the rest. (At equilibrium, the arb relationship has to be consistent with Hotelling's Principal, right? The scarcity rent forces up the resource price at all times, but there is still no storage arbitrage.)

But the question is, how can "index specs" bid up 1-year forward contracts so consistently, when someone has to take the other side. For the last 7 years, have there really been specs on the other side, betting on a burst bubble, losing every time? Why haven't producers forced a downspike by insisting on more deliveries that the vast speculative community can deliver? If your answer is "why would they, they like high prices", I don't disagree, but then you agree that the problem is more likely that arbitrage is not being aggressively pursued on the short side than that there are bidders on the long side. The longs are disunited, dumb, and big. It's the short side that's running this show, if there is any show. Ironically, shorts are always criticized for driving he price of assets downwards. The issue here is that they might not be driving prices downward aggressively enough, that they might find it profitable to take losses on their short futures positions to fan the flames exuberance about their net worth.

burrite — I agree with you. It is really very difficult to say what the "right" price of oil should be. I don't think I like the notion that replacement cost is competitive and DCF not, I think fundamentally they'd take you to similar places. I think that the place you are going is not dissimilar to where James Hamiltion arrived, that is that scarcity rents have been priced into oil recently in a way that they were not previously. Some of the reasons for that, though, as you outline, is that consumers have not bargained very hard with producers, so there has been little cost to pricing at the maximum consumers would pay, that is extracting all the "surplus", as the economists say. I think the reasons for that non-bargaining might turn out to be as much political as anything else. I mocked conspiracy theorists who bought oil cos when Bush was elected, but of course they turned out to be right. It'll be interesting to see if and when consumers stop acting like price-takers and instead become game-theoretic actors doing what they can to get what's there's, how much ability oil producers (whose own longevity is always at issue, they are not infinite horizon agents) will have to bargain all the surplus for themselves.
5.30.2008 6:31pm
Steve Randy Waldman (mail) (www):
Richard — I'm a big Krugman fan myself, but it's no fun having a conversation if we're just gonna defer to authority. I don't take a strong position on whether or not there is in fact speculative inventory build-up, but while Yves Smith has an easier job (she argues we don't know, Krugman argues that we know it's not happening), so far I'll side with her. I think no one has good enough information to quantify how much storage — in obvious tanks, in odd places, in foregone production — is happening in real time. That said, I agree with Krugman (and disagree with Masters) about the years-long ramp of commodity prices. I think we would have observed storage over such a long period of time, and would have seen more arbitrage by shorts. Overall, I agree with Hamilton, that the long march up has been driven by things other than a "speculative bubble", but something funny's been happening for the last nine months or so.

Re "The Hangover Theory", I'm very familiar with it. I'm a bit more Austrian-ish than Krugman, so I take a bit of umbrage, but his point is well taken that some Austrians Victorians seem to think that all good times must, as a karmic or moral matter, be balanced by bad, and that's just dumb. We should strive for as much fun and little hangover as possible, sure. But I think Krugman's position on this, which is very mainstream among American economists (Mark Thoma and Brad DeLong have expressed similar views) has been ultimately destructive, in that by emphasizing that bad times needn't follow good, we stopped trying to discriminate between the kinds of good times that would lead to bad and the kind that were just good. Instead, we've adopted a utopian position that if it's a good time today, that's cool however we're doing it, and we'll find a way to make a good time tomorrow, because no pipers must be paid. In fact, we've adopted the point of view that the best policy is always to prevent short-term bad times, however we got where we are, because no pipers must be paid, and that's simply untrue. To use the hangover analogy, Krugman is right to point out that having lots of fun needn't provoke a hangover, but it would be wrong to extend that by suggesting no kinds of fun, including fun by drinking to oblivion, need be followed by a hangover, and therefore it's good to have another hair o' the dog to provide quick relief if that's where you are. It's worth noting that Krugman is an intellectually flexible guy, in the good way (ie Keynes, when the facts change, I change my mind), and his writing this decade on this as many topics has been more nuanced than his writings over the nineties. He's penned several NYT pieces that have suggested we've mismanaged things so starkly since Y2K that some bad times are inevitable. Very Austrian, really, though he might not like it put that way.

IA — Here's a decent ungated "compensated dollar" reference, which I need to read more carefully. I spend a lot of time trying to think of monetary arrangements that would not be simple commodity money (that would be in some sense "elastic", and not tied to supply and demand for an arbitrary good), but that would be more automatically attached to value than central-bank-managed fiat money. I sense this is an interest of yours as well. From skimming the article I found, I'm not sure how excited I am about this proposal, but I need to skim it somewhat more deeply, and perhaps look elsewhere as well.
5.30.2008 7:00pm
Steve Randy Waldman (mail) (www):
IA — BTW, the proposal in the next post is partly motivated by monetary thinking. Easily tradable claims on a very wide variety of stored goods via "commodity banks" (potentially options on ready service capacity as well) are naturally moneylike, and ideally, if such claims are sufficiently easily exchanged and there is sufficiently broad goods &services coverage, the "supply" of money would naturally grow with absence of scarcity and ease of storage, which seems right somehow. Of course, this is just like going back to non-fractional gold bank money, except not restricting the money/banking commodity to gold.
5.30.2008 7:12pm
Richard H. Serlin (mail) (www):
Steve,

I think Krugman generally thinks that even with the best monetary policy, there's just too much randomness to do it perfect, and so there will be some bad times, but good monetary policy will make them shorter and less bad.

Much of what you say refers to the long run. Good monetray policy can keep employment overall pretty high, but long term growth requires capital creation and technological advance, and that requires smart saving and investing, which we've had much too little of -- one of the most harmful results of conservative slogan economics.

For long term growth, my favorite economists are Nobel prize winner Robert Solo, who really influenced me as an undergrad and yet another Paul, Romer.
5.30.2008 8:04pm
Independent Accountant (mail) (www):
SW:
Thanks for the reference. I will read it and let you know what I think. To be clear, I do not favor Irving Fisher's "compensated dollar" program. I am an Austrian and subscribe to von Mises "index number problem". I have seen how we CPAs tried to make a usable "inflation accounting" in the 1970s. Nothing seemed to come out "quite right". Do we use: CPI, PPI, replacement cost, etc., etc.? Today I believe the best we can hope for is: gold. That we can have an "unbiased price level", which next year is as likely to be higher as lower, is all I can realistically see. Price level stability is a chimera to me. A "random walk" of the "price level" would be fine.
6.2.2008 10:24pm
Note: You don't have to register to comment! Just use the pull-down menu to comment as "Guest".

Post as: [Register] [Log In]

Account:
Password:
Remember info?