When John D Rockefeller created the vertically integrated oil company in the late 19th century, he became probably history's richest man, worth US$663 billion (Dh2.43 trillion) in today's money.
But last Thursday, the US oil giant ConocoPhillips, an offspring of his creation, decided to undo his work by splitting in two. Is this the beginning of the end for his model?
Vertically integrated oil companies have ownership in the whole chain of the business: from oilfieldsto pipelines and tankers to refineries and finally the retail outlets where motorists buy petrol, diesel and Mars bars.
In the past, the major oil companies produced their own crude oil, sent it in their own tankers to their own refineries and sold the fuel under their brand name.
Now, in nearly all cases, production, refining and retail are separate profit centres, while pipelines and tankers are generally owned and operated by independent companies. A Shell station is unlikely to sell petrol made from Shell's crude. Transparent prices along the chain eliminate the tax benefits from vertical integration.
The remaining arguments for integration are less compelling: creating bulk, smoothing earnings volatility, giving flexibility in capital allocation, and exploiting technology and trading synergies.
In the late 1990s, the pressure of low oil prices and cost-cutting drove a wave of mergers that created the modern "super-major" oil company.
Now ConocoPhillips, one of the super-majors, will divide into two pieces: the world's largest pure refiner, and the biggest non-state exploration and production (E&P) company.
The refining company will probably be worth about a quarter of a combined value of about $120bn, depending on whether it contains the pipelines, storage and chemicals businesses.
Marathon, a smaller player, did the same in January.
Three factors drove these moves. Refining, apart from a brief mid-2000s golden age, has generally been a low-margin, low-growth business. Fuel demand in developed countries is dropping; environmental legislation becoming more stringent.
Managing complex, diverse businesses dilutes management's focus. E&P demands risk-taking and the ability to strike deals with host governments. Refining is about cost-cutting and efficiency. Retail requires responsiveness to customers' needs.
Investors find the integrated companies hard to value - they combine high-margin, fast-growing exploration activities with volatile and low-margin refining, and safe but unexciting "utility" businesses such as pipelines and storage.
ConocoPhillips was valued at 5.5 times its earnings before interest, tax, depreciation and amortisation, compared with eight times for large E&P companies and six times for pure refining companies, showing the immediate benefit from a de-merger.
Since the start of 2007, the shares of pure E&P companies such as BG, Anadarko and Occidental doubled. But even though three quarters of its business is E&P, and oil prices rose substantially over the period, ConocoPhillips's stock price was essentially flat.
ConocoPhillips's aggressive acquisition policy of recent years landed it with a number of unprofitable refineries, notably on the US east coast and in Europe.
And it was the smallest of the super-majors, producing about 1.7 million barrels of oil equivalent per day, compared with Total, the next smallest with 2.4 million. It did not have the muscle of the larger companies to compete for the large and technically challenging projects offered for partnership by national oil companies. On the other hand, it was not agile enough to compete with smaller companies.
This was most clearly displayed when it was awarded a deal to develop the challenging Shah gas field in Abu Dhabi, then, after two years of negotiations withdrew, to be replaced by Occidental, less than half its size. And, unlike BP, Shell, ExxonMobil and Total, it did not win any of the big contracts in Iraq. The E&P rump of ConocoPhillips is now a highly attractive acquisition target for a super-major.
Similar break-ups of the other super-majors have been mooted, for BP in particular, both before and after its disastrous Gulf of Mexico oil spill last year. Although the independent analyst Oppenheimer & Co estimates that a split could increase ExxonMobil's value by $80bn, it seems more likely that ConocoPhillips's peers will stick with their current structure for a few more years yet, while shedding mature oilfields and poorer-quality refineries.
What about the Middle East? With a few exceptions, such as Dubai and Oman, a single state company operates oil production, refining and retail. Oil production remains highly strategic, while overseas marketing networks offer security of demand.
But is it also time for the vertically integrated national oil companies to rethink Rockefeller and stop managing rusting refineries and selling Mars bars?
Robin Mills is an energy economist based in Dubai, and the author of The Myth of the Oil Crisis and Capturing Carbon