Fire fighters in a city such as Baltimore have a better defined job than the police.  Fire fighters do not set fires.  They put out fires and save those at risk.  Fires occur at random times in random parts of the city.

The quality of the fire fighting unit can be measured using;  1. time until the unit arrives on the scene.  2. time until the fire fighters put out the fire.  3. total cost for putting out the fire.

Each member of the fire fighting unit knows his role and the individual team members know whether other complementary members of the team did their job. 

A KEY POINT.  Once the fire breaks out, this is not a strategic game.  The fire is a passive foe that makes no strategic moves.   A good applied scientist should be able to almost perfectly predict its path.  When the police try to engage in prediction, their "R2" will be much lower and this inability to predict what will happen next is the cause of many of our social problems. The police do not know the intentions of the specific person they are interacting with and turn to rules of thumb that have a much lower ex-post (R2) (i.e statistical discrimination often does not yield the right prediction).
The police have a much harder job because they interact with people (not fire).

I am guessing that after each fire, the team reviews their respective performance (like a NBA team watching game film) and figure out how to do better next time.  A good researcher could use data from #2 above to study unit specific learning by doing effects. In English, do more experienced fire fighting teams take less time to put out a similar quality fire?

Since fire fighters can die in the midst of a fire, they have strong incentives to weed out "bad apples" (those who do not know how to do their job) and to build up social capital within the unit and to build up productive human capital to fuel the unit's adaptive skill. In this case, note that social capital within the unit actually helps society because it improves the quality of the organization's ability to fight fires.   Dora and I explore group social capital in our Civil War work on desertion.

In the case of the Police,  social capital within the unit may be more likely to create an "us versus them" mentality and may help to protect "bad apples".  The police have private information about their interactions with people that even body cameras cannot capture.  The patrol partner of a given officer observes this information but does he have the right incentives to report it up the chain of command?   Do captains have an incentive to identify "bad apple" units before a tragedy occurs? Will such top down questioning affect morale of the unit? I would ask readers to skim my recent Al Pacino Serpico piece.

We know little about the social interactions within police units.  Who is the "Alpha" who leads these units?  How did this person achieve this respect level?

So, the point of this blog post is that fire fighters have a single dimension job;  put out the fire.  The fire puts society at risk and it puts the firefighters at risk.  This creates an alignment of incentives such that society and the unit are both stronger if the firefighters are individually skilled and work well as a team.

Policing raises several issues at the frontier of economics related to game theory (the strategic interaction with the people the police are interacting with), asymmetric information (the police do not know who they are interacting with, the leadership of the police do not know what the officer is doing in a given situation),  and empirical work;   what is the causal effect of policing on actually reducing urban crime and reducing civic trust?

Fire fighting does not face these questions.   Has a sociologist studied who chooses to join the police versus the fire fighting department in a given city and what are the respective effects of experience in each of these cultures on individuals as time passes?

Who self selects to be a fire fighter versus a police officer?  How does experience in these organizations shape one's values and world outlook?

In 2019, Ian Coxhead wrote a tough Journal of Economic Literature review of my co-authored 2016 book Blue Skies Over Beijing: Economic Growth and the Environment (joint with Siqi Zheng of MIT).   In this blog post, I won't comment on specifics about his review but I would like to restate our main thesis.  I hope this helps young scholars think about some open research questions.

Our main thesis is that rising educational attainment by increasingly sophisticated urbanites creates an upper-middle class who demand blue skies.  Given that China has developed later than Japan, Korea, the U.S and Europe, cleaner technology is readily available now and China is even becoming a leader in green technology both to green its home market and for future export targets.  Rising demand for blue skies in the major coastal cities creates incentives for introducing policies, upgrading energy sources, and industrial composition shifts (away from coal burning and heavy industry to cleaner service and tech and finance industries) and the net effect are much greener cities. 

Siqi and I approach the topic of urban quality of life in China from the perspective of applied urban and environmental economics.  I am not a big fan of the new generation of quantitative trade models in urban economics.  These models are too precise about writing down "structural" production functions" and transportation functions that in my opinion are not structural (and cannot survive the Lucas Critique). Given that they change as the economy shifts (consider the recent rise of Zoom), I am not comfortable with relying on them for out of sample predictions.

So, my preferred approach is to assemble a set of correlations that tell a consistent story.  Here is the story that Siqi and I tell about the rise and decline of pollution in urban China.  Dr. Coxhead keeps pushing that we are "Environmental Kuznets Curve" adherents and there is some truth to this but I hope you see a more nuanced approach below.  Our real "causal" variable is the rise of an urban educated and sophisticated population (income correlates with this).

Back in the late 1970s, Deng Xiaoping took over at the leader of China. China was poor and rural.  He was no dummy and he saw what Mao had unintentionally built.  Starting in the early 1980s, China's urban manufacturing factories (state owned enterprises) started to increase production.  This poor nation relied on coal for its energy source and urban pollution soared.  When China entered the WTO in early 2000s this further accelerated the scale of coal consumption and industrial production.  The nation's air pollution and water pollution soared and its greenhouse gas emissions sharply increased.

Given the huge count of rural people in China, a domestic passport system was introduced (the Hukoo system) to limit migration to cities by denying basic services to migrants. Migrants to a city could work but their children couldn't attend the local schools.

As China's Eastern cities grew richer and more educated, an interesting development took place among married couples with 1 child.  Such families became more interested in obtaining "blue skies" for their family.  More Educated Chinese people traveled to Japan and many studied in the United States and saw how others live.  Worried about their sole child's future, they sought blue skies for their family and safe food and clean milk.  I wouldn't load all of this on income. I would say that the rise in human capital in China's cities has fueled the environmentalism.

In the United States, cities such as Pittsburgh and New York City and Boston and even Baltimore have become much greener cities as manufactured left these cities.  In our book, we argue that the same phenomena is playing out in China's east coast cities as the price of land rises and as the SOE companies are phased out.  Such past industrial land is reclaimed and cleaned up for new residential towers.

A point that our reviewer missed is the correlation between pollution and productivity.  In a manufacturing economy based on coal, these two are positively correlated. In a skills economy (think of Seattle or San Francisco), they are negatively correlated.   My co-author Josh Graff-Zivin and Matt Neidell have a very nice JEL paper on this point.

As Charles Tiebout would predict, China is becoming a system of cities and people are voting with their feet to move to places that meet their needs.  Cities with low quality of life will experience a brain drain and in the modern skills economy, they will become poorer.

Note that this spatial equilibrium idea is nowhere embodied in the Environmental Kuznets Curve literature.  The EKC literature implicitly treats geographic places as isolated islands.  In contrast, our book is explicit about the dynamic spatial equilibrium as we study how the introduction of bullet trains across cities affects urban quality of life. We use hedonic real estate methods both across and within cities to report correlation evidence indicates the willingness to pay for environmental protection.  The relaxation of the Chinese Hukoo system (outside of Beijing) has created a system of cities that we explore in detail.  Applying the ideas of Rosen and Roback in a dynamic framework (we have a supply side of pollution while they treated spatially tied locational attributes as exogenously determined) is one of the many new features of our work.

Many are quick to note that China does not have elections.  Westerns jump to the conclusion that the powerful elites thus can ignore "the people".   We argue that both the Central and the Local Governments have strong incentives to promote the green cities agenda.  Part of our arguments can be found in our 2014 paper.   Again, the key intuition is that "green cities" attract and retain the skilled and young people acquire more human capital in cities with lower pollution levels.

In this sense, China is no different from the United States. The urban skilled demand blue skies and the elites in power have strong incentives to supply these.  America's cities that have not delivered quality of life (including my Baltimore) suffer a brain drain.  The same point holds in China.

Competition between cities creates an incentive for local politicians to supply quality of life.  This dynamic Tiebout equilibrium is a novel point that merits more research.

The reviewer critiqued us for relying on informal interviews with 2nd tier city mayors.  My view is that all evidence is evidence.   When is a data point a story and when is it data?  I would be disappointed if the reviewer was suggesting that we fudged our data.  This is not the case.  In sociology, there are active efforts to incorporate long interviews into research designs.
My colleague Robert Moffitt has a very interesting paper on this topic.

As I re-read our book, I see a relatively simple story that explains many facts about the lives of 1.4 billion people.  Those who seek pollution progress should cheer on the rise of human capital accumulation in China and the global market for technology.  Returning to Paul Romer's points about blueprints, once we have a good idea --- the Chinese will adopt it!

My 2011 New York Times piece was unpopular back then but read it now. It still makes sense today.

The reviewer is correct that our book is written for a general audience but there are many ideas embedded in the book for those seeking new research questions.  The reviewer wants a much more complicated institutional point of view that features "explicit Chinese characteristics".   But, there are tradeoffs when you add additional parameters to a model.  Siqi and I believe that a low dimensional model of supply and demand for urban pollution (with a growing urban educated upper middle class) can explain the key facts that we see in China today.

I am aware that special features of the Central Government policy introduce nuances and I have explored that in co-authored papers such as my 2015 AEJ paper and my 2018 JEEM paper.

A final thought.  As I skim through the reviewer's research output, he has not published applied work on China's cities and pollution.   He has written much more on Vietnam and the Philippines.  It would interest me if our core thesis in China's cities can explain those Asian city's pollution dynamics. Siqi and I have approached the Asian Development Bank to study this very issue. Stay tuned!

At the end of the day, Blue Skies Over Beijing is my tribute to Sherwin Rosen.  I hear his voice in my head every day and I try to make progress.  Read his 2002 AEA Presidential Address and you will see the links between my work and his ideas.   Read Ed Lazear's touching tribute to him in the National Academy of Sciences.  

Almost 20 years ago, Brian Jacob, Steve Levitt and Mark Duggan taught economists how to detect cheating teachers and cheating sumo wrestlers.    Why haven't urban economists been as successful in identifying corrupt police officers?

The answer must be that the data on their daily performance output are not available in the public domain.  In past research, I have examined the annual compensation for the Police in Baltimore, Boston and New York City.  You can read our report here.

Even with those data, the administrative data does not report the worker's age or gender.  These data provide no performance information.

Basic economics teaches us that an employer will create a job if the extra revenue a worker generates is greater than the cost of hiring her.

For a non-profit, how do we measure "extra revenue"?  When a city hires one more police officer, how much safer is the city?  This is a very difficult question to answer.

In earlier work, Steve Levitt tried to answer this by estimating cross-city crime "production functions" where he correlates a city's crime rate with the number of police hired.  Ignoring chicken and egg reverse causality issues (i.e cities with more crime hire more cops), this is a very challenging exercise because does each city have the same "crime production function"?  I don't think so but researchers have to pool observations in order to simplify the statistical inference problem.

If researchers could collect daily data on public sector worker performance, we could do a much
better job estimating person specific production functions.  The challenge would arise that those
who collect the data would know that an academic would be processing the data and this would
create an incentive to misreport the data to protect the public sector worker.

The key here is the word "public".  A private sector firm maximizes profit and the residual claimant has strong incentives to identify productive workers and to get rid of bad workers.  In the absence of strong liability incentives, does the public sector have strong accountability incentives?

Economists celebrate the "Big Data" access we have today but note that we do not have access to such public sector data. I am arguing that this is no accident.  How do quantitative researchers advocate for more accountability through more public release of productivity data while protecting worker privacy?

There would be fewer "bad apples" if there are Big Data screening techniques to identify them. Such individuals would not join government and they would act differently in the "sunshine" as the quants identify them and track them.

For those who need an Al Pacino refresher on his movie Serpico click here.

Information is highly valuable and costly to acquire.  Some of us know more about some things than others. A mother of a young child knows more about the child's health and personality than does the child's teacher.  A friend knows more about you than a stranger.