A headline on the website of the Harvard Business Review last week might have stopped traffic on Wall Street and possibly some hearts in corporate boardrooms. It read: "Government regulation that actually works."

The Review (est. 1922) is an official publication of the Harvard Business School (est. 1908) at Harvard University (est. 1636).

The business school is arguably the most rigorous and prestigious institution in the world for training business leaders. It's not exactly a spawning ground for radical socialists. Its many notable alumni include current presidential aspirant Mitt Romney; former U.S. President George W. Bush and Henry Paulson, his Treasury secretary; Chase Carey, the deputy chairman and likely successor to Rupert Murdoch at News Corp.; Jamie Dimon, CEO of JPMorgan Chase; and John Paulson and Stephen Schwarzman, noted hedge-fund managers.

The notion that The Review might favor government regulation seems, well, counterintuitive, given that the official battle cry of contemporary American business and the politicians it finances is that regulation kills jobs, kills sales, kills salaries, kills profits and generally kills businesses.

As it turns out, The Harvard Business Review was not endorsing such a notion. It was, however, promoting a remarkable new study by business school associate professor Michael Toffel and two academic colleagues, Matthew Johnson of Boston University and David Levine of the University of California-Berkeley. Their highly technical statistical analysis was published in the May 18 issue of Science magazine (est. 1880).

Using 10 years of data from the California Division of Occupational Safety and Health, known as Cal/OSHA, the business professors constructed a study sample and a matched control group. They measured the effects of random government safety regulation inspections of workplaces in high-injury industries.

Their findings? "We now have the data to prove it," the researchers wrote. "Workplace inspections not only improve safety; they cause no discernible damage to employers' ability to stay in business and no reductions in sales or credit ratings, according to our research. Nor did we identify any effects of workplace inspections on wages, total payroll or employment."

Even more directly, they found that a regimen of random inspections to determine compliance with government safety regulations "reduced the number of injuries leading to workers' compensation claims by about 9 percent and lowered the medical expenses and wage replacement paid from those claims by 26 percent."

The academic team acknowledged remaining uncertainties and stressed that more research is needed to test if its results would hold up in other states. The advantage of conducting the initial research in California was a state law that requires Cal/OSHA to conduct some of its safety inspections on a random basis.

The more common practice is to target firms with a noticeably poor record of safety. Cal/OSHA's random data allowed researchers to create a control group for comparison purposes, giving their results much greater statistical validity.

The important point here is that good public policy needs to be based not on partisan ideologies and campaign slogans but on facts based on honest, responsible research using scientifically valid methods to investigate areas of genuine importance to American society.

Science trumps ideology every time.

REPRINTED FROM THE ST. LOUIS POST-DISPATCH