Part 2 (End):
As I dug into this new data, it suddenly dawned on me why so many
people get confused at what oil and gas economics are about. It turns
out that GAAP accounting, combined with technology changes, has
created what could almost be described as meaningless financial
numbers for companies in the exploration and production of oil andgas.
Look at the picture which GAAP accounting paints for the 24 largest
U.S. energy companies in 1996 and 1997. In both years, these companies
had revenues exceeding $82 billion. More remarkably, their earnings
before interest, taxes, depreciation, depletion and amortization or
EBITDA, was an astonishing $50 billion, or 60% of exploration and
production revenues. Many analysts think of as "free cash" before any
investment decisions are made. These type returns almost make
Microsoft look like a laggard.
On a per barrel basis, the apparent cash cost to create oil is just
over $6 per barrel including overhead. Even after deducting DD&A, some
other expenses and paying a hefty $14 to $15 billion in taxes, these
24 companies, just in their exploration and production business,
generated $20 billion after-tax or almost $4 per barrel in both 1996
and 1997. These are remarkable returns and clearly indicate that
prices could come way down before these 24 lucky companies would feelany pain.
Look at the stunning difference, though, when you change from using
GAAP accounting and return to an old fashioned "cash-in, cash-out"
system. Remember the $50 billion of "free cash"? It ignores cash
expenditures of $26.5 billion in 1996 and $34 billion in 1997, which
were capitalized onto their company balance sheets. These costs
ultimately do get expensed as DD&A over a fuzzy estimated life of
various oil and gas fields, but whenever prices fall, "impairment to
value" tests cause DD&A to get written down. So, while the cost to
maintain flat production soared, the reported exploration and
production expenses using GAAP accounting methods fell.
It turns out that in 1996, which was the best year for hydrocarbon
prices in almost two decades, these 24 companies ended up with only $9
billion of net cash after all exploration and production expenditures
before paying dividends of over $16 billion to their shareholders. On
a net cash basis, these companies were deficit spending even in 1996.
In 1997, the 24 largest exploration and production companies had only
$300 million of net cash left, again before paying dividends in excess
of $16 billion. Their break-even cash cost after-taxes totaled almost
$16 per barrel.
Make one simple change to these numbers to illustrate how devastating
low oil and gas prices are, even in the full bloom of high technology.
Drop 1997 prices by 35%, to the level actually experienced in the
first quarter of 1999, and suddenly these same companies end up with
$28 billion of negative cash. The real results would not be quite this
bad as taxes would fall way off, but these 24 companies have a
negative cash flow of almost $14 billion even if no taxes are paid.
It could be argued that most of the massive funds that got capitalized
rather than expensed, were used for future oil and gas projects yet to
begin production. But, look at what these top 24 companies spent
compared to what they produced in a 1995 to 1997 time frame. In just
three years, these companies spent almost $100 billion. And, their
daily oil and gas production increased by a grand total of five-tenths
of 1%. This is exactly the same picture we saw over a decade in the
"best in class" analysis.
These numbers show in graphic detail how much money any significant
oil and gas producer must spend merely to keep production flat.
Technology did wonderful things for the oil business through the past
decade, but it also accidentally created dazzling illusions that costs
were dropping down when they were actually starting to soar.
Before leaving these "best in class" numbers, you can argue, based on
these figures, that the price of a hydrocarbon, on a barrels of oil
equivalent basis, must be in excess $20 per barrel if these
participants want to make a return on total capital equal to their
cost of capital. The idea that a clearing cost for an efficient oil
company was around $6 per barrel was simply an illusion.
So the picture becomes even more confusing. If the oil glut is
possibly an illusion, and if the real cost to produce oil is far
higher than $10 per barrel oil allows, then why did the industry
suffer through the worst collapse ever? The answer is simple. We now
live in an era of paper barrels. Oil prices are set on the floor of
the NYMEX. Whatever the paper barrel prices end up each day becomes
the de facto cash price for oil almost every day.
While few industry analysts really understand how paper barrels work,
they are now the industry's pricing mechanism, for better or for
worse. There is also a remarkable series of data points showing the
impact that the hedge and commodity funds, who actively bet on oil",
have had in influencing paper barrel oil prices.
Over the course of the past two-and-one-half years, in the few times
when the funds had a net long position in NYMEX crude contracts, oil
prices had risen, generally by a wide margin. Conversely, when they
changed their view and turn back into net short holders, oil prices
fell, often by a lot. Whether this makes any sense is an entirely
different matter. But, the data is straightforward and hard to refute.
In the world of paper barrel pricing, oil could drop to $5 per barrel.
In fact, paper barrel prices could go anywhere. The system has no
circuit breakers. There is not even the mechanism to only short an oil
contract on an up-tick, like exists in the rules of shorting common
stocks. But, paper barrel pricing only lasts until supply and demand
meet. Then, fundamentals regain the lead. Real physical supply and
demand constraints always win out over paper markets. Oil prices would
have ultimately have risen, even had OPEC not engineered this week's
production cuts, because $10 oil will have a savage effect on the
world's current oil supply.
The question this raises is what will non-OPEC supply be by the time
1999 comes to a close, regardless of what oil prices do over this same
period of time.
Remember the numbers we reviewed earlier about the massive downward
revisions to non-OPEC supply? That it had only managed to stay flat
over the past three years, despite almost every rig in the world
actively drilling? This was before the exploration and production
budgets were slashed in response to $10 oil and rigs started to get
laid down while projects were postponed or canceled. From the best
published data, the average non-OPEC producer has now cut its
exploration and production budget by 25% in 1999, with most of the cut
heavily front-end loaded. So, how will such cuts impact non-OPECsupply?
According to the IEA analysts, non-OPEC supply will actually grow this
year, though the number keeps dropping with each new Monthly Oil
Report. As of last August, the IEA estimated that 1999 supply,
measured by fourth quarter run rates year-over-year, would grow by 1.6
million barrels per day. By January, the number was cut again by half.
By March, it had been cut in half once again.
Several other analysts are now predicting, that non-OPEC supply will
actually drop, and these predictions were all prior to this week's
cuts. Goldman Sachs recently published an estimate for non-OPEC supply
to drop at least 300,000 barrels per day. Petroleum Intelligence
Weekly just updated their estimate to show a drop of almost 500,000
barrels per day. The Simmons & Company research team has the most
aggressive projected drop in print at an estimated 1 million-barrelsper day.
My worry is that even our estimated drop will be too light. There is a
chance that by year-end non-OPEC supply could drop by as much as 3.5
to 4 million barrels per day. While this is not a most likely case, it
is not impossibility, either.
I came up with such an aggressive drop by simply playing around with
the numbers. The analysis begins with a simple observation that 1998
non-OPEC supply fell by 4 to 4.5 million barrels per day over what was
originally forecast, not because of a drop in drilling but due
primarily to depletion of the existing base. Most of the new projects
embedded in the IEA's original supply forecast actually happened. The
existing production base just happened to also drop, as it did for the
best in class companies. It is getting harder and harder to maintain
flat supply and when budgets get cut, supply drops.
I come up with my "worry" forecast by merely examining the likely
production drops from three non-OPEC producers: the U.S., Canada and
the former Soviet Union. In all three cases, oil production will drop.
It is only a matter of how much. In my opinion, for the U.S. 1999 oil
supply, "best case" is a drop of only 1 million barrels per day.
Canada's best case is a drop of 250,000 barrels. And, the big wild
card is the FSU, but with almost zero funds to spend, even if prices
begin to rise, their production could easily be off by 750,000 barrels
per day. If these three happen, the balance of the non-OPEC producers
only need a drop of 7% to get to my mid-point fall in non-OPEC supply.
If you look at my worst case for the three countries, the balance of
non-OPEC producers have to only drop by 2%, which tells me that a 3.5
to 4 million drop is not a worst case.
The fall in U.S. supply is now in full gear. The numbers from the API
and the DOE keep moving around. Both have a big lag in the way most
producing states report their production data. But, both data sources
now show an estimated drop in U.S. oil supply through just the first
two-and-one-half months on 1999 totaling around 450,000 barrels perday.
When I see the rig count in a state such as Kansas, which two weeks
ago had only two rigs at work, or the state of North Dakota, which
last week had not one single rig at work, let me assure you that it
will be easy for the U.S. to lose 1 million barrels of daily oil
supply. Whether we can recover from this bout of $10 oil is an
entirely different matter.
It is easy to see why so many analysts still predict rather stable
non-OPEC supply. These estimates presume that supply only drops if
wells are intentionally shut in. No one would ever shut in a well if
the price of oil stays above lifting costs.
But, the reality of the oil and gas business is that there is no
"supply momentum." That is what the best in class data we previously
reviewed is all about. This is an industry that needs to spend
virtually all the cash that $15 oil generates, simply to maintain flatsupply.
You do not need to shut in a well to effect a drop in supply. You
merely have to experience "depletion."
While I am sure no one responsible for any individual oil and gas
field has ever made this mistake, the oil and gas industry, as a
whole, essentially forgot about depletion. Almost none of the
published oil supply estimates have any model for the decline rate of
the existing production base production, let alone how this decline
rate will likely increase. Every field ever found declines at some
point. And, when the decline begins, it generally accelerates.
Perhaps the cruelest of all oil field technology illusions was a sense
that depletion had disappeared. Some of you probably remember our Hill
Country fall board meeting three years ago, when a panel discussed
whether the Gulf could remain the backbone of the U.S. energy supply.
In this discussion, our independents highlighted how fast new gas
wells were being sucked dry and Ray Galvin of Chevron eloquently
commented that we were headed into a race between technology anddepletion.
I think we now have the evidence that depletion won.
The concept of depletion is as real as night and day, but the subtlety
of getting one's hands around what depletion rates are all about is
far trickier than first meets the eye.
Begin with what happens in the production profile of any large field.
A field could theoretically be predicted to produce at far higher
rates than ever occur. But, production is intentionally choked back to
preserve Mother Nature's natural lifting power for as long a possible.
So, for years the field creates an illusion of having flat production.
Ultimately, all fields "roll over" and begin their natural decline.
This natural decline rate, which our firm is now calling "gross
depletion", rarely shows up. Instead, the field's operator spends
considerable sums of money and uses many additional rigs and the
decline lessens. This creates "net depletion," which is always a
smaller rate of decline. But, it also requires lots of money and rigs.
Again, this is what all the money spent by the "best in class"
companies was all about.
The larger the field, the longer flat production is maintained and the
easier it is to slow decline rates. The problem is that few giant
fields are being found. In their place have come hundreds of "high
technology" boutique fields like the Tordis Field in the Norwegian
Sector of the North Sea. Here is a field that reaches peak production
of 80,000 barrels per day in the course of 30 months. It then rolls
over and begins a gross depletion rate in excess of 25% per annum.
But, the net rate is just under 20% as satellite fields come onstream
to mask the real underlying rate.
A field like Tordis is too small to choke back production. The
project's economics would be killed. So, it replicates the same
"blowout" depletion rate that the giant field would have experienced
had nothing been done to hold back production.
It turns out that technology never killed depletion. It fed the
monster and made it far more of a threat to maintain stable supply.
I have become a real believer in depletion. It is truly the most
powerful force that will impact the next decade in oil and gas. And,
as strange as it sounds, there are no reliable published estimates for
current, let alone future, worldwide decline rates.
Look, though, at the what the fury of depletion rates are all about.
The world's current oil and gas production base is just over 110
million barrels of oil equivalent per day. When I first began
preparing a graph to show how much fresh supply needs to be added to
cope with depletion, I used a 3% estimated depletion rate. Before
long, I realized that a 3% decline was far too low. So, I produced the
graph you now see which shows the impact of both a 5% rate and a 10%
rate. However, this was before I realized the difference between gross
depletion and net declines. By definition, this dotted line must be
gross, before anything is done to halt the decline.
Our firm has now studied too many basins around the world with
identifiable net decline rates at 10 to 20% per annum. So, I now
believe that a 5% rate is also too low. The real rate must be about
10% on average, or more.
Thus, to merely keep the current production base flat over the next 11
years, so we begin the year 2010 with the same amount of hydrocarbon
supply as we now enjoy, the industry must add an astonishing 83
million barrels per day. And, this has zero impact on any further
growth in demand.
What makes this task even more staggering is to examine the strange
bifurcation of the world's 110 million barrel per day production base.
It turns out that just over 70% still come from giant oil and gas
fields that have been producing for over 30 years. The average age of
these fields is probably around 40 to 45 years each. Some of them
still produce at flat rates and many have just started to roll over.
Almost all the depletion pressure seen so far is coming from the top
end of this pyramid, which includes just the fields brought onstream
since 1970. And, this base only includes three giant fields. The
balance became smaller and smaller, reaching peak production faster
and faster and then declining at ever-steeper rates.
The issues this pyramid raises are many. First, are there any giant
fields left to be found? If so, they will almost certainly be in
deepwater or the Middle East. If not, how many small fields must be
brought onstream to meet this rat race? And, how many new rigs must be
added to make all of this possible?
It seems almost certain that most of the giant fields in the bottom
70% of this pyramid will begin their declines as we move through the
next decade, but what will the composite rate become when both ends of
this pyramid are falling? And, what will the cost become to try and
stem these rising gross depletion rates? And, how many rigs and added
people are needed to make such additions more than a mere pipe dream?
These are the staggering issues that the industry will face over the
next 10 years. But, come back to 1999, the last year of the 20th
Century, which Dan Yergin so eloquently called the Century of Oil.
What if my guess that the U.S. will lose at least 1 million barrels of
daily oil supply really happens?
The pinch gets back to logistics, an issue I first discussed at an
NOIA meeting in Seattle in the summer of 1991, just 60 days before
Iraq invaded Kuwait. The issue I raised at the time was a prospect
that U.S. oil supply might continue to fall while no major new
logistics were added to our petroleum supply system. I called the
problem a potential "domestic embargo" where we create our own supply
shortages simply because oil or finished product cannot be transported
to where it is needed.
At the time, U.S. supply was still over 7 million barrels per day.
Today, we are below 6 million barrels and falling fast. We added a
tiny amount of new oil logistics, but these additions are all full.
Whether we have the capacity to substitute a fall of an added 1
million barrels per day by increased over the water imports is a big
question mark in my mind. And, count Canada out as a swing producer,
as their supply is also falling.
I would almost bet that our two landlocked refinery districts, PADD II
and IV, have the logistical test of a century as the year progresses.
So, now we have a race between the high point of the summer driving
season and a fall in domestic oil supply, with a genuine risk of
gasoline lines breaking out somewhere in these two refinery districts.
PADD's II and IV might sound like no big deal, but the territory
covers 19 important states with many roads and heavy summer traffic.
These 19 states also account for almost 30% of U.S. oil consumption.
The irony of having our first taste of petroleum shortages in 20 years
hitting home as early as this summer is the angst this would trigger
throughout America. And if it happens, everyone will immediately blame
the domestic oil industry and OPEC for creating this mess.
Sadly, both the domestic oil industry and OPEC were innocent
by-standers and were rapidly becoming insolvent as this whole mess
played out. If it happens, it is all because oil fell to $10 per
barrel. The real villains turn out to be the illusions of technology,
the "IEA's infamous Missing Barrels", the NYMEX paper barrels and
Iraq, who might well have been conducting a carefully organized
campaign to see how low oil could go and who would feel the most pain.
Since Iraq was not receiving money anyway, they turned out to be the
only players immune to the risk of $10 oil.
When this whole crisis is over, some serious post-mortem
re-engineering of the oil and gas industry must be actively debated.
I sadly give the industry, on the whole, poor marks for understanding
its own economic dynamics. Issues like the IEA's Missing Barrels were
either ignored, or worse, simply chuckled about. Too many senior
executives immortalized the concept that technology had truly created
$10 oil through talk after talk. And, almost everyone missed the power
of depletion. The real oil story was so different. Yes, we had an
oversupply, that is not even a question. But, the issue was how much,
and the facts now appear to be "tiny" by virtually any industrial
measure. Yes, observed stocks are also high, but when you examine each
stock component and what is happening to demand, most of the build was
a logistical necessity, not symptomatic of a worldwide oil glut.
Through bad data and sloppy analysis, the industry almost destroyeditself.
It turns out that there is nothing remotely normal about $10 to $12
oil. This price turned out to be a serious, life threatening risk to
almost the entire industry, while GAAP accounting procedures "mask"
most of the pain. It was one more illusion.
The old adage that "nothing corrects low oil prices like low oil
prices" is still alive. And, we will lose a significant amount of
non-OPEC supply, in addition to the highly publicized OPEC cuts. It is
just a question of how much. By year's end, we will likely test the
capacity of what OPEC has really "shut in." I bet that will also be asurprise.
Maybe at the root of these problems is the biggest mistake all of us
have made. We have been raised in a world that described oil and gas
as simply another commodity. We all know that commodities are not
particularly special and their prices always go way up and way down.
There is almost an embedded industry-wide belief that no serious
commodity ever gets by without a continuous string of booms and busts.
I have heard the oil business described as just another commodity so
often that I finally decided to look up what the definition of a
commodity is. According to the Concise Oxford Dictionary, a commodity
is one of two things. It is either an article or raw material that can
be bought and sold, in contrast to a service. Its second definition is
"a useful thing."
It turns out that most everything we buy is simply a commodity in one
form or another: Microsoft's software, personal computers, diamond
rings, and even cars. However, hydrocarbons might rank as the world's
most precious, costly and certainly most useful commodity, even
exceeding the Internet!
The industry is badly in need of re-engineering to remove these
debilitating boom and bust cycles. This last bust might have already
doomed our ability for the industry to increase daily oil and gas
supply by over 80 million barrels per day. If the industry can
recuperate, someone needs to insure this never happens again. |
