Oil continues to spill into the Gulf of Mexico at a rate of 800,000 liters a day from the BP oil rig that sank in late April.  BP Chief Executive Tony Hayward described his company's efforts to seal the oil leak to the BBC this morning, and he said it was not possible to estimate how much BP will have to spend to contain the leak.  

The first concern--for federal and regional government officials at least, if not for most people--is on the potential damage to the environment and the economic impact of that damage, rather than the cost to BP.  Still, the question of cost for BP is an important and interesting economic exercise, at the very least.  Ben Fissel, a PhD candidate at the University of California, San Diego, has taken a stab at analyzing the economic impact of the oil spill on BP for Econbrowser.  Fissel evaluated the change in stock value to measure what the market anticipates to be the long term costs:

When an event, such as this oil spill, impacts a company it will also impact its long run profitability. The divergence of the stock price from what we would have expected had the event never happened is a measure of the net present value of the cost incurred by the oil spill. Event study analysis gives us a framework to answer just this question.

BP prices since Jan. 1^st have been plotted below in Figures 1 and 2. The black line gives the real prices and the red line shows the model estimate of what would have happened if the spill had not occurred. Table 1 lists the prices and returns over the event window for both the real time series and the estimate. Event studies use other factors in the market to estimate what BP's stock price would have been. A list of these factors can be found on Table 4. The event window spans 7 trading days April 26^th - May 4^th and is centered on April 29^th. A 10-day event window buffer was used to separate the estimated model from the event window. The 250 trading days prior to the event window and buffer were used to estimate the model.

Figure 1

Figure 2

Read Fissel's full post at Econbrowser, here.