Did we overreact to the Boston bombings and underreact to the West, Texas, fertilizer plant explosion? That's what Richard Kim argues, characterizing the national reaction to terrorism as "total social warfare" and to industrial accidents as "callous indifference."

I agree with Kim on Boston. Terrorism remains a rare cause of death for Americans: a risk of just 1 in 20 million. Shutting down the whole city of Boston on Friday was a costly overreaction. It entailed large human costs, including lost wages and lost jobs, which probably totaled more than $100 million, and wouldn't have been worth it even if the shutdown produced a marginal increase in safety. And you only have to look at the invasion of Iraq to see our overreactions to terrorism can be far, far worse.

But the subdued reaction to the explosion in West has been largely correct. The 14 deaths there are tragic. Government investigators and the courts will sort out what went wrong, and there will probably be some useful policy recommendations on industrial safety and zoning that come out of that process.

But just as terrorism deaths are rare in the U.S., so are deaths from industrial accidents. We should be cautious about making sure that any policy changes sparked by the West explosion produce benefits that exceed their costs.

Let’s take a look at the statistics. About 4,700 Americans died in workplace-related incidents in 2010. That’s down from 6,200 in 1992, even though the number of employed Americans rose from 109 million to 130 million over that period. For an American worker, the odds of being killed on the job fell from 0.0057 percent a year in 1992 to 0.0036 percent in 2010.

As Matt Yglesias notes, this isn't an artifact of sectoral shifts away from manufacturing toward services. Manufacturing work is safer than average, and its on-the-job death rate has fallen almost by half since 1994. Construction, a relatively dangerous sector, has also gotten much safer. Sectors where safety hasn't improved include agriculture, forestry, fishing and hunting (which includes some of the most dangerous jobs in the U.S., logging and fishing), and transportation and warehousing.

The bulk of deaths are not due to "industrial accident" events of the type seen in West. In 2010, 40 percent of on-the-job deaths were due to transportation accidents, and an additional 18 percent were due to violence. America's main workplace safety problems aren't directly related to the workplace at all: They're subsets of our general problems with road safety and violent crime.

The U.S. has one of the highest rates of road fatalities as a share of population among members of the Organization for Economic Cooperation and Development, mostly because we drive a lot. Measured per million vehicles, the U.S. road death rate is only a bit above the OECD average, but that's mostly because we do much better than places like Mexico and Turkey. Even after adjusting for number of vehicles, we do worse than every country in Western Europe save Belgium, and we have about double the road death rate of the U.K. or Japan. This seems like a likely area for cost-effective safety improvements.

Murder and other violent crime, though down markedly from two decades ago, are still much more common in the U.S. than in other advanced countries. Policies to reduce crime -- from gun control to better policing to parole and probation reform to deleading -- could make Americans safer at work and at home.

And what about the other 42 percent of deaths on the job, caused by falls, contact with equipment, exposure to hazardous substances, explosions and other similar events? When you exclude transportation accidents, our rate of on-the-job deaths is about the same as that of France and Spain, despite our workplace violence problem. If we got our rate as low as the U.K.’s, meaning 0.9 fewer on-the-job deaths per 100,000 workers, we could save more than 1,000 lives a year.

The question, then, is what the cost of such policies would be. If we could achieve such safety improvements with regulations that cost $1 billion a year, that would be a good deal. But if they cost, say, $50 billion, it wouldn't be worth doing. And some of the plausible recommendations in reaction to West, such as distancing homes from facilities holding explosives, would be quite expensive -- which is why France, which set out to separate fertilizer storage from homes after a 2001 incident similar to the West explosion, has struggled to do so.

Those cost-benefit questions are worth asking. Unlike Kim, I don't think a louder popular outcry over the freak accident in West is likely to make that conversation more productive. An exploding fertilizer plant and a terrorist bombing are both somewhat like a shark attack -- likely to arouse popular panic that is out of scale with the risk actually experienced by the public.

There are common-sense measures we should take against industrial explosions, terrorism and sharks. But it's better if public attention shifts away from spectacular modes of death and toward the routine events that kill so many more people, such as obesity, handgun violence and car accidents. Far from "callous indifference," the reluctance to ascribe large policy implications to the tragedy in West represents restraint.

(Josh Barro is lead writer for the Ticker. Follow him on Twitter.)