Heck yes we need new refineries.
For a long time we didn't. There was a boom of refinery building in the '60s which sent the industry into overcapacity -- there was the ability to make more refined products than the US was consuming. Even as the US was an importer of crude oil, we were an exporter of finished products (mostly of heavy stuff like bunker oil and fuel oil, and mostly close to home, to be sure -- I don't want to make it like we were shipping gasoline to Saudi Arabia or anything).
The industry stayed balanced away from equalribium for a long time because of two factors. One was increased gas mileage. Now, most of the increase in gas mileage of the US auto fleet was quickly eaten up by more miles driven. But those were optional miles. If gas got too expensive people would drive less. So that made gas prices sticky downward. The other was "capacity creep" at the newer refineries. New technology allowed refineries to produce in excess of their "nameplate" capacity at a rate of 2-4%/year, enough to absorb all those new miles driven and still drive out of business a bunch of older refineries. Even as refinery capacity in the US has increased at 1-2%/year, the actual number of refineries has decreased by half in the past 30 years.
So far this is all good news for consumers and bad news for refiners. Throughout the 80's and 90's refiners were not making sufficient profits to justify building new ones. Indeed, refineries traded at well below their replacement value or even below their book values. A lot of the big oil companies sold out of refining partly or even entirely in the US -- their capital was better spent finding new sources of crude oil overseas or investing in plastic plants. The other good news for consumers was that pollution was decreasing because these newer refineries' expansions were cleaner than the cruddy old refineries they were putting out of business.
Now, industries always like to overexpand. Even when conditions are poor, there's always a dreamer who thinks that the market will turn, or that he has a better technology, or that he has some marketing edge or whatever. So even in this environment there were companies which wished to build new refineries. What they found is that they couldn't do it -- at any price. Refineries smell, they still pollute, and for them to be of any use they have to be surrounded by ugly and smelly sources of crude oil -- terminals, docks, pipelines, the whole bit. And they have to build new pipelines to get finished product from the refinery site to market. The one place in the United States where a builder has a chance in heck of building a new refinery is the one place where they're not needed -- the Gulf coast from Houston to New Orleans.
Now a couple of things have happened. Capacity creep has largely stopped, at least for the time being. That's mostly because refineries have spent most of their capital for the past few years on preparing for new fuel rules. Each state has its own winter and summer gas mixes, and that's expensive. New sulfur regulations are going into effect and that was expensive. It may come back in a few more years, but finished products consumption has continued to grow -- the balance is now away from customers and toward refinery profits and it will take a long time to catch up even in the best of circumstances.
Additionally, Katrina exposed a geographical imbalance. Those new refineries were mostly on the Gulf Coast. They'll be fine, once they restart. But the old refineries which went out of business were in Chicagoland and Georgia and places like that. Add in population shifts and you've also got a refinery shortage out west. Not just California, but in the four corners states. A localized problem, one which is predicted to recur with the return of the hurricane cycle, causes nationwide problems.
So yes. We need more refineries, we need them in places which haven't allowed them and we need to find a way to make that possible. That is, if market equalibrium is to be restored domestically. There are two alternatives. One is to bring in more finished product from overseas. This is expensive and annoying and increases our energy dependence on foreign sources and (if you're an environmental type) is kind of imperialistic because it basically exports US pollution to wherever the refinery is (currently, mostly places like Aruba and Trinidad) and makes it more difficult to get the correctly formulated gas to the right places at the right time. The other is to let supply and demand take care of the problem without increasing supply. In other words, to tolerate extremely high refining margins much of the time. This is not a non-starter as an issue. Consumers didn't change their behavior from $1.50 gas all the way up to $2.80. But they've shown real resistance at $3.00. It would hurt the economy and probably cause generalized inflation to have consumers making their vacation and even purchasing choices based on the price of gas, but if $3.00 indeed turns out to be the point where real price elasticity kicks in those effects can be quantified and policymakers can make decisions based on that.
