Industrial regulation and its unintended health consequences

Countries with limited regulatory capacity face a challenge when designing regulation of industrial externalities. Most industries generate multiple types of externalities, often in completely different domains. If those externalities interact, agents’ response to the regulation of a particular externality may affect the extent of other externalities. If so, “piecemeal” regulatory design may help account for the frequent and often dramatic regulatory failures we observe, especially in developing countries, where state capacity is limited.

In their BSE Working Paper (No. 809) “Firms’ Response and Unintended Health Consequences of Industrial Regulations,” Christopher Hansman, Jonas Hjort, and Gianmarco Leon explore the unintended health consequences of piecemeal regulations that ignore firm’s responses to new incentives by looking at the case of the 2009 Peruvian industrial fishing sector reform.

From aggregate quotas to individual property rights

Peru has the largest single stock fishery in the world. The industrial fishing industry has been extracting anchovies to convert into fishmeal since the 1950s, to become the second largest exporter in the world, accounting for more than 10% of the global supply and 3% of the country’s GDP. However, unregulated depletable resources face the well-known problem of the tragedy of the commons. Up until 2009, the sector was regulated using a Total Allowable Catch (TAC) system, which set the seasonal catch quota for the entire sector. In 2009, the government switched the usually recommended regulatory regime, Individual Transferable Quotas (ITQ), in which each boat was allocated property rights over a share of the global quota.

The process of granting industrial ship owners transferable individual property rights had the intended consequence of increasing fish stock sustainability, reducing the overcapacity, and increasing the sector’s productivity and profitability. However, any new regulation introduces incentives that can also affect the way firms organize production and determine which firms survive and thrive. Moreover, as in many other industries, the Peruvian fishing sector generates more than one externality. Besides using as its main input a depletable resource, it also generates air and water pollution when converting anchovies into fishmeal. If the industrial responses to new regulation are not taken into account, they could affect the other externalities of an industry (such as pollution), and have welfare implications that go beyond those intended by the initial regulation.

Any new regulation introduces incentives that can affect the way firms organize production and determine which firms survive and thrive.

Pre-reform effect of fishmeal production on health

The authors first estimate the effect of fishmeal production on adult and child health by linking various datasets from hospital admissions (for which they consider solely respiratory admissions):

  • Household surveys
  • Fishmeal production
  • Measures of air pollution

To get at the causal effect of production on health, the authors compare the health status of people living close to fishmeal plants (5km and 20 km away when considering hospital admissions), with those who live farther away, as well as people interviewed during the government imposed fishing and non-fishing seasons.

A key feature of the identification strategy exploits both spatial and temporal differences, allowing the authors to isolate the effects of specific seasons or locations on health from that of fishmeal production.

The authors find that there is significant worsening in the health of adults and children when fishmeal plants are in operation. The patterns found hold for self-reported health problems, medical expenditures, and hospital respiratory admissions. The effects found are similar regardless of whether they consider production in the previous 30 or 90 days, using either the amount of fishmeal produced or the number of production days of the facilities. For example, hospitals near fishmeal producing plants, during the fishing season, can expect 2 more admissions for every 10 additional days of production (or when a 50% increase in fishmeal production occurs); likewise, medical expenditures can be expected to increase by 4.6%. Incidence of cough or other respiratory diseases among children (less than 5 years old) rise by 2% as a response to fishmeal production occurring in the past 90 days in a plant close by.

What drives these health effects?

The health effects observed in the data are not related to changes in the labor markets that happen during the fishing season, nor are they specific to workers in the fishing industry. Using air quality measures from 5 stations in the capital city of Lima, the authors show that the effect of fishmeal production on health is due to air pollution (and mostly unrelated to water pollution). Particularly, during the production of fishmeal, plants release large quantitates of pollutants such as NO2, SO2, and organic particles, which have been shown in other contexts to cause detrimental health effects.

Specifically, a standard deviation increase in the emissions in any one of the pollutants in the closest port of Callao causes a roughly 2.5% increase in self-reported health issues. These findings are roughly in line with epidemiological literature, even though those findings are correlational while the techniques the authors use here lead to casual findings.

Post-reform health effects: duration vs. intensity of exposure

Although the 2009 reform had the intended consequences of improving fish stocks to maintain the future of the industry without harming (and in fact increasing) its profits, the authors find that the reform significantly worsened the health effects of the industry on the population living nearby. The reform caused an increase in hospital admissions of 7.3% (adding 12.2 more admissions per month per hospital), self-reported health problems increased by over 10%, and medical expenditures by 24%. Notably, the data shows that the large negative effects of the reform are linked to the increase in the number of days of production, rather than the total fishmeal production, which remained more or less constant before and after the reform.

The figure below shows the regression trend lines pre- and post-reform, where the vertical solid line indicates the reform. As can be seen there is a stark upward shift post reform for incidences of respiratory admissions, self-reported health issues, and medical expenses for those living near fishmeal plants (compared to those living farther away).

Figure 1. Plotting Health Outcomes Across Time Pre- and Post-Reform

In the paper, the authors develop a theoretical model that shows how firms respond to the incentives introduced by the ITQ regulation. Under the TAC regime, as soon as the fishing season starts, boats have incentives to “race” for the fish. Given that fishmeal resulting from fresh fish has a higher price in the international market, the concentration of the fishing activity at the beginning of the season means that most of the production (and pollution) is concentrated in a short period of time. Additionally, the excess supply of input earlier in the season drives down the price of fish, allowing firms with higher cost of production to profitably operate. The introduction of the ITQs spreads out production in time. Fish prices increase, and thus less productive firms are led out of the market.

A key prediction of the model is that the time profile of fishmeal production (and its resulting pollution) dramatically changes with the reform, going from a concentration of large amounts of production early in the season, to the same amount being produced over a longer time period. Further, this change in the time profile of production should be starker in ports that concentrate more productive firms, compared to those where high cost firms are located.

Consistent with epidemiological studies, the authors hypothesize that sustained low intensity pollution can be worse for health than short periods of intense pollution. The data is consistent with this hypothesis. The data analysis shows that the negative health effects of the reform are related to the changes in the time profile of production. Further, the results are more stark for people living in the northern region of Peru, where the new regulatory frame had a larger effect spreading production in time, whereas in the previously unregulated southern region, where the regulation reduced the number of days of production, the reform effect was zero or even positive, improving health in some locations. Likewise, in more efficient ports, the negative effect of the reform was larger than in ports that concentrate low productivity firms.

Overall, the authors conclude that the 2009 introduction of individual property rights over fish, meant to reduce overextraction and overcapacity, dramatically exacerbated the impact on health of air pollution from the plants that manufacture fishmeal. Further, the adverse health impact arose because a given amount of air pollution is worse for health if dispersed over time. This new evidence on the shape of the health production function has far-reaching implications on how to regulate polluting industries. These results speak to the common policy debate on intensity versus extension of production.

This new evidence on the shape of the health production function has far-reaching implications on how to regulate polluting industries.

Was the reform worth it?

It seems that the main problem with the Peruvian ITQ reform of 2009 is that dispersion of air pollution over time that led to worse health outcomes was either not anticipated or, if it was, it was scrapped altogether to sell the reform. Though the reform presumably ensured the sustainability of fish stocks and increased the profits of the industry, it came at a cost of hurting the health status of the population.

So overall, did the reform increase or reduce society’s welfare? Using publicly available financial information from firms operating in this market, the authors estimate that sector profits one year after the reform rose by USD 219 million. On the other hand, the reform caused 55,000 additional hospital admissions for respiratory diseases, which, when added to the higher medical expenditures reported, give a whopping cost of USD 343 million, indicating that the unintended consequence of the increased duration of air pollution exposure surpassed the benefit of the reform. While some strong assumptions are made in these calculations, the authors note that they may be underestimating the true health impact, since they consider only health issues that lead to a respiratory hospital admission and ignore effects on mortality.

How do these findings impact regulation design?

The results highlight the shortcomings of piecemeal regulatory design in the presence of multiple externalities. While commonly advised regulation aimed at internalizing one externality can increase welfare, the social benefits of these regulations can be offset by their effects on other simultaneous externalities.

Ignoring how incentives reorganize production and the effect on other externalities is the main risk in implementing piecemeal regulation.

The results highlight that the common policy trade-off between duration and intensity of exposure to air pollution can be of first order importance. In the particular case of industries that rely on extraction of a common pool resource processed into a final product by polluting factories, policymakers face an additional trade-off. Ignoring how incentives reorganize production and the effect on other externalities is the main risk in implementing piecemeal regulation. Careful planning and consideration should be taken at the design stage of any piece of regulation.