Correction to This Article
This article about the relative importance of safety and cost savings in the corporate culture at BP misstated the amount of oil released into the Gulf of Mexico since the April 20 well explosion. The spill is estimated at more than 4 million barrels (168 million gallons), not 4 million gallons.

At BP, safety vs. cost-saving

By Joel Achenbach
Washington Post Staff Writer
Saturday, October 9, 2010

METAIRIE, LA. - There's been a lot of talk about the "safety culture" at BP in the wake of the Deepwater Horizon disaster in the Gulf of Mexico. Now, federal investigators are probing what looks to them like BP's save-money culture.

At a federal hearing this week, an investigator revealed that BP's top manager on a drilling rig is given a performance evaluation that includes the category "Every Dollar Counts and Simplification."

Of 13 employee evaluations reviewed by investigators, 12 had documented ways they had saved the company large sums of money, typically six-figure amounts, and one had put together a spreadsheet showing that he could account for $490,000 in savings, said Jason Mathews, an investigator for the Bureau of Ocean Energy Management, Regulation and Enforcement, which is conducting the joint inquiry with the Coast Guard.

BP witnesses have acknowledged this week and in previous hearings that oil drilling is a business, and costs have to be taken into account. But they've said safety always comes first, and they denied compromising safety to save money.

The government this week raised the question of whether BP took incremental shortcuts because it felt rushed to finish the problem-plagued Macondo exploration well, a job costing BP about a million dollars a day. Simply leasing the Deepwater Horizon rig from Transocean was costing $525,000 a day. Eleven people died when the mile-deep well blew out April 20, and more than 4 million gallons of oil spewed into the Gulf of Mexico in the nation's worst offshore oil spill.

One document introduced into evidence this week showed that the Macondo job, originally projected as a $96 million operation, was on track to cost $58 million more than anticipated.

BP kept track of estimated completion dates for drilling jobs, and tried to figure out where the rig could go next so that it could stay active and have minimal down time. E-mails discussed by attorneys this week showed that BP hoped to squeeze in a quick job in May - plugging a well called Nile - before tackling a more complex job at another site, the Kaskida well, that might take six months.

Mathews asked John Guide, the BP well team leader for the Macondo job, whether cost-consciousness colored decisions involving safety.

"No sir," Guide answered. He added: "It is a business, so yes, there is consideration for monetary things. But like I said before, even though you could phrase it that way, safety was always the Number 1 priority. Safety was never compromised in the operation."

Also testifying this week for the first time was a key BP engineer, Gregory Walz, the engineering team leader for the Macondo well. Walz participated in some of the most critical decisions regarding the cement job, which BP's own internal investigation has cited as a major factor in the blowout.

Notes from BP's internal interview with Walz on July 29 were discussed at the hearing Thursday. Walz told the BP investigators that they were concerned about using only six centralizers to position the casing prior to the cement job, but thought they could mitigate any potential "channeling" by repositioning the centralizers and using a less-dense, nitrogen-foamed cement recipe.

Walz had a "brief discussion" with Guide about a Halliburton model showing that there could be a "severe" gas flow problem in the well if only six centralizers were used, but "they had a foamed cement slurry which was designed to protect against gas migration," the BP internal notes say.

Walz thought that 15 additional centralizers that he'd ordered flown to the rig had turned out to be the wrong type and would create potential mechanical problems in the completion of the well. According to the BP notes, Guide and Walz "did not discuss stopping and ordering the correct style of centralizer." BP's internal inquiry, released in September, said the 15 additional centralizers were the correct style after all.

As a backup - a contingency if they had signs of a bad cement job - the BP well team decided it could run a test called a cement bond log.

Walz testified that the well team followed an established "decision tree" when it came time to decide whether to run that test. When the first analysis of the cement job looked positive, including no signs of mud being lost in the well, Guide said in a conference call that he saw no reason to run the test, and Walz and other participants in the call concurred, according to Walz in his BP interview.

BP would have had to pay a contractor, Schlumberger, about $128,000 to run the test that would have taken 12 to 18 hours if there had been no problems, Guide testified. But BP decided to forgo the test and the Schlumberger workers left the rig on a helicopter the morning of April 20.

"Based on the information I had at the time, I felt we acted properly," Walz testified.

Eleven hours after the Schlumberger workers left the rig, the well erupted.

© 2010 The Washington Post Company