Industrial accidents can be devastating to the people and communities involved. In addition to the immediate loss of life, such accidents can leave a lasting mark. For example, residents and victims of the 1984 Bhopal gas leak incident in India still live with the environmental damage created by Union Carbide, and protest yearly to publicize their continuing efforts to obtain financial compensation for damages as well as justice for the remaining perpetrators. Scientists are still trying to determine the long-term effects of the 2010 BP oil spill in the Gulf of Mexico.

However, little systematic research has explored the underlying causes of such accidents. In a recent study, we examine one crucial, yet overlooked, factor: Economic globalization.

We thought it was likely that two interrelated aspects of globalization – growing international economic interactions and the state policies that facilitate them – are likely to increase the probability of industrial disasters. Regarding the former, the increased complexity of global supply chains could increase the likelihood of industrial accidents due to such factors as technological differences between home and host countries or culturally divergent perceptions of risk and workplace safety. On the latter, firms might intentionally seek out countries with less enforcement of safety regulations or choose to move particularly dangerous industries, such as ship-breaking or toxic-waste removal, abroad (known as the “race to the bottom” dynamic).

To test these linkages, we examine the possible impact of two different measures of globalization – one focusing on economic interactions, the other focused on policies meant to encourage these interactions – on the probability of major industrial accidents across 137 countries over 42 years (1971-2012). The economic interactions variable captures cross-border economic interactions that include trade (imports and exports), foreign direct investment, and portfolio investment. The economic policies variable covers state policies that restrict or facilitate more global economic openness, including tariff rates, non-tariff barriers, taxes on international trade, and capital account restrictions. Both the interactions and policies variables are coded on a 100-point scale, with higher cores indicating more economic flows and fewer restrictions, respectively. We gathered the globalization data from the KOF, a Swiss research institute that specializes in applied economics. Our major industrial accidents data comprise workplace accidents that result in at least 10 reported fatalities and/or at least 100 people reported injured or requiring immediate assistance. The data are drawn from the Emergency Events Database (EM-DAT), which provides a comprehensive listing of natural and man-made disasters since 1971.

Statistical results suggest that economic globalization increases the likelihood of industrial accidents. The findings are robust even when we control for several other major predictors of industrial accidents such as economic wealth, the share of major industries in the total economic output, bureaucratic quality, and political corruption. How big is the impact of economic globalization on the occurrence of industrial accident? Our data provides some picture of the strength of this effect. For example, the mean score of the economic policies variable in our sample is 47 (on our 1-100 scale). When we increase it by one standard deviation (from 47 to 70), the probability of an industrial accident goes up by about 54%. Similar shifts in our economic interactions measure raise the probability of an industrial accident by about 14%.

While these shifts in indices are a bit of an abstraction, they demonstrate that even after we account for other relevant factors, globalization has a substantial independent impact in increasing the probability of major industrial accidents. These findings corroborate various case-study evidence that cites the role of economic globalization in major accidents – for example, the Rana Plaza disaster in Bangladesh, which resulted in over 1,000 worker deaths, contained five garment factories that were connected to multiple global supply chains, and some attributed the disaster to the “greed and pressures of globalization.” One of the primary factors underlying the Bhopal disaster was the ability of Union Carbide to exploit lower safety regulations in India in order to lower costs.

In terms of implications, our study reveals a major conflict between worker safety and globalization, as a policy commitment to economic openness may come at the cost of more industrial accidents.

Governments play a key role here, as it is the state that is responsible for the relevant labor and safety regulations, as well as enforcement of other environmental and building codes which should ostensibly reduce the probability of industrial accidents. However, state officials have a natural tendency to underinvest in accident prevention, as politicians are more inclined to enact policies with short-term benefits even if such benefits have larger costs in the long run.  While major accidents impose political “costs” on state officials – particularly if they bear the blame for the occurrence – potential costs are discounted because their occurrence is a future probability rather than a present certainty.

Moreover, it is possible that blame for such accidents, should they occur, could be borne by future administrations. Thus, though such investment may be cost effective, its political benefits are elusive. As former U.S. Congressman Barney Frank once noted, “No one will ever get reelected for avoiding a crisis.”

To a certain extent, the dilemma we uncover is similar to the dilemmas that surround all sorts of other long-term public goods such as infrastructure, preparation for natural disasters, or environmental protection. Specifically, it cannot be simply assumed that either market rationality or political interests will necessarily result in improved worker safety. Firms and businesses cannot be left to their own devices to ensure a safer workplace. Globalizing states face a major governance challenge in this area, as the short-term dictates of both the marketplace and the political arena may further reduce the willingness of states to take the steps necessary to lessen the likelihood of industrial accidents.

Concerted efforts at multiple levels – including states, firms, international organizations such as the ILO, as well as NGOs and civil society groups – might be necessary to reduce the occurrence of these tragic, and ultimately preventable, disasters.