The consequences of Lean at BP

In his new book on BP, Tom Bergin blames lean management principles for the Deepwater Horizon disaster. Here is a summary of Bergin's argument, taken from a review by Mattathias Schwartz, LRB 6 October 2011.

The beginnings of the Deepwater Horizon disaster, Bergin argues, can be found in the reorganisation Browne undertook, applying to BP the leaner management principles he learned at Stanford. The company was divided into ‘strategic business units’, independent companies within the company, each of which could allocate its capital and manage projects as it saw fit. Managers were held to short-term ‘performance contracts’ focusing on high production and low cost. Those who could extract the most oil while spending the least money were rewarded with promotions and bonuses. Promising junior executives were shuffled between posts all over the world, rarely staying anywhere long enough to bother replacing outdated equipment or rusting pipelines. ‘Go to the limit,’ Browne told his managers. ‘If we go too far, we can always pull back later.’

Bergin argues persuasively that such practices amounted to ‘moral hazard’, with BP not quite consciously rewarding the senior employees who engaged in the riskiest behaviour. The cost-cutting continued under Hayward, who trimmed BP of drillers, geologists and other specialists, outsourcing technical tasks to contractors and filling the company’s top ranks with traders who knew how to allocate capital and whip subordinates into meeting the next quarter’s targets. The demands for rapid production and low cost grew even more intense as Hayward instituted ‘stretch targets’ whereby the results achieved by one outperforming business unit were touted as company-wide goals.

Much the same sort of thing has been going on elsewhere, in manufacturing and retail in particular, since the late 1990s, when a new wave of Taylorism swept through management theory. Under the banner of euphemisms like ‘accountability’, workers’ earnings and job security were linked to ever rising performance goals. For a retailer like Wal-Mart, there were few upper limits on efficiency targets – impossible goals could be passed down the chain of command until ambitious managers felt compelled to lock their minimum-wage employees in stores overnight. But oil and gas extraction were a special case. At the bottom of the production chain were the implacable realities of geology, whose limits could not safely be breached. ‘Thus began a continuous effort to go beyond what BP’s own engineers considered physically possible,’ Bergin says of the stretch targets. One of the most important measurements was raw speed – how fast project leaders could get a hole drilled – calculated in ‘days per 10,000 feet of drilling’. It was as though BP’s senior executives in London had sent their workers into a room full of flammable gasoline vapours with a box of matches and a live chicken, offered prizes to whoever could produce a cooked chicken fastest, then handed the workers safety manuals, closed the door and turned their backs.

Mattathias Schwartz, LRB 6 October 2011 
reviewing

• Spills and Spin: The Inside Story of BP by Tom Bergin 
• A Hole at the Bottom of the Sea: The Race to Kill the BP Oil Gusher by Joel Achenbach

See also my post Black Swans and Complex System Failure.

Market Forces and Effort

In a rational world, we might expect people to work harder when there is more demand for their labour, and take a rest when there is less demand. A recent study focused on taxi drivers because they have considerable autonomy over the number of hours they work. (Many other workers have rather less autonomy, and so their preferences are less easy to discern from their behaviour.)

We may suppose there is more demand for taxis when it is raining. Although taxi fares are fixed, economic theory still predicts some rebalancing of supply and demand. The total cost of a taxi ride to the passenger is higher when it is raining. This is because it takes longer to find a taxi, and it takes longer for the taxi to reach your destination (because of other traffic). The taxi-driver earns more money in an hour, because the cab is occupied most of the time. (On a sunny day he might drive around empty while people prefer to walk.) We should expect supply to go up, as taxi drivers work longer hours to take advantage of the higher demand. We might also expect taxi drivers to find leisure a more attractive option when the weather is good.

But this is not how taxi drivers behave. Many set themselves a daily revenue target, and they work until they have reached this target. On a rainy day, they reach this target more quickly and therefore go home earlier. They actually work longer hours on sunny days when business is slow.

source: BBC Radio Four
The Economy on the Couch
(December 2nd, 2004)

We can draw several important lessons from this.

1. People's behaviour cannot always be explained by economic theory. (Empirical observation of people and organizations may produce surprising results.)


2. In particular, economic incentives to work harder are not always effective. (Arguments for higher remuneration are almost always based on moral persuasion or political power rather than rational economics. This applies equally to executive pay and to shopfloor pay.)
   
   
3. The behaviour of a system may depend critically on the policies by which people choose to govern their behaviour. (In our example, the daily revenue target acts as a personal policy for each driver.)