Thursday, February 3, 2011

Why Oil Matters More Than Rubber

I was thinking about Paul Krugman's Cross of Rubber column, and in particular the associated blog post Commodities: This Time is Different.  My take on Krugman is that he's an extremely brilliant guy who's been thinking about economics for a good long time.  His enormous knowledge and insight are invaluable, and I pay close attention to his writing.  However, I also think he's gotten into some pretty deeply scripted habits of thinking based on past events and isn't paying close enough attention to the ways in which the present and the future are likely to be different than the past.  In particular, his frame of reference for the events of the last few years has been past deflationary episodes such as the Great Depression in the 1930s and Japan in the 1990s.

However, he's smart and pragmatic enough that he at least pays attention to facts that don't altogether fit his narrative, and in particular in recent weeks, he's been paying increasing attention to the fact that commodity prices are rising rapidly, even though the US and European recoveries have barely gotten going.  To me, ever since I started thinking about economic issues five or six years ago, these "finite world" issues have seemed to be a central fact of the current era that we need to come to grips with, but to Krugman, it's a bit of a distraction from his main theme: our biggest problem is lack of aggregate demand in the advanced economies, and we need more government fiscal stimulus and more liberal monetary policy to solve that problem.

So in the blog post he linked above, he starts by looking at the last 12 months change in prices of a range in commodities, and orders them according to that:

Obviously, many commodities are going up in price, and this reflects in part the fact that China, and to a slightly lesser extent, other developing economies are trying to raise their standard of living to western levels, and the global economy is struggling to come up with the raw materials to make that happen.

Now, obviously, some of these commodities are not in any long-term supply shortage. Iron, for example,  is #2 on the above list, and yet iron represents about 5-6% of the earth's crust.  It's going to be a very long time before humanity runs short of iron in some absolute sense.  Any price rises have to be due to basically temporary factors that can be ameliorated by opening some more mines (at some environmental cost in digging up some more of our beautiful planet).   Obviously, by focussing on a 12 month time frame, Krugman also opens the possibility of just meaningless noise getting into the list.

Still, looking at the systematic list caused me to think more deeply about my own focus on oil as the central commodity to worry about (with food second).  Why is oil more important than rubber, or iron, or cotton?

The answer comes from thinking: what happens as commodity prices get higher?   What is the economic effect?  For the most part, consumers aren't directly exposed to commodity prices.  No-one goes to the mine and buys iron ore (well, except this guy), or buys raw latex from rubber plantations.  Instead, we buy cars and toasters with these things in, and we are only going to modify our behavior with respect to buying and selling if the price of cars and toasters changes, not the price of iron and rubber.  And of course, inflation in the price of finished goods like cars and toasters is extremely subdued in the western world.  The reason is that commodities are (currently) a small part of the cost structure of most manufacturers, and since there is overcapacity in the economy, manufacturers cannot pass increases in their input costs onto consumers.  So these commodity price increases aren't going to have a visible effect on measures like core inflation for a while.  And as long as they don't, they can basically be ignored in thinking about macro-economic issues (which is exactly what Krugman would prefer to do - ignore them).

However, oil is the exception!  We may not buy raw rubber or sheet steel very often, but we do buy gasoline and diesel, and those products have very limited value-add over straight crude oil.  Refining margins are a small fraction of the final cost of those products, and crude oil and gasoline prices move in very close tandem.  And we really, really, want the oil that we want.  We aren't indifferent and willing to  happily substitute something else - instead we pay and pay and insist on driving anyway.  In economist language, oil demand is very inelastic in the short term (the elasticity in recent years has been about -0.05, though I wouldn't be surprised to see it get a bit more elastic in the next decade, now that we can't put it on the credit card quite so easily).

So oil is the one commodity price that cannot be ignored in the developed world.  If the world economy does indeed continue to recover rapidly, oil supply is not going to keep pace, oil prices will go up sharply, and that is something that will have macroeconomic consequences that Krugman will not be able to ignore.  I wouldn't be at all surprised to see the shock come this summer, but it could be delayed another year or two.  But come it will.

Food commodities are slightly different in their impact.  In the western world, we largely eat processed foods which behave economically more like cars and toasters - they only respond slowly and moderately to commodity prices.  Where food commodity prices really hit is in the urban third world.  The average resident of a big developing world city is poor, so they have to eat more basic foods with a high input of commodities, and, unlike their rural brethren, cannot grow much food of their own.  So, when food prices go up, we get riots in big poor cities.  This is something rulers have had to understand since the Roman Empire (which subsidized wheat for the poor in Rome).   The wave of hasty food buying and subsidy increases across the Middle East in recent weeks is a symptom of rulers quickly relearning this fact in response to events in Algeria, Tunisia, and Egypt.

But still, that does not mean that we in the privileged west can ignore these increases either.  In an interconnected global world, revolutions in far away countries can cause strategic realignments or economic problems that will affect us all.

So, it's not rubber, cotton, and iron we need to worry about.  It's oil and food.


Anonymous said...

And yet we live in a country where virtually all of the economic policy makers think, talk, and act in terms of "core inflation", that is, inflation ex. food and energy.

Seriously, WTF? It's like driving a car by watching the battery light indicator instead of the road.

Glenn said...

People tend to think of "Bread and Circuses" as a key in the downfall of the Roman Empire. They forget that the Western Roman Empire maintained both Panem and Circusum for about 500 years before the final collapse, and the Eastern, or Byzantine Empire another thousand years after that. I don't think the U.S. or the industrial world will match either though, despite high fructose corn syrup, computers and television. We're going through our resource base _much_ faster than previous empires. Ain't technology grand?


Stuki said...

If, as the data seems to indicate, commodity prices increase while final goods prices stay the same, the difference comes directly out of the take available to those involved in the intermediary steps. Who are more likely to reside in the developed world, than those who benefit from higher commodity prices. Norwegians and perhaps Icelanders excepted.

This will either show up as slower wage growth, or declining margins; the latter which will also over time also result in less labor demand, hence slower wage growth. There's really no way around this; hence at least part of Krugman's rich world demand side weakness, is due to to rising commodity prices blunting stimulus spending.

Just like everything else, there are diminishing returns to buying more demand as well (via low rates, QE and deficit spending), and one of the channels through which this manifests itself, is the input costs channel.

Greg said...

My reading of Krugman leads me to think that he doesn't have one main theme, but several, relevant to this post:-

o In the USA, the immediate-term big problem is unemployment, which is caused by lack of aggregate demand (the problem is unemployment; low demand -- and not structural problems or anything else -- is the cause)

o Europe's immediate problem is that it is not an economically sensible unit: policies appropriate for Germany (and the Netherlands and Finland) are utterly wrong for Ireland, Spain and Portugal

o globally, the chief short-term (not immediate-term) problem is the lop-sidedness of China's economy (in the medium term, this will be China's political problem more than others' economic problem, perStein's law;

o high commodity prices are a transient and not particularly important phenomenon: in the short and medium term, people will use substitutes for persistently high priced commodities, and in the long term, the structure of demand adapts (fewer cars = less steel and rubber and more yoga/meditation lessons to help cope with using public transport); the owners of those commodities that are still scarce will get a larger share of the pie. (Remembering that scarcity is a relative attribute.)

It's on the short-term part of this last point that there is room for debate -- especially the "not too important" bit.

Given that US oil demand is inelastic, the question is, what is the oil-price elasticity of other consumption? For households at or above the mean income, oil is probably a small part of the budget; many of those below the median income receive some sort of transfer (food stamps, say) which would support their consumption. High oil prices may just increase the federal and state deficits.

As you say, we'll likely find out in six or eighteen months.

Alexander Ac said...


Paul Krugman still thinks that QE1 and QE2 were disappointing because less money was thrown onto economy than needed!

Finally, I am not in that agreement that there is NO speculation in food (and oil) prices, Ilargi writes Zombie money kills real people:

Food prices are rising partly because, let’s not forget, China, unlike the US, does have inflation, with its money supply going through the roof. But much more than that they're rising because we have elected to kill off the principles of our own western economic systems, which were once supposed to be based on free market ideas, that dictate that success is rewarded and failure punished.

Etc... worth read it all.

And finally, I think australian economist Steeve Keen (debt deflation blog) has much higher S/N ratio than Krugman, even though he only recently got in touch with Albert Bartlett and his explanation of exponential function...


Anonymous said...

Iron is relevant:

"Look at the way that rules governing the availability of other resources go haywire when applied to energy. When North America’s deposits of high-grade iron ore were exhausted, for example, the iron industry switched over to progressively lower grades of ore; these contain less iron per ton than the high-grade ores but are much more abundant, and improved technology for extracting the iron makes up the difference. In theory, at least, the supply of iron ore can never run out, since industry can simply keep on retooling to use ever more abundant supplies of ever lower-grade ores, right down to iron salts dissolved in the sea.

Try to do the same thing with energy, by contrast, and two awkward facts emerge. First, the only reason the iron industry can use progressively lower grades of ore is by using increasingly large amounts of energy per ton of iron produced, and the same rule applies across the board; the lower the concentration of the resource in its natural form, the more energy has to be used to extract it and turn it into useful forms. Second, when you try to apply this principle to energy, you very quickly reach the point at which the energy needed to extract and process the resource is greater than the energy you get out the other end. Once this point arrives, the resource is no longer useful in energy terms; you might as well try to support yourself by buying $1 bills for $2 each."