In the wake of the disastrous oil spill in the Gulf of Mexico, BP has decided it needs new leadership. In analyzing its own efforts during this latest calamity on the Gulf Coast, the federal government might come to a similar conclusion concerning its disaster response operations.

The BP oil spill, like Hurricane Katrina five years before it, illustrates again the federal government’s inability to cope with megadisasters. In both Gulf Coast tragedies, federal agencies failed to act decisively and authoritatively, coordinate with one another and work cooperatively with state and local authorities.

“GulfGov Reports,” a research and dissemination effort led by the Rockefeller Institute in which I participated, comprehensively documented the governmental missteps in the wake of Katrina. For example, a Mississippi city tried to rebuild a highway and the infrastructure underneath but was stymied by conflicting federal requirements. One federal agency required it to give hiring preference to minorities for work on the project while another agency insisted that hiring be done on a nondiscriminatory basis. Louisiana used the bulk of its federal recovery aid to assist homeowners to rebuild in exactly the same place despite the lack of adequate assurances that they would have flood protection or that the city would provide them with water, sewage and other vital services.

More recently, as oil spilled into the Gulf, the federal government — the Army Corps of Engineers — prevented the state of Louisiana from erecting rock and sand berms to keep the oil off its shores. There may indeed have been sound scientific reasons for rejecting the particular schemes Louisiana proposed. But why didn’t the federal government initiate and fund better alternatives rather than content itself with stifling state initiative?

Actually, the Army Corps was not proactive because that is not its function. Federal agencies are geared towards dealing with routine circumstances and events, not the unpredictable and devastating effects of catastrophes. There are good reasons for such caution. Officials are duty-bound to obey the letter of the law and follow congressional directives. What disaster victims see as red tape is really a set of yellow lights prompting affected agencies to fulfill their congressionally mandated obligations, which require a full review of potential environmental impacts and strict compliance with federal labor and civil rights laws.

The key to improving federal megadisaster response and recovery is therefore not to pillory “bureaucrats” but rather to create temporary detours around bureaucratic routines in the face of truly catastrophic emergencies.

To provide such emergency response and recovery capacity, Congress should give the president standby authority to appoint an officer in charge (OIC). The OIC would coordinate the federal government’s response and recovery efforts. The OIC would be expected to ensure that the relevant federal officials and agencies were consulting one another and supporting each other’s efforts. The OIC would be empowered to end the deadlocks caused by conflicting federal regulations and to resolve conflicts among federal, state and local authorities.

But the OIC would be more than a referee and facilitator. In megadisasters, as in war, no preconceived battle plan can substitute for bold, determined, flexible and creative leadership in the field. The job of the OIC would be to mount a powerful and well-orchestrated campaign geared to contain megadisaster damage and stimulate recovery. Armed with its extensive (albeit temporary) powers and presidential mandate, the OIC would push agencies like the Army Corps to find solutions — rather than act as naysayers.

Each OIC would be appointed under a specific executive order unique to the megadisaster at hand. This is necessary because the precise limits and extent of the OIC’s authority should be tailored to the specific physical, political and economic circumstances being addressed. The person the president chooses should possess the stature, wisdom and experience that would enable the OIC to energize the national government response and, if appropriate, recommend to the president and to Congress a course of national recovery action.

To its victims, all disasters are “mega.” But the OIC scheme is only appropriate for truly extraordinary events. Unless a high megadisaster threshold is established, governors and state congressional delegations will pressure the president to declare every serious hurricane, fire or blizzard to be a megadisaster, with the expectation that this designation will produce more federal aid than could be obtained through ordinary channels. This concern is not merely hypothetical. During the Clinton years the criteria for obtaining federal disaster relief were loosened and, despite a lack of truly major incidents in this period, the number of disaster declarations and the amount of disaster relief spending rose.

It has become commonplace in the aftermath of a megadisaster to urge the president to appoint a “czar” to deal with it. Czars did not do all that well by Russia, and the U.S. should avoid them. An OIC is not a czar. The OIC’s authority would be temporary, and its discretion limited. Responsibility for funding megadisaster response and recovery would remain with Congress.

Without exceeding these un-czarlike limits, an officer in charge would still be capable of providing the strategic vision, communication among diverse participants and the tactical decisiveness that have been sorely missing in the two Gulf megadisasters.