Oil Dollar Euro ULP

I’m interested by the claim repeated in numerous news analyses that “investors are buying oil to hedge against the sliding greenback” – it seems to me to be a tad more significant that world oil demand has exceeded world oil supply for the last several months, and looks like doing so for at least the rest of this year. My forecasts for year end 2008: $145 barrel; A$1=US$1; ULP at $1.87 per litre. Enjoy.

I’m interested by the claim repeated in numerous news analyses that “investors are buying oil to hedge against the sliding greenback” – and also the likewise repeated claim that we are only now in record territory, because the oil price reached an inflation adjusted USD100+ in 1980. Both of these statements point us toward looking carefully at the history and current position in terms of what has actually happened in oil prices, inflation generally, exchange rates, and various measures of energy intensity – ie what you actually get out of a barrel of oil in terms of, for example, mileage or production generally.

US Inflation

Let’s start with the inflation claim. A useful CNN article from a couple of weeks back explains some of the complications:

Crude prices are within the range of inflation-adjusted highs set in early 1980. A $38 barrel of oil then would be worth $97 to more than $104 today, depending on the how the adjustment is calculated.

A direct comparison with daily Nymex prices is difficult because historical data, gathered before the crude futures contract was created in 1983, are based on average monthly prices posted by oil producers.

Different analysts have varying benchmarks for an inflation adjusted high. For example, John Kingston, director of oil at Platts, the energy research arm of McGraw-Hill Cos., arrived at an all-time high of more than $104 a barrel, which he said adjusts for delivery costs to and from Cushing, Okla., the Nymex crude oil delivery terminal. Using his own inflation adjustment, A.G. Edwards & Sons oil analyst Eric Wittenauer arrived at a widely-quoted estimate of $103.76.

However, an inflation calculator maintained by the Bueau of Labor Statistics estimates that $38 in 1980 is worth $97.34 today. A Federal Reserve Bank of Minneapolis calculator puts $38 in 1980 dollars at $99.43 today.

Aussie Inflation

From our point of view at the edge of the world economy (geographically, anyway), we can add further complications of our own exchange rate and inflation. Fascinatingly, the Aussie in March 1980 was worth … US92¢ ! – so no significant exchange rate adjustment needed. Australian inflation since 1980 (using the RBA inflation calculator) is a little over 250%, so, working it through: a US$38 barrel in March 1980 was A$41.30, inflated to today is A$145.80, which at  today’s rate of 94.6¢ is US$138. I leave it to the economists among you to work out just why the difference – and the underlying question of why the exchange rate and inflation rates have diverged so much. Actually, to help you with this piece of homework, you probably only have to wonder about the last two years, since the equivalent figures in April ’06 would have been: Aus inflation 232%, inflated barrel A$137.30 at US74¢ = US$101.60, which is pretty much exactly in line with the inflated figure for the US at that date.

Exchange Rates 

Unwinding ourselves from the convolutions of the calculations in the last paragraph, we can see that the 26% increase in the USD value of the Aussie dollar over the last two years has been a substantial analgesic against the pain of the 57% increase in the oil price over that time.

I am, of course, being Aussie-centric in that paragraph: the trade-weighted index for the Aussie is only up 12% since April ’06, and the USD is 11% of the TWI, so around two-thirds of the analgesic came from the general fall in the US dollar against everyone else, and one-third from the appreciation of the Aussie against everyone else.

This brings us back to that claim that the price rises are currently being driven by the fall in the dollar. In 2008 so far, the headline West Texas Light Crude oil price is up a nice round 10%, from $100 to $110, while the dollar against a not-randomly-chosen euro is down 3.3% (from 0.666€¢ to 0.645€¢). In fact, it is probably worth at this point looking at a slightly longer history of oil prices in euros vs dollars (and we’ll get the Aussie in at the same time, for completeness). To save space, I’ve only included the points at which the oil price crossed a headline round ten-dollar increment.

WTLC   when?     €:$    price in €  A$:US$  price in A$
$110    Mar-08    1.55    € 70.97    0.93    $118.28
$100    Dec-07    1.5      € 66.67    0.87    $114.94
$90    Oct-07     1.43    € 62.94    0.9      $100.00
$80    Aug-07    1.38    € 57.97    0.83    $96.39
$70    Apr-06     1.23    € 56.91    0.74    $94.59
$60    Jun-05     1.22    € 49.18    0.77    $77.92
$50    Jan-05     1.31    € 38.17    0.77    $64.94
$40&nbsp
;   May-04    1.2      € 33.33    0.7      $57.14
$30     Oct-00     0.82    € 36.59    0.53    $56.60
$20     Jul-99      1.03    € 19.42    0.66    $30.30

The first and most obvious thing to say from this table is that the dollar has been pretty steadily depreciating against current strong currencies like the euro and the Aussie for a very long time: the euro to dollar rate is pretty much a straight line from 2000 to 2008, with a pimple on it in early 2005 and just a tad of acceleration in the last six months. The Aussie to greenback chart is almost identical apart from a wider pimple.

The oil price chart, on the other hand, wobbles around all over the place until 2004 (there are three separate places where the $30 mark is passed for significant periods of time), climbs fairly steadily at around a $ a month until mid-2007, then accelerates to a steady $10 per quarter from then on. There is no real relation between this graph and the dollar depreciation graphs at all.

Well, it may yet be part of the explanation, but there’s no evidence here for the “depreciation drives price” theory: the (over-simplified) euro-denominated price graph is interesting enough to include here:

The underlying shape is the same as the USD oil price graph but has those two interesting flattish areas – on the face of it, there is some evidence to suggest that  the oil price has been fairly steadydenominated in euros for quite long periods of time, but has in fact accelerated over the last year in exactly the same way that the USD price has.

Also, prima facie, unless whatever forces are driving the oil price change, we can reasonably expect this trend to continue, and see the next few milestones as $120 in June, $130 in August/September, and the mid-$140s at the year end. Of course, if we took our trend of the last two weeks,  it would all be a lot faster than that.

What happened in mid-2007?

One thing that didn’t happen was that world economic growth didn’t accelerate – in fact, it began to slow down, which on the face of it should have taken pressure off of prices. Here’s CNNmoney again:

Back in October, when oil prices were near $90 a barrel and the economy was still humming along economists said high oil prices shouldn’t cut into economic growth. The economy used oil more efficiently than it did in the 1970s, and spending on gas was just a small percent of people’s budget, the experts said.

Fast forward to March and you’ve got a sputtering economy, and economists saying $105 oil deserves a big part of the blame.

We’ve already noted that the dollar depreciation curve did accelerate, but by nothing like the extent that the oil price did.

So, what did happen? Well, here’s a handy table extracted from the International Energy AgencyWorld Oil Balances report:

    Date       World Oil Supply      World Oil Demand
1Q2006         85.4mbd               85.5mbd
2Q2006         85.1mbd               83.4mbd
3Q2006         85.7mbd               84.4mbd
4Q2006         85.4mbd               85.7mbd
1Q2007         85.6mbd               86.0mbd
2Q2007         85.2mbd               85.0mbd
3Q2007         85.2mbd               85.6mbd
4Q2007         86.4mbd               87.2mbd
mbd = millions of barrels per day

Now, can we notice something here? What’s the trend in the supply column? And what happened in 2007? And does it help if we point out that the IEA forecasts demand to average 87.6mbd through 2008? So, unless supply grows, demand will exceed supply by a little over 1mbd throughout 2008,

And might we apply a little very basic economic science here to predict what happens to prices when demand exceeds supply consistently every day?

What about OPEC? Are they holding us to ransom?

That extra 1.2mbd in the last quarter of 2007 essentially all came from O
PEC, despite the fact that OPEC now only accounts for 42% of world supply, Since the IEA forecasts no increase at all in non-OPEC supply in 2008until the fourth quarter, if there is no increase in OPEC supply, then there will be no day in 2008 in which supply meets demand, hence continuing upward pressure on prices – and presumably at some point real market failures as the low bidders fail to acquire enough oil to do whatever they wanted it for.

On the other hand, OPEC now claims that there is no supply problem that justifies a further increase in OPEC production.

OPEC members meeting in Vienna decided to hold production levels flat, insisting oil markets are well supplied and blaming record prices on factors outside the cartel’s control, including speculators and the “mismanagement” of the US economy

The most recent OPEC Monthly Oil Market Report, February 2008 [1MB pdf], says:

The demand for OPEC crude in 2007 is expected to average 31.9 mb/d, an increase of 0.3 mb/d over the previous year. In 2008, the demand for OPEC crude is expected to average 31.5 mb/d, a decline of 0.4 mb/d.

Why do they think this? – not because they anticipate a worldwide recession:

World growth is now expected at 4.6% compared to average growth of around 5% in the last five years.

On yet another other hand:

World oil demand is forecast to grow by 1.2 mb/d in 2008 to average 86.99 mb/d, slightly down from our last MOMR estimate. (p24) [and] Non-OPEC supply is expected to average around 50.53 mb/d in 2008, an increase of 1.07 mb/d over 2007 (p28).

… so world demand is going to grow, and non-OPEC supply is going to grow less than that (and the IEA doesn’t agree it’s going to grow at all) – but OPEC demand will fall – does not compute, surely … Well, actually it’s an artifact, driven in the detail of the forecast entirely by a much larger forecast drop in demand in the second quarter, as the northern hemisphere warms, than that forecast by the IEA, combined with the assumption that the non-OPEC supply growth continues and OPEC demand is the difference between the two.

Just why is OPEC playing down the potential demand? Are they “holding us to ransom”, as is so often proposed? Why won’t they increase the production quotas? But for the last several years OPEC has done nothing at all to discipline countries that breach their production ceilings, and current and forecast OPEC crude production exceeds the current ceiling by more than 2mbd. Many observers believe that OPEC is up against a different kind of ceiling – not a quota, but actually producing all the oil their current facilities will allow, and will not be able to increase beyond this whatever happens to prices or demand. If so, the “weapon” of OPEC quotas has no ammunition left.

The US Energy Information Administration is, as usual, more optimistic, opining that OPEC has 2-3mbd of spare capacity. However, 1) that was before the 1mbd increase in OPEC production over the last five months, and 2) what remains unused is not the light sweet crude under demand, but heavy sulphurous stuff from Saudi. Even if the EIA is right, the bringing of all OPEC capacity into full flat-out production would only just satisfy demand – and any interruption of any sort by refinery accidents, hurricanes, civil disturbance, whatever, will drop total supply back below total demand, and kick the prices up again.

Not Peak Oil

Note that this is not a Peak Oil argument, but a simple extrapolation from known data for the last and next years: supply is not currently meeting demand, and won’t do for the rest of this year.

Of course, if the Peak Oilers are right, then this might be a permanent position, but we’ll restrict ourselves to the known for the rest of this piece.

What does it mean for oil prices?

Being a very classical economist for a while, the simple answer is that oil prices will continue to rise until they have suppressed demand back to match supply. Taking the various forecasts we’ve already referenced, demand is forecast to grow by 1.8% in 2008 (year-on-year growth from 86mbd to 87.6mbd), and this is from a forecast world economic growth of around 4.5%. If supply is short by 1mbd, then the oil demand increase must be ch
oked off to 0.7% (2008 average 86.6mbd) – less than 40% of the forecast increase.

The original forecasts imply an increase in energy intensity of around 2.6% – ie, the amount of GDP we get from a barrel of oil is forecast to grow by that proportion. This is higher than the historic average, but consistent with a world in which fairly major efforts are being put into energy efficiency and non-fossil-fuel energy sources. The implied supply-constrained world GDP growth is thus around 3.3% ceteris a lot of paribus.

How high would the price have to go to reduce world GDP growth by 1.2%? Big question that really needs a month or two of simulation runs on a high-grade economic model, but let’s have a SWAG (sheer wild-assed guess) at it. We’ll start by reference to one of those CNNmoney articles we quoted earlier.

When economists were predicting that oil wouldn’t negatively impact the economy, they based their assertion on a price of about $80 a barrel.

But if oil stays at $100 a barrel for the next 12 months, consumers will have shelled out an extra $100 billion on oil by next year. That’s an extra $100 billion not being spent at the mall, mega-mart or multiplex.

“The entire stimulus package could be drained by higher energy costs,” Lafakis said, referring to the $120 billion lawmakers will refund to taxpayers in an effort to keep the economy out of recession. “That has the potential to turn a mild recession into something more dark.”

So, extracting the meat from the fluff, it suggests that a $20 price increase takes $100billion out of US consumer spending on other stuff. The US total GDP is around $14trillion, so if we need to take it down by 1.2% we need to take out $168billion, which, allowing for multiplier effects on both the increase in spending on gasoline and the decrease in spending on other stuff, would be achieved by an oil price increase of somewhere in the $20 to $40 range, starting from the Dec ’07 baseline of $100.

A $20 to $40 increase in the oil price might depress the US economy by the required amount, but would it be enough for the whole world? Well, not if the US dollar continues to depreciate, as that depreciation will mitigate the demand suppression impact of the increased price on the rest of the world’s economies – to the extent that the dollar depreciates, the price increase in USD has to go higher to compensate.

The trade-weighted USD depreciation over the last six months (the period over which the oil price has risen $20) is 4% – so if that trend continues, we need to up the oil price by another 4-8% to get our 1.2% decrease in world GDP.

Voila, our SWAG for the average oil price in 2008 is between $125 and $150 – which at the low end is consistent with (though a little above) our pure short-term trend forecast for $120 in June and $145 at year end, and at the high end suggests a potential further acceleration in the price beyond even the current level – an average of $150 for the year would bring us near the $200 barrel by the year end, but that would almost certainly be much more damaging to world GDP growth than a mere 1.2% shortfall.

Price predictions, and petrol pump prices

So, I predict, with caveats all over, that the headline oil price will reach US$145 a barrel or more by the end of the year. While I’m at it, I predict that the Aussie dollar will reach (at least) parity with the US dollar over that same period.

What will this mean for petrol pump prices? After taking out 38¢ per litre in excise duty, and 10% GST (and assuming that the forthcoming budget doesn’t change these), today’s price (top of weekly cycle) of 146¢ per litre is pretty close to $1 per litre for the actual stuff. Since the $110 barrel hasn’t kicked in yet, we can make a handy parallel from the $100 barrel to the $1 + tax litre. My forecast $145 barrel (at A$1=US$1) is thus $1.35 plus tax or $1.87 per litre at year end. If the current oil price continues to go up $1 a day (NYMEX peak overnight was $111), then we might hit the $120 barrel and $1.65 litre a lot sooner. Enjoy.