The New York Times devotes a lot of effort to bashing the Chinese Communist Party, thus it is interesting that today the Chinese leadership receives praise from the NY Times.   The backstory is the fact that President Trump is repealing President Obama's low carbon policies.  The New York Times now is hoping and saying that China will have to lead the world's low carbon agenda.  

For those interested in academic economics, I suggest that you take a look at a couple of my papers on this subject;

our 2016 book

our 2017 Journal of Economic Perspectives paper

my 2011 Energy Policy paper

My piece in The Conversation also summarizes my thinking.

The NY Times has published a great piece about Acemoglu's co-authored piece on the economics of robots.   The "headline" claim in Daron's new paper is that robots cost us jobs (substitutes not complements).   I'd like to sketch a few claims that might explain their fact;

1.   Robots don't need health insurance.  In the language of economics, to employ a worker requires fixed costs (health insurance, life insurance) and variable costs (wages and benefits).   As fixed costs of hiring workers rise, this increases the average cost for firms per hour worked and this encourages substitution from labor to capital (i.e hiring more robots).

2.   Robots don't sue their employers.  Do firms with higher K/L ratios have lower litigation bills as they face fewer lawsuits?  

3.  Robots don't go "postal" and harm fellow employees.

4.  Robots don't engage in workplace sexual harassment (i.e again, fewer lawsuits).

5.  Robots don't abuse alcohol and drugs and cause workplace accidents.  This last point should be quantified. Of course robots can malfunction, but relative to the replacement humans do they malfunction as much?  Has this been quantified.

6.  Robots don't unionize and ask for wages above their marginal product. Robots don't go on strike or engage in work slowdowns.

Facing all of these expected costs, it is impressive that firms hire people.  Do you see that the marginal product of a person must be much higher than the marginal product of a robot , given all of the expected costs from #1 to #6 above.

7.  Now, in adapting to climate change --- are robots or workers more resilient to indoor heat?  As climate change heats up factories, will this favor robotic firms or worker based firms?  This would be a nice extension of my 2016 paper with Graff-Zivin.

I am very happy to announce that the Brookings Institution has just published my new paper  titled "Protecting Urban Places and Populations from Rising Climate Risk".  Back in 2011,  David Levinson and I published "Fix it First" and that paper became well know. I'm proud to write a sequel.

Nick Bloom has published a very nice general interest piece in HBR discussing his new co-authored empirical work documenting rising inequality across firms.   Suppose I handed you Google's payroll database in the year 2005 and 2017 and I handed you similar data for Uber, McDonalds, General Motors, etc.  Nick Bloom is finding that those firms that are focused in the information technology sector are paying higher average salaries and higher salaries at each percentile of the distribution.  For example, the 90th percentile would be a lower bound for  pay for the top 10% of workers at the firm.

While I have not carefully read his academic research on this topic, this raises the usual issue of selection versus treatment. What types of workers choose to work in which industries and firms?  Does Google self select great workers and then teach them nothing? Or is there a matching process such that Google knows the profile of the men and women it wants and then shapes them to be even more productive inputs for the firm?

Now let's switch to this blog's focus.  If you think of U.S Economics Departments as "firms", has pay inequality across these firms increased?  We know that private universities pay more than public universities.  Universities are not profit maximizing firms but they do seek out price adjusted quality inputs.  There is a competitive hedonic pricing gradient for talent that is convex as the superstar economists command a high price premium.  Some ambitious private universities are known to bid more for elite talent.

Using data from the 9 UC campuses, one could study pay inequality across the campuses over the last 10 years.   I think you would see a rising return to superstar status but that the ranks of the mean pay over time would be very similar.    There are not stock options in academic economics and the production function is stable over time.  There is little need for a risk premium in academic economics because all scholars over the age of 40 are tenured with no mandatory retirement restrictions.

In the IT world, is there risk?  Can executives at Uber hedge the risk they face if the bulk of their compensation is in stock options?   Is "inequality" an ex-ante or ex-post concept?  

My serious question for Nick Bloom focuses on the nature of the underlying production function.  For the modern firm such as Facebook, what is the human capital mix of workers at the firm?  Do any high school graduates have interesting jobs at the firm or do they just get lunch and clean up?  If the superskilled are self-selecting into narrow concentrated firms, is this a pareto-optimum?  Is cross-firm inequality "bad"?  Would a progressive (and Milton Friedman) argue that it could be bad if rich firms use their $ to buy political power?

The NY Times has published a touching piece about London in the aftermath of the recent terror attack.  Several years ago, I published a piece that contrasts adaptation to terror risk versus climate change risk.  I think it is useful to restate my key points (here is an unpublished draft).

1.  The rational terrorist chooses a set of targets such that there is no predictable pattern of the attacks. In the language of time series statistics, there is no serial correlation in the attack targets. If you hijack airplanes on 9/11/01 then you hit another target at a later date.  Why?   The victim is likely to over invest and to make airports (post 9/11) too safe.  This investment costs time and $ and also leaves other alternative targets more vulnerable.   True terror is created by popping up in unexpected places.

2.  In the case of climate change,  there is a serial correlation to shocks.  The same geographic areas are hit with similar shocks over and over.  This predictability creates both learning and allows firms and households to update their evolving beliefs about these risks.  I will be talking about this at Brookings on Monday.

The spatial predictability of climate shocks make them more easy to adapt to then the "random" terror shocks.  You can think about the strategic game that the victims play against the terrorists versus against Mother Nature. In the former case, the terrorists are aware that surprise greatly helps their cause. Since Mother Nature is not a strategic opponent, she just strikes without thinking through how to cause real harm.  This predictability allows forward looking households and investors to plan accordingly and this investment under uncertainty helps us to adapt.

President Trump is helping a future generation of economists to publish natural experiment papers.  In a "natural experiment" research design (see QJE back in the 1990s), a sudden policy change provides exogenous variation that allows for a test of whether X causes Y.  For example, X may be a new government savings program and Y might be private savings.  So, if government expands Social Security, do people save less for their retirement.

With this preamble, let's turn to President Trump's new policy related to H1B visas.  This NY Times article discusses that many rural towns rely on immigrants to be their doctors.    If these immigrants are no longer allowed to work in the U.S, will small towns have no doctors?  The NY Times is worried about doctor shortages.

Spatial economics makes the prediction that rents and wages across space (think of wages and rents in San Francisco versus Detroit) adjust so that people are roughly indifferent between living in either area.  If doctors are now scarce in some rural Iowa place, then their wage will rise.  While a power couple might now want to live close to there, marginal analysis predicts that there will be some young doctor who will be willing to live in this area.

A key issue here is transportation costs.  Suppose that I'm a skilled general practitioner MD
and I locate my practice in a small town.  Suppose that there are 40,000 people living in a 40 mile radius of me.  Each of them can reach me in a 30 minute drive.   If 1% of them are sick at any point in time, then there will be plenty of demand for my services.

So, I predict that rural doctor domestic wages will rise if President Trump repeals H1B visas.  How much they rise will provide new clues about this displeasure that the skilled feel about living in small cities.  In this age of Amazon and Netflix and the Internet, this big city premium should be shrinking.  Heterogeneous tastes for city size will also keep this wage premium lower than it would need to be to attract me to live in one of these areas.

To an economist, the interesting question in the H1B visa case relates to heterogeneity in spatial preferences.  Suppose that Matt and  Rashid  (an Indian doctor) are equally good doctors but Matt demands a $35,000 annual earnings premium to be in rural Iowa while Rashid requires only a $22,000 premium (over what he would earn in Chicago) in order to lure him to rural Iowa  (keep in mind that I'm factoring out housing rents so nominal wages do not have to be higher in Iowa than in Chicago).

In this case, it is efficient for Rashid to take the Iowa job because he is "better assigned" to it.  If Rashid now cannot work in Iowa because of the new visa rules, then the people of Iowa will need to pay an extra $13,000 to attract Matt.  The  people of rural Iowa bear the incidence of this visa change.  I wonder if people understand this point?

This Houston Chronicle piece is worth reading.   The new Trump Administration budget calls for eliminating funding for the U.S Chemical Safety Board.   This agency investigates chemical spills.  In the absence of such an entity (and thus a weakened deterrence effect), will polluters take less care and devote less effort to reduce spill risk? If this occurs, will there be more chemical spills? Who will bear the brunt of this damage?   Future environmental economists have a nice "natural experiment" here (provided by the Trump Administration) for measuring the benefits of safety regulation.   A nice event study design will be used by these researchers. Canada could be used as the control group because their regulation has not changed.

My new Harvard Business Review piece is based on this recent NBER paper.

Climate change will increase the risk of temperature extremes. Induced innovation could offset some of this threat. This paper explores the demand and supply for climate adaptation innovation in a market economy. Climate change induces this innovation because the rising temperatures increase demand for self protection products and for profit firms respond to these incentives. We then augment the model to introduce climate skeptics. Such skeptics reject the claim that the world is warming and thus do not increasingly demand adaptation products. We study how the economy's rate of adaptation innovation, cross city migration, real estate pricing and the welfare of agents with rational expectations are all affected by the presence of such skeptics.

The LA Times has published a fascinating piece pointing to California as an important state where Republicans wield little political clout. In such a "progressive paradise", is all well?  Or , to repeat this again;  when progressives are in charge does the wise governance of the TV show West Wing's President Bartlet emerge?

So, there are two issues I want you to consider.  What public goods does a government seek to provide (and thus how would this product mix be affected if Republicans had some clout in California)?  Second, what is the input mix (i.e quantity of labor and capital used) to produce the desired output?  (If Republicans have some clout, would this put some discipline on California's famous labor unions to be more cost effective).

I have lived in the state for 10 years and based on my research and watching current events here is what I see.

1.  Progressive cities build too little housing by regulating developers too much.  Read my papers here and here.

2.  Bullet trains in airplane connected places are over-rated and union captured.  More generally, when California does big infrastructure and transit projects, what is the "bang per buck"?   The Boston Big Dig featured countless cost overruns and delays. In California, do construction unions take their time and pad their costs as they know that they have influence over progressive politicians who they have helped elect?  Do people actually ride the new trains that billions are spent on?  Past work in transport economics (see the work of Don Pickrell and John Kain has argued that planners over-estimate the ex-post actual ridership of these shiny new projects).

3.  Liberal cities pay their public sector workers more and thus feature higher taxes per unit of services.  Read my paper here.    Since California has unique amenities (that attract brilliant people),  government officials recognize that they face an inelastic tax base that can be taxed.   One open question is whether California's projects such as "CalTrain" significantly improves this local quality of life or whether these are Keynesian make work projects that mainly seek to attract federal cross-subsidies.

4.  While California is innovative in its environmental policy (and I mainly support this except for the consequences on #1 above),  California has been slow to introduce road pricing in its major cities.  If Republicans had more power, would there be more experimentation with road pricing and dynamic parking pricing?  (I recognize that San Fran was an early mover on implementing Don Shoup's ideas related to dynamic parking pricing).

5.  Gov. Brown has mixed feelings about the University of California.  He can't decide if he wants "great" universities or "accessible" universities.   The state can't have both.  He appears to be choosing the 2nd option.  Would under a counter-factual world where Republicans in the state have some clout, would "continued excellence" have won the day?

6.   Given California's generous redistribution policies (and wonderful climate),  how many homeless from other states and areas have moved to California?  Taking every other state's policies as given, how has California's socio-economic mix been affected by state level and local level redistribution generosity?  Has this increased local crime?  Has this affected quality of life in downtown San Fran and LA?

7.  Most California public schools are not of high quality.  If Republicans had some clout, would there be more experimentation with different educational approaches?   Does the local teachers union inhibit such experimentation?      For example, in what cases can we substitute robots for teachers? See the example of Paul Romer's Aplia.  

8.  An old question in urban economics asks;  Do cops reduce crime?  Would crime be lower across the state if Republicans had more clout and invested  in more deterrence strategies?

9.  How much higher would air and water pollution in the state be if Republicans were in charge?  If Republicans (in the name of attracting factories) gutted air and water regulations, would the factories return?  Given the high price of land and union labor, I would say "no".  Read my papers here and here.

10.  If Republicans had some local clout, fewer California cities would enact $15 minimum wage laws.  These laws will accelerate the adoption of robots, and fewer young minority workers will be hired.   The $15 minimum wage natural experiment will teach us several lessons about Econ 101 models of perfect competition.  The laws of supply and demand cannot be repealed by the power of good intentions.

If a state or city has high taxes per unit of public services, then research in economics has shown that its home prices will be lower (see Gyourko and Tracy 1991).

Two years ago, I was chatting with one of the current editors of one of the "top 5" economics journals. I sketched my climate change adaptation arguments and he grudgingly agreed with my optimism (this occurs quite often in private discussions).  He then said that a major challenge that cities will face will be disruptive storms that will lower urban productivity.  Is he right?  I don't think so.

Today's major snowstorm in the Northeast offers a natural experiment.  In this Internet connected age, does severe weather lower or raise our productivity?

point #1;  most jobs do not feature a key need for face to face contact.  Yes, a dentist must treat a patient face to face. Yes, a teacher must be face to face with students.  Yes, a starbucks worker must hand you the coffee.  Snowdays will affect their production.  But, there are many other jobs where people will be more productive on snowdays because there will be peace and quiet and because there will be no commuting on that day.  More and  more firms are allowing workers to work from home.  This conserves on commercial real estate and the firm can pay the worker less.  Team production does not need to take place in a face to face setting. Most workers are not assembling a final manufacturing product.

point #2  Any lost output on a snowy day will represent "harvesting" that can be offset on the next day when the snow melts.   Yes, if you have  a major delivery of perishable berries delivered to your store, you may not be able to sell them on a bad weather day.

point #3  Due to our physical real estate (think of shopping malls), our cities are insulated from heat and snow.  Connection to the Internet may be the ultimate adaptation technology as people have real time forecasts of when they can go outside and what spatial and temporal risks they face.   Access to Uber means that people are not "trapped in their homes".

Imagine what Uber could achieve in a region that is experiencing a flood (such as New Orleans in Katrina).  Yes , the company might engage in price gouging but this would mobilize tens of thousands of drivers to come to the rescue (the elastic supply curve).

A more interesting resilience question would study heterogeneity. If you are 10 years old, 30 years old, 50 years old 70 years old, 90 years old, how is your day affected by extreme weather? Are you as an urbanite equally productive as you would be if the weather had been average?

What quantitative metrics could be constructed so that an academic could judge this?  We can't just do a survey of people's stated valuations.

While I live in sunny LA, I do remember that the snow is inconvenient and this is a cost but moving forward in our Internet cities are such storms and nasty high frequency weather a major disruption or a day of peace and productivity?