Persistently hazardous air quality suggests that Pakistan lacks a framework to mitigate pollution. However, Pakistan has a rich history of environmental legislation — successive governments have instituted departments, laws, and mandates over time to manage air quality. For example, national and provincial legislative acts establish environmental protection agencies, delineate the agencies’ governance structures and responsibilities, and mandate air quality and pollutant-specific standards. Despite such frameworks, Pakistan has struggled to clean its air.
In a recent working paper titled “Charting Pakistan’s air quality policy landscape” and supported by the International Growth Center (IGC), my co-authors — Ali Habib (HIMA^Verte) and Amna Shahab (ECOlogical) — and I deconstruct the following question: Why has Pakistan — and Punjab in particular — continued to suffer acute air pollution over the past several years despite strong regulations on paper and vociferous citizen uproar?
We demonstrate how regulatory deficiencies, resource and capacity constraints, and imperfect information prevent environmental institutions and regulations from achieving their objectives. In a nutshell, air quality outcomes will continue to stagnate when authorities keep operating in a weak institutional setting, regardless of environmental legislation.
1: The air pollution problem
What does it cost?
Air quality in Pakistan’s major urban centers has considerably deteriorated over the past five years. Lahore and other cities regularly feature in the list of the most polluted cities globally.
Figure 1: Lahore’s daily average PM2.5 concentration (data from the US Consulate’s AirNow monitor). Note: I thank Mahnoor Kashif for generating this graph.
Figure 1 depicts the daily trend of Lahore’s average particulate matter 2.5 (PM2.5) concentration from October 1, 2019 to October 21, 2021. PM2.5 constitutes fine particles — with a size less than three per cent the diameter of a strand of human hair — which the blood stream absorbs when inhaled, gravely risking health.
Lahore’s daily average PM2.5 concentration exceeded the World Health Organisation’s and Punjab Environmental Protection Department’s standards on most days in this period — even on days when Covid-19 led to lockdowns. In winter, Lahoris experienced PM2.5 levels up to 13 times the EPD’s maximum permissible limit.
This year — at the cusp of the smog season — Lahore’s annual average PM2.5 levels stand at 10 times the WHO’s standard and seven times the EPD’s standard. Poor air quality is not just a winter phenomenon — we experience poor quality air throughout the year.
These statistics reveal that Lahoris bear considerable costs of poor air quality. Direct costs include infant mortality, debilitating diseases (cancer, asthma, and bronchitis); and acute morbidities (obesity, mental illness, and cognitive dysfunction.
Indirect costs include lower labour supply and productivity, higher incidence of violent crime, migration of skilled labour, and volatility in the financial sector.
Figure 2: Potential PM2.5 reduction and its impact on life expectancy (EPIC 2020).
Owing to these costs, humans exposed to PM2.5 live shorter lives. Figure 2 — a snapshot of Pakistan’s Air Quality Life Index (AQLI) — reveals that at current PM2.5 levels, an average Pakistani loses 2.7 years off their life, while an average Lahori loses 5.3 years of their life.
What are the main sources?
We lack data and research on pollution sources and their contribution to overall emissions — studies that determine these statistics are called source apportionment studies. In 2018, the Food and Agriculture Organisation (FAO) conducted perhaps the only comprehensive source apportionment study in Pakistan. It shows that the main polluting sectors include transport (43 per cent share in total emissions), industry (25 per cent), agriculture (20 per cent), and power (12 per cent).
Older vehicles and two-stroke motorcycles and autorickshaws explain a large share of vehicular emissions. These vehicles burn fuel inefficiently, generating considerably higher emissions. The number of motorcycles in Pakistan has risen considerably over the years, comprising 74 per cent of the total number of registered vehicles.
Poor fuel quality exacerbates the problem. Pakistan’s fuel quality falls under the European Union’s outdated Euro 2 standard — years behind the Euro 6 standard adopted by many high-income countries. The federal government’s signals to switch to Euro 5 compliant fuel have stuttered.
Pollution ferociously sprawls across Punjab in the winter. Two phenomena explain this: 1) thermal inversion; and 2) crop residue burning.
Thermal inversion — a meteorological phenomenon — traps particulate matter and other pollutants, which mix with condensed water vapour to form smog.
Just before the onset of winter, farmers in both Pakistani and Indian Punjab begin burning rice stubble — left on their fields after the fall harvest — to prepare land for sowing wheat. Farmers find this cheaper than hiring labour. Crop burning amplifies emissions, leading to prolonged smog.
Do emissions flow from India to Pakistan? Transboundary pollution depends on prevailing wind patterns and direction. Evidence suggests that fire-related pollution primarily flows from Pakistan into India but could reverse direction. Determining how pollution in each country affects the other requires models supported by fine-grained data — which we do not have.
2: Regulatory deficiencies
Inactive environment protection councils: Federal and provincial Environment Protection Councils serve as independent oversight bodies, tasked with monitoring and enforcing the EPDs’ rules and regulations. These councils have either held far fewer meetings than mandated or not met at all in the last few years — a foundational failure of Pakistan’s air quality management.
Figure 3: Environmental Protection Councils: Non-government members and meetings.
Shortcomings of the command-and-control approach: Provincial governments in Pakistan manage air pollution through what economists label a command-and-control (CAC) approach. This approach entails mandating various standards through law and then enforcing them through state machinery — inspectors, courts, fines, and threats of shutdown. For example, the Punjab Environmental Quality Standards — monitored and enforced by the Punjab EPD — establish standards in seven domains, including ambient air quality, industrial gaseous emissions, and motor vehicle exhausts and noise. Note: Punjab established rules to measure and levy pollution charges on sources several years ago but has never enforced them. The CAC approaches are impossible to implement when regulators with limited resources must keep tabs on millions of polluters — including motor vehicles.
Inadequate environmental quality standards: Implementing standards require several departments to coordinate and share responsibilities. Provinces have not delineated the extent to which the EPDs can inspect, penalise, and advise factories, motorists, and farmers — which traditionally fall under the Industries, Transport, and Agriculture Departments.
Setting emission standards for industries and vehicles does not imply that ambient air quality will improve. Linking source emissions to ambient air quality requires scientific models which take meteorological factors (temperature, humidity, and windspeed) as inputs. The EPDs have not publicly disclosed how they linked emission and ambient standards. Vehicles can deteriorate air quality even if they meet the existing standards. For new vehicles, the EPDs limit either emissions per kilometer or emissions per unit of fuel consumed — not the number of vehicles on roads nor the number of kilometers each vehicle can cover. As more vehicles enter roads and cover greater distances, air quality worsens.
Uniform standards are inefficient: The EPDs apply the same emission standards to most sources. This substantially raises the cost of abating emissions across all sources. Across industries, firms produce different outputs, while within industries, firms often employ different technologies. The costs of reducing emissions will differ greatly across firms if they must cut emissions till the same level. Reallocating emissions from firms that incur relatively higher costs of reducing emissions to firms that can cut emissions more cheaply can save significant costs. The only way to reduce abatement costs through CAC is to set source-specific standards. But this requires knowing what costs each source will incur for cutting emissions, which the EPD does not because of information asymmetries — sources know far more about their technologies and costs than the regulator.
Lack of frameworks to adopt technology standards: As part of their mandates, the EPDs must facilitate polluters to adopt cleaner technologies. They do this through technology standards. The recent move by the Punjab government to convert existing Bull’s Trench Kilns to Induced-Draft Zigzag Kilns offers an example. Zigzag kilns pollute less compared to traditional Bull’s Trench kilns, but other technologies (Vertical Shaft kilns) pollute even less. Mandating an absolute technology standard takes away kiln owners’ flexibility to adopt better technologies beyond Zigzags. They also erode incentives to adopt other measures to limit emissions. Note: the Punjab government labelled Zigzag kilns “environmentally friendly”. This label is false — Zigzag kilns are cleaner but by no means clean.
Budgetary constraints: Given the large number of emission sources and their incentives to ignore official directives, the EPDs require considerable outlays to monitor and enforce their standards. But provincial governments allocate insufficient funds to EPDs. For example, the Punjab government allocated only 1.8 per cent of its development budget to environmental protection — not just air quality — in 2020. Paltry budgets weaken the EPDs’ abilities to carry out their regulatory functions.
Questionable air quality data: The Punjab EPD has gradually begun to publicly report air quality data. But these data and their reporting are fraught with serious problems. The EPD lists the data only on its websites, putting it out of reach of the digitally illiterate and those without digital access.
The EPD operates six air quality monitors, which cover only Lahore, though several other cities in Punjab also experience poor air quality. It also struggles to consistently report data with large chunks of daily readings missing. This underreporting commonly stems from malfunctioning equipment, with many monitors going offline because of power outages or expired internet packages.
Figure 4: Differences in the Air Quality Index advisories.
Air Quality Index does not conform to international standards: The Punjab EPD uses data from its air quality monitors to construct and report an Air Quality Index (AQI) — a weighted average of various pollutants. But this AQI deviates from internationally accepted standards — it omits a safety margin, which environmental agencies often include in their indices. As Figure 4 shows, the Punjab EPD labels AQI values between 301 and 400 as “poor” and values above 500 as “severe” (highest category). On the other hand, the United States Environmental Protection Agency (USEPA) labels any value above 301 as “hazardous”. It is concerning to see that what the USEPA considers “very unhealthy”, the Punjab EPD considers “moderately polluted”.
No source apportionment: Without source apportionment studies (described earlier) across space and time, regulators cannot understand how different sources affect air quality. But so far, we only have one comprehensive study — by the FAO in 2018 — to inform policy. This study relies on imprecise measures and is already outdated.
3: Filling policy gaps
Better measures of economic and social costs: We require more robust work on quantifying the full costs of long-term exposure to air pollution. The work must not only focus on mortality, morbidity, and cognition but must also cover pollution’s impact on behavioural decisions, such as fertility, migration, time-use, and defensive expenditures. Such evidence on impact will underscore air pollution’s potential costs and motivate policymakers to act.
The government can overestimate growth figures if it ignores the effects of worsening air quality on growth. Projecting the flow and stock of emissions and quantifying their effect on growth will allow policymakers to buffer the adverse shocks of development policies.
Data on source apportionment: We do not know enough about the spatial and temporal dynamics of emission sources and require continuous research to understand these. The apportionment studies must employ established techniques involving lab testing and diffusion modeling to generate precise and accurate data. Identifying sources and their industrial composition, location, and contribution to overall emissions will enable policymakers to better target polluters, revise existing emission standards, and set clearer objectives.
Understanding willingness to pay: We currently know little about how much people are willing to pay for improved air quality and how this willingness to pay changes with information. Methods that involve measuring the demand for pollution avoidance — particulate filtering masks, air purifiers, and air quality information such as forecasts and real-time readings — offer a way to approach these questions. Ongoing work by my co-authors and me, which experimentally measures the willingness to pay for pollution forecasts and particulate filtering masks in Lahore, will fill a part of this gap.
Explore incentive-based strategies: Pakistan should explore options to transition from command-and-control regulations to incentive-based abatement strategies such as emission charges and tradable permits. These strategies incentivise polluters to abate cost-effectively and provide them greater flexibility in determining the best abatement measures.
India’s Gujarat state is currently piloting an emissions trading programme for particulate matter. The results of this programme could inform the design of Pakistan’s own permit trading system. Note: for details on how emissions charges and tradable permit work, I direct the reader to the paper which this piece summarises.
Informal regulator as an alternate channel: Amid regulatory failures, alternative stakeholders (“informal regulator”) offer a complementary channel to pressure polluters to comply with standards and governments to improve monitoring and enforcement.
Leveraging informal regulators such as civil society organisations, academic and research institutions, and industrial associations can create more transparency in air quality data sharing, signal a sense of ownership in the regulatory process, and motivate voluntary initiatives.