Thursday, May 21, 2015

On the Road

Today, I'm at Drexel University speaking at this sustainability event  and Tuesday I return to the LSE to talk about my China research.     Yesterday, I learned a good lesson about Uber. You never know who you are going to meet.  A Hollywood Movie Star drove me to the airport.  Those of you who are fans of the Texas Chain Saw Massacre may remember LeatherFace.  

Tuesday, May 19, 2015

Does a Large Increase in the Minimum Wage Cause Unemployment? A Natural Experiment

The City of Los Angeles is raising its minimum wage to $15 an hour.    Do David Neumark and David Card agree on the consequences of this well intended policy?  Neumark's papers on this subject are listed here.   Card's research on this subject is available here.   The current minimum wage is $9 so a 66% increase is not "marginal".  How will partial equilibrium modelers evaluate the effects of a policy that we have never seen before?  Will structural modelers claim that they can predict the consequences?

Are micro economists useful in predicting the consequences of new policies?  The new $15 minimum wage poses a test of our value and it poses a test of standard neo-classical economics.

A first step a researcher would need to grapple with is to be explicit about which jobs in the City currently pay less than $15 per hour.  For each of these jobs, will employers hire less labor when they must pay $15 per hour?   If employers must pay for health benefits also, some bosses may choose to hire fewer workers and ask them to work longer shifts (to hold total hours supplied constant).  This would reduce the fixed cost of paying health benefits because fewer workers would be hired.

Will firms now take fewer chances hiring workers who appear to be marginal in terms of cognitive and non-cognitive skills?  Recall that a profit maximizing firm hires until the marginal value of hiring a worker equals the marginal cost. If the marginal cost is now higher because of the minimum wage then firms will screen out the least productive low "marginal value" workers.  This will effect people who have been in prison and other groups who are viewed as potentially low productivity.

Will firms stop reporting taxes and create more informal jobs whose payments are in cash and are not observed by the tax system?  The minimum wage can't bind if there is "no job" because it has vanished into the black market.

Will progressive economists argue that the minimum wage will pay for itself due to efficiency wage theory that workers who are not better paid will exert more effort and this will raise their marginal product?

For small businesses whose profit margins are cut, how will their families be affected?  Who starts small businesses in Los Angeles?    Again, what types of firms in what industries hire low skill workers?  How will they adapt to now facing a 66% price increase for labor?    Will they become family firms that just hire family members and then pay them with equity in the firm?

The only way to answer these questions would be to have a Census of Los Angeles employers and to explicitly model their labor demand choices before and after the new policy.  Such research would have to figure out what substitution possibilities such firms can engage in --- in the short term and the medium term.  Will LA's firms hire more robots to replace the low skill guys?


UPDATE: I just received this message.  I do not agree with its core claim.  These guys need to take a course in Econ 101.


For Immediate Release: May 20, 2015

Contact: Anna Zuccaro, anna@fitzgibbonmedia.com, (914) 523-9145

Restaurant Opportunities Center United Responds to Minimum Wage Increase in Los Angeles

Despite Aggressive Lobbying, L.A. City Council Rejects Attempts to Illegally Exempt Restaurant Industry from $15 Minimum Wage
 
The following statement is by Saru Jayaraman, co-founder and director of the largest national restaurant workers’ rights advocacy organization in the country, Restaurant Opportunities Centers (ROC) United:
 
“This policy means that tens of thousands of restaurant workers across Los Angeles — the second largest city in the country — are closer to a more livable minimum wage. What’s especially significant is that yesterday’s 14 to 1 vote by the L.A. City Council would maintain one fair wage for all workers, without exemptions for the tipped workforce. California is one of seven states without a separate, lower minimum wage for tipped workers and the state is among the nation's strongest in restaurant industry growth, with $72 billion in sales last year and 11% workforce growth over the past two years.
 
“Although California is considered a leader in progressive policies, its legislators are no stranger to the onslaught of corporate lobbying that has ramped up significantly over the last several months. Corporate restaurant lobbyists aggressively sought a two-tier wage system for restaurant workers, threatening to increase poverty, wage theft and even rates of sexual harassment for tipped workers. ROC-LA, the L.A. Raise the Wage Coalition and L.A. Coalition Against Wage Theft tirelessly attended marches, demonstrations, and hearings, like the one yesterday, to speak truth to power and demand that their city take action to ensure that all workers are treated fairly and ensured the dignity of a fair and stable wage, rather than being forced to live off tips. In addition to the increased wage, ROC-LA and allies won a significant victory by establishing the City's first-ever local enforcement agency with increased powers, such as permit revocation for employers who steal wages and undercut honest businesses.
 
“As L.A.’s city attorney moves forward in drafting legislation, and we await a final vote, we implore members of the L.A. City Council to prioritize those who are the backbone of our economies above corporate interests and reject anything less than a $15 minimum wage for all workers.”
 
###

Co-founded by leading workers’ rights advocate Saru Jayaraman (“One of the top 50 most influential people in the restaurant industry” – Nation’s Restaurant News) ROC United has grown to close to 14,000 worker-members across over 30 cities in the US, winning 15 worker-led campaigns, totaling $8 million in stolen tips and wages.


Causal Policy Effects Induced by Disasters

In the aftermath of the Amtrak train disaster, railroad unions are demanding that a second engineer ride in each train.   This poses an interesting public policy question related to behavioral economics.  A second engineer would be costly to hire and the union would love to have more well paying union jobs.  But, a benefit of the second engineer would be that the probability of a disaster would decline.   A key question is how much would safety be improved by having "four hands on the wheel"?   It has been pointed out that train riding per mile is much safer than auto travel per mile but behavioral economists would point to Matt Rabin's law of small numbers research claiming that people overweight salient events when updating their subjective assessments of risk. If this is the case, then the union could grow rich because of this train disaster.   Why? The public will demand that the 2nd engineer be mandated as new safety policy for "risky" trains.

Will any neo-classical economists have the guts to make the claim that the 2nd engineer on the plane does not pass a cost-benefit test? In the aftermath of the 9/11 attacks was it "good policy" to have U.S Marshals flying on planes?  Did this sharply increase our safety relative to the extra costs?

I am quite interested in how policy entrepreneurs use disasters to try to nudge voters to adopt new policies that help them.  While a second engineer on the train would help the union, how much would it help the people?  How much are we willing to pay save a statistical life?

Intuitively, if there are 200 people on the train and having the 2nd engineer on the train reduces the probability of a crash by .0001 and 1% of the people on the train would die in a crash, then having the extra engineer saves 2*.0001 statistical lives per train trip.

For some economics of the rising value of a statistical life read my 2004 paper with Dora Costa.

For some research on how liberal environmentalists in the U.S Congress use environmental disasters such as Chernobyl to introduce more radical legislation, read my 2007 paper.  

Environmental disasters as risk regulation catalysts? The role of Bhopal, Chernobyl, Exxon Valdez, Love Canal, and Three Mile Island in shaping U.S. environmental law

Unexpected events such as environmental catastrophes capture wide public attention. Soon after five major shocks—Three Mile Island, Love Canal, Bhopal, Chernobyl, and the Exxon Valdez oil spill—Congress voted on new risk regulation. This paper conducts an event study to test whether individual congressional representatives were “shocked” by these environmental disasters into increasing their probability of voting in favor of risk legislation. On average, representatives were less likely to vote in favor of bills tied to these five events. Significant heterogeneity in representatives’ responses to these shocks is documented. Liberal Northeast representatives were most likely to increase their pro-environment voting in the aftermath of these shocks. Copyright Springer Science+Business Media, LLC 2007

Sunday, May 17, 2015

Does California Face Limits to Growth? Evidence from Alfalfa

The NY Times has published a long pseudo-Freudian piece about the respective legacies of Gov. Pat Brown of California and his son Gov. Jerry Brown (the current leader of California).   In a nutshell,  Pat was the leader during a time of growth and optimism.  He liked people and he wanted more people to move to California. In contrast, Jerry is a dour dude who openly declares that there are "limits to growth" and he is not afraid to say "no".  He faces constraints that his father did not face.  A structural budget deficit and the absence of rain. Is California doomed?  Of course not, but it will need to use scarce resources more efficiently.  A silver lining of the combination of bad policies (i.e low water prices and overly generous property rights to water for agriculture and the absence of water markets) combined with drought is that these circumstances nudge the people of California to take a second look at how we are currently allocating scarce resources.

80% of water is used by farmers in this state.  50% of urban water is used outside.  We face prices of less than 1/2 cent per gallon. Could we be using this scarce resource efficiently?  Some facts about agriculture.  Note that rice and alfalfa are not on this list.

Here is our production of alfalfa.  6.1 million tons per year.     It takes 135,000 gallons of water to grow one ton  of alfalfa.   If we sacrificed all of our alfalfa production (which is not one of the 10 biggest commodities for CA), each of the 40 million Californians could have an extra 20,500 gallons of water a year each (56 gallons per person per day)!   No crisis!   No limits to growth. Just simple choices.  Raise water prices and alfalfa land will be converted into a higher use activity.

California's top-ten valued commodities for 2013 are:
  • Milk — $7.6 billion
  • Almonds — $5.8 billion
  • Grapes — $5.6 billion
  • Cattle, Calves — $3.05 billion
  • Strawberries — $2.2 billion
  • Walnuts — $1.8 billion
  • Lettuce — $1.7 billion
  • Hay — $1.6 billion
  • Tomatoes — $1.2 billion
  • Nursery plants— $1.2 billion










Past Evidence on Trends in Within City Health Inequality

Dora Costa and I have just published an urban economics piece about the trends in the geography of health inequality in major U.S cities 100 years ago.  Here is the source: AER May 2015 Papers and Proceedings piece.

Here is our Intro

Today, there is great interest in the geography of opportunity.  Recent research using linked income tax records has documented that poor children born in Atlanta have a lower probability moving up the economic ladder than poor children who grew up in San Jose (Chetty et. al. 2014).   Such research emphasizes the causal role of “place” as a key factor in determining a person’s economic performance over the life cycle.  A necessary condition for moving up the economic ladder is to survive and to be healthy enough to learn and work.   In the United States in the late 19th and early 20th century, large cities had extremely high death rates from infectious disease.   The urban mortality premium relative to rural areas in 1880 was ten years in Philadelphia  (Haines 2001).     However, within major cities such as New York City and Philadelphia, there was significant variation at any point in time in the mortality rate across neighborhoods.


We will show that between 1900 and 1930 neighborhood convergence took place in New York City and Philadelphia such that the death rates in the highest baseline death rate areas declined by more than in the relatively safer areas.     Reductions in the death rates in the most deadly parts of the city meant that a convergence in overall quality of life took place in such cities and this is likely to have mainly benefited lower income individuals and immigrants.  Our investigation of mortality rate dynamics within cities contributes to the urban quality of life literature and the public health literature (Rosen 1988, Gyourko and Tracy 1991, Cutler and Miller 2005). We close by discussing what is known about the causes and consequences of the trends we document.

Saturday, May 16, 2015

Some Links for Today

Sisters meet for the first time at a Columbia University writing seminar.

Entourage and the declining influence of CAA and WMA?

Desalinization as a tool for augmenting California's water supply during drought ;  challenges posed by environmentalists

FAO Schwartz leaves 5th Avenue  --- bricks and mortar vs. the Internet

Friday, May 15, 2015

My 40 Student Fall 2015 Elective Class at USC: Econ 487

I will teach undergraduate Environmental Economics at USC this fall.  If you click here,  you will see that my class enrollment is capped at 40 students and each of my students will have already taken intermediate micro.    After having taught 120 person enviro econ classes at UCLA where roughly half the class hadn't taken principles, this will be a very good experience for me.   I will push the students to work hard and will teach an accelerated course focused on blending applied econometrics with micro theory to think about the causes and consequences of pollution.   Our textbook will be my $1 book and we will have the time (and they will have the tools) so that we can cover everything in the book.


Thursday, May 14, 2015

The University of California Pension Shift Offers a Test of Behavioral Economics

Today, UC's President Janet Napolitano distributed a letter which features some very good news for the University of California.   Instate tuition will remain frozen at roughly $12,000 per year and Governor Brown will provide the UC with extra $.  This deal opens up the "win win" of providing the resources for continuing to have an excellent university while protecting the middle class from tuition increases.

As an economist always seeking "natural experiments", one part of the Governor's letter caught my eye.

Pension changes

·     The agreement's $436 million in one-time funding over three years to help UC pay down its pension liability recognizes the State's obligation to help support UC's pension plan.
·     In exchange for the pension funding, UC would adopt, upon approval by the Regents, a new pension tier by July 1, 2016.  The new tier, which would affect only new employees hired after it is implemented, would provide, at the employee's election, either:
-- A defined benefit plan with a pensionable salary up to the California Public Employees' Pension Reform Act of 2013 (PEPRA) cap (currently $117,020), plus a supplemental defined contribution plan for certain employees,

So What?


The current Pension Cap for incumbent UC Faculty is $240,000 per year.   This means that a UC Professor who serves for 40 years at the UC and whose salary is above $240,000 receives a yearly defined benefit of $240,000 per year when he/she retires.

Under the new rules, a newly hired Professor (hired after 7/1/2016) whose salary is over $240,000 and serves 40 years at the UC would receive a yearly defined benefit of $117,000.

If you do the math, you will see that future professors  (mainly at the Business School, Law School, Medical School and Economics Department) will be paid $123,000 less per year in retirement (assuming they stay 40 years) than current professors who signed on at UC before July 1 2016 at a salary of over $240,000 a year.

So, will Professors in these fields who UCLA tries to hire after 7/1/2016 demand much higher salaries than incumbent faculty to compensate them for their lost pensions?   An ambitious researcher could use the UC Pay Database to track this.  Such a researcher would need to assume that the new hires under the new rules are of the same quality as the incumbent faculty.   This is an economic incidence question.  When future pensions are cut, does current pay rise?  Neo-classical economics would say "yes" , would behavioral economics say "no"?  Why does neo-classical economics predict , current pay will rise for the new hires?  Compensating differentials and competitive labor markets!  To attract talent, the key is the present discounted value of the total compensation plan. If the compensation at the backend (pensions) are cut, then to hold the PDV constant requires paying more money in higher salaries upfront.   A behavioral economist would posit that people discount highly the future pensions and can't calculate PDVs.  We have a clean test!

When I did a calculation, I see that the UC must raise pay by $20,000 per year of salary to offset this 50% pension pay cut at the end of life.

Measured in 1,000s of dollars and using a 3% interest rate; these two salary flows have the same PDV for an individual who joins UCLA at age 30 and retires at age 70 and dies at age 85.


 list age salary1 salary2

     +-------------------------+
     | age   salary1   salary2 |
     |-------------------------|
  1. |  30       150       170 |
  2. |  31       150       170 |
  3. |  32       150       170 |
  4. |  33       150       170 |
  5. |  34       150       170 |
     |-------------------------|
  6. |  35       150       170 |
  7. |  36       150       170 |
  8. |  37       150       170 |
  9. |  38       150       170 |
 10. |  39       150       170 |
     |-------------------------|
 11. |  40       200       220 |
 12. |  41       200       220 |
 13. |  42       200       220 |
 14. |  43       200       220 |
 15. |  44       200       220 |
     |-------------------------|
 16. |  45       200       220 |
 17. |  46       250       270 |
 18. |  47       250       270 |
 19. |  48       250       270 |
 20. |  49       250       270 |
     |-------------------------|
 21. |  50       250       270 |
 22. |  51       250       270 |
 23. |  52       250       270 |
 24. |  53       250       270 |
 25. |  54       250       270 |
     |-------------------------|
 26. |  55       250       270 |
 27. |  56       250       270 |
 28. |  57       250       270 |
 29. |  58       250       270 |
 30. |  59       250       270 |
     |-------------------------|
 31. |  60       250       270 |
 32. |  61       250       270 |
 33. |  62       250       270 |
 34. |  63       250       270 |
 35. |  64       250       270 |
     |-------------------------|
 36. |  65       250       270 |
 37. |  66       250       270 |
 38. |  67       250       270 |
 39. |  68       250       270 |
 40. |  69       250       270 |
     |-------------------------|
 41. |  70       250       270 |
 42. |  71       240       116 |
 43. |  72       240       116 |
 44. |  73       240       116 |
 45. |  74       240       116 |
     |-------------------------|
 46. |  75       240       116 |
 47. |  76       240       116 |
 48. |  77       240       116 |
 49. |  78       240       116 |
 50. |  79       240       116 |
     |-------------------------|
 51. |  80       240       116 |
 52. |  81       240       116 |
 53. |  82       240       116 |
 54. |  83       240       116 |
 55. |  84       240       116 |
     |-------------------------|
 56. |  85       240       116 |
     +-------------------------+







Tornadoes in Oklahoma Offer a Test of Behavioral Economics

As we celebrate the publication of Richard Thaler's excellent Misbehaving,  must neo-classical economists surrender and apologize for our long embrace of Mr. Spock and forward looking investment in the face of uncertainty?   Today in the NY Times, an economist named Kevin Simmons  helps the "rebel alliance" to launch its counter-attack.   Here is his webpage at Austin College. 

He argues that people who live in at risk areas are more likely to invest in pre-cautions to reduce the damage caused by natural disasters such as tornadoes.

A Direct Quote:

"Academic research in the 1970s found that in cases of low-probability hazards with serious consequences, most people ignored the likelihood that the hazards could happen to them. This suggested that people would be unwilling to pay for features that would provide protection against tornadoes.

But research I did with Daniel Sutter, an economics professor at Troy University in Alabama, suggests this belief may be wrong. We found that people were willing to pay more for homes with safety rooms, and to rent lots in mobile home parks with community shelters — features that increase the odds of surviving a deadly storm. Following the 1999 tornado in Moore, some builders there began to strengthen their construction voluntarily, and one of them told me the better construction was responsible for about half of the increase in sales of his homes."

As I have argued before, climate change adaptation offers a great lab for testing behavioral economists' theories versus rational expectations economists predictions about investment under uncertainty. I count Kevin Simmons' work as a data point for the "old" University of Chicago.

Sherwin Rosen and Robert Lucas would posit that self interested households will forecast what challenges are posed by living in a given location and will update these beliefs as new information becomes available.  These households will make investments that reduce their exposure to these risks (see Ehrlich and Becker's 1972 paper on self protection).   Would behavioral economists predict that people will not update their risk perceptions as new news arrives? Or would behavioral economics predict that people will over-react (the law of small numbers see Rabin) to the new news?  

Dr. Simmons sketches a simple model of rational forward looking risk averse individuals taking pro-active steps to protect their families. This is high stakes stuff and this is a clear prediction from old school econ 101.





Wednesday, May 13, 2015

The Economics of Moore's Prediction of "Moore's Law"

Back in the 1990s, Owen Lamont wrote this paper about the incentives of macro forecasters over their life-cycle.   One theory (which I believe he rejects) is that young forecasters seek to stand out in a crowd and they announce a "crazed" prediction (the Dow Jones Industrial Average will rise by 77% this year) and if this occurs they are certified as a genius and they can then cash in on this for the rest of their life.  If they turn out to be wrong, they can exit the sector and go do something else.  So, this option value creates an incentive to take an early daring risk in choosing a forecast to announce.

My mind returned to the economic incentives of forecasters  as I read Tom Friedman's piece about Gordon Moore.   Moore is the guy who claimed that  "Moore predicted that every year we’d double the number of transistors that could fit on a single chip of silicon so you’d get twice as much computing power for only slightly more money. "

Here is a direct quote from Gordon Moore:

"I guess one thing I’ve learned is once you’ve made a successful prediction, avoid making another one,” Moore said. “I’ve avoided opportunities to predict the next 10 or 50 years.”

Amazon Teaches its KDP Authors About the Regression Discontinuity at the 70% Royalty Kink

I am a Amazon Author of the $1 text; Fundamentals of Environmental and Urban Economics.   In this book, I introduce almost all of the ideas of modern environmental economics. The book is targeted both to undergraduates and general readers.  Today, I revised the book and uploaded my revised manuscript.  Amazon tried to gently educate me that I have set the price of my book too low. $3.99 would maximize my profit!   In this graph below, Amazon tries to explain to me (I already knew all of this) that there is sharp discontinuity at $2.99 such that my royalty rate would jump from 35% to 70% if I price my book at $3 or higher.  Amazon understands that demand curves slope down so the continuous downward sloping line is the typical demand curve for a KDP book.  So, by setting the price at $1, I attract more buyers but I get 35% of the $1.  If I charge $3.99, I get .7*3.99*(books sold at a price of 3.99) > .3*1*(books sold at a price of 1)  so I am throwing away profit!  Should I call Richard Thaler to report this anomaly?  Or do we augment my utility function to highlight my eagerness to redistribute $ back to my students and to spread the word about University of Chicago style free market environmentalism?

Amazon should write a paper on whether these nudges lead which authors to jump the notch to charge just over $3.  Regression discontinuities and bunching are fun stuff.  For readers who know what a price elasticity of demand is; note that you can calculate this from the data Amazon provides below.  While the quantity of books sold is not reported here, each author receives this information each month.




Monday, May 11, 2015

Sugarfish in Beverly Hills

This is an infomercial for Sugarfish in Beverly Hills.   If you like great food and you seek to stimulate our economy (that's a joke), then you should go there.   After you eat at Sugarfish, you can walk two blocks to Rodeo Drive and pretend that you are Thorstein Veblen and study the "leisure class".   Here are some images you will see.  Note the blue sky and the 75 degree temperature almost every day of the year.  






Image result for rodeo drive beverly hills



Thaler's New Book as Autobiography and History of Intellectual Thought

Richard Thaler's Misbehaving has been published.   I own an e-copy and I've read 4 chapters now.  Thaler writes as he speaks.  He is lucid and witty. He doesn't write like an academic!   While each academic has an "intellectual journey", most of such journeys do not merit a full length book (perhaps an Amazon Single?).  Thaler has earned the right to tell his story as a 400 page book.  Type his name into Google Scholar and you will see a superstar at work.

His work is a thorn in the side of typical University of Chicago economics research so it is quite ironic and appropriate that he is a Professor at the University of Chicago.  He does appear to be quick to dismiss rational expectations and other key ideas in the neo-classical toolkit.

The book starts by telling his autobiography in chronological order.  At the start of his career he was brilliantly iconoclastic and lucky.  He got to spend an early year at Stanford with two giants from Psychology  and this helped him to launch and to battle the neo-classical establishment.

While I have much more reading I need to do here, I believe that the key issue for the relevance of behavioral theories for economics more broadly is a population mixture issue. What % of people are "behavioral"?  In the real world, the population is a mixture of Spocks (rational expectations guys) and Homer Simpsons (the Thaler behavioralists).  Since Thaler is not a structural econometrician, he will not embrace the Heckman and Singer approach for disentangling what % of the diverse population is of each group.   This is the key issue.  Economists study general equilibrium and interactions between various groups. A key issue is what happens to the well being of heterogeneous groups when they trade and interact with each other? The early noise trader literature noted this point but I don't know where this literature has gone.  What treatments can nudge behavioral guys to debias themselves? What role does culture and family play in developing preferences and a world outlook that converges to the rational expectations view of people?

Issues of the Arrow Impossibility Theorem also arise. If the population is diverse with respect to behavioral tendencies, then what is good public policy?  Benjamin and Shapiro have written a key paper that provides some insights here if you are worried about the 99%.




Sunday, May 10, 2015

Anticipate, Invest and React

Two weeks ago Justin Gillis reported that weather extremes will become more severe and more likely in the near future.   Now that we know what to expect, how do we adapt? How effective will our investments be in protecting us?

A Direct Quote:

"If global warming can be brought under control as rapidly as many environmental activists would like, keeping global warming below three degrees Fahrenheit, the new study found that heat extremes might increase only 14-fold later this century, compared with their frequency in the preindustrial world.  But runaway emissions, causing the planet to warm by more than five degrees Fahrenheit, would lead to a 62-fold increase in heat extremes, the researchers found. Other studies have forecast levels of heat and humidity by late this century that could make it dangerous for people to work and play outside, possibly for weeks on end."

So, in the ugly "high carbon scenarios", how will we adapt?  We will air condition and as shown by Singapore today -- economic activity will take place in Malls and indoors and people will come out at night. We will reorganize our days so that hours outside will be minimized during these anticipated extreme times.  What will we lose by rescheduling our days?  The Internet will help to reduce outdoor exposure.  In an extreme case, people will wear spacesuits to reduce their exposure to the outdoor heat.

I would like to see some serious economists write down models in which an anticipated trend (i.e more violent heat waves)  over the next 50 years causes serious economic losses for an urbanized economy.   I am now working on this topic in two independent projects.

A rational expectations economist would say that an anticipated trend can't cause serious damage because we will make investments (that yes will be costly but since the capital stock does not live forever -- we must always reoptimize) that will help to defend us.  Behavioral economists would argue that we are idiots and we will not make these investments because we will naively assume that tomorrow will be like yesterday and we will under-estimate future heat and thus won't be ready when the heat arrives.

In studying future heat's impact on our economy,  you can't use current hedonic real estate pricing comparing Phoenix (a hot city) to San Francisco ( a temperate expensive cit).  These cities differ on too many dimensions beyond just climate.  Also, such a researcher is assuming that no technological change will occur that helps us to adapt.  There are many million dollar homes in hot Phoenix today.  Rich people are living comfortably there today.  A home is a durable asset. If people are paying over $1 million for a home in Phoenix then they are voting with their wallet that they believe that Phoenix will continue to be a good place to live and work even as climate change unfolds.  Do you believe that these buyers are idiots? Or do you agree that they are climate change adaptation optimists?

A second key question is whether poor people can continue to live pretty well in such cities. In this sense, adapting to climate change becomes an environmental justice question. To raise the chances that the poor can live well in affected cities requires capitalist innovation and helping the poor to raise their income levels through better schooling investments (pre K for all) and reducing labor market barriers such as the minimum wage.



Negative Pollution Externalities Associated with Nail Salons in the Consumer City

Urban economists have created an exciting new literature examining the rise of "Consumer Cities".  Glaeser, Kolko and Saiz started this literature (see this and this) by arguing that cities have always been centers of production but that a new trend is that cities are centers of consumption (eating, culture, entertainment, shopping).   For people who do their work "On the Cloud" and thus can locate anywhere, they still need to go to a physical restaurant to eat (your food can't be emailed) and big cities with people who have similar tastes as you will have the scale economies to provide the varieties of food that you like.  This means that big cities with educated, wealthy people will have really upscale targeted varieties (see the work of Rebecca Diamond and Joel Waldfogel).  In recent work, Jonathan Dingel and Don Davis and E. Morales have ingeniously linked up the benefits of the decline in urban crime to the gains people enjoy from living in a city where there are more consumer opportunities (i.e women spend more time outside when they feel safer and thus the consumer city is more valuable to them).

But, what is missing from the "Consumer City" literature is an examination of the negative externalities from local consumption of specific services.  Think of car repair places, gas stations and nail salons and dry cleaners.  These consumption service places are nasty in a local radius for both the workers and people in a vicinity of them as they use chemicals to provide their services.  The NY Times provides an in depth discussion of the localized externalities at  Nail Salons and provides some geography maps.  As an economist I have a few questions;

1.   Do Nail Salons pay "combat pay" to their workers to reward them with higher wages for the nasty work conditions?  A labor economist who believes in competitive markets would argue that the workers in the nail salons would merely need to be educated about the risks  they face by working at the current nail salons and that this symmetric information would be sufficient to affect their labor supply and in aggregate the equilibrium wage that they are paid for their efforts at a nasty workplace.

2.  Do the nasty smells at nail salons exit the building and stink up the surrounding area?  If "yes', then suburbanization and low density offers a solution.  Recall the old line that the "solution to pollution is dilution".

Is the NYC Nail Salon problem in part caused by the fact that NYC is so dense. If it were spread out would there still be a nasty externality?

So, the point of this blog post is that when we live at high density , we have easy access to consumer amenities we want (such as a Starbucks down the corner) but we also have easy access to disamenities. What is the optimal density to live at? Where do we want noxious activity to locate? In my new research on China, we argue that land intensive polluting activity is leaving big cities because of standard opportunity cost logic but nail salons have an incentive to stay close to the population centers which demand these services. When there is a negative externality associated with consumption, what is the optimal "pollution tax"?  Will it be implemented?


Saturday, May 09, 2015

Economic Inequality and the Future of Progressivism

Mayor De Blasio is heading to UC Berkeley to give a talk thanks to an invite from  Prof. Bob Reich.  While Mayor De Blasio is unlikely to hire me to write his speech, I thought that I should offer a sketch of how I would present his talking points.   His talk will be titled; "Economic Inequality and the Future of Progressivism".   Here is my version;


Dear Friends,

May 2015 is a funky time in U.S history. We are at peace.  The stock market and the unemployment rate and the price of gasoline suggest that this is a time of prosperity.   Yet, there are millions of angry people.  We fear that China is running past the U.S as the world's superpower.  Many immigrants face challenges of becoming legal  citizens, police brutality is a daily media topic, climate change lurks, the federal budget deficit lurks, entitlements for senior citizens and rising health care costs dominate our 30 year fiscal horizon.  Center city public schools are not properly preparing our young to achieve their full potential.  Housing in many major center cities is priced at extremely high levels.

What can be done?

1.   Auction off 3 million passports for the United States with a minimum price of $500,000 each.   This will increase our nation's population size by less than 1% and will collect at least $1.5 trillion in revenue.  Homeland Security will screen who these individuals are.   An age restriction could be tied to the auction to attract younger people.

2.  Use the new revenue money from #1 to provide good pre-K for every child in the United States.  If pre-K costs $5,000 per child and if 1 million children sign up for this new program then there is plenty of $ left over.  Experiment with both private vouchers for pre-K and public vouchers and compare the cost per trained child under the two regimes and compare test score and non-cognitive skill development to see whether the public sector does achieve a higher bang per $ than the private sector.  Use part of the new revenue to modernize our military hardware and drone technology.

3.  Abolish the minimum wage and relax firing laws to encourage employers to expand hiring.  Increase the earned income tax credit.

4.  Change housing zoning laws to allow for 30 story buildings within ten miles of the city center on any plot of available land.

5. Gradually raise the age for receiving Social Security by 3 months per year and include a greater co-pay for Medicare tied to total lifetime Social Security earnings.

6.  Reform drug sentencing to reduce jail time for non-violent crimes.

7. Introduce a cap and dividend carbon tax so that any revenue received is recycled back to the public on a per-capita basis each year.

8.  Reduce the cost of voting (Smart Phone App?)  so that everyone who is of voting age actually votes.  This would reduce the power of special interests in the primaries and would lead politicians to be more "inclusive".

9.  Introduce driverless urban buses that allow the population to move across space but without facing the labor intensive cost of union drivers driving such buses.

10.  Create a mechanism such as Glen Weyl's quadratic voting to allow citizens to reveal their desires for improved infrastructure ranging from airports, to parks, to highways.  Allocate $ to improve such public goods based on such a revelation mechanism.

Would Big Bill give this talk?  What would Dr. Reich think?  While I doubt that he would like the means, I'm confident that he would salute the end outcomes.   If the U.S adopted this agenda, we would be richer and happier and safer.






Friday, May 08, 2015

Convert NYC's La Guardia Airport into New Housing

The NY Times often repeats itself over and over again.  Few new ideas are explored.  Today though, George Haikalis has written a brilliant Opinion Piece.  He makes a compelling case that La Guardia Airport should be closed and its 680 acres of prime New York City (in Queens) real estate should be converted into residential housing.  Based on his algebra, 30,000 new units could be built.  Suppose that each sells for $1 million dollars.  This $30 billion dollars in new real estate would look great and the Mayor of New York City would collect roughly $500 million dollars a year in new property taxes each year.  Mr. Haikalis makes the case that Newark Airport and JFK Airports have the space and the spare capacity to handle the flights that would land in the metro area if this ugly La Guardia Airport ceased to exist. He also understands externalities.  The La Guardia Airport occupies some very valuable real estate. If the noise and pollution externality from the airport vanishes then some great homes could be built.  He optimistically argues that  a fast train could be built to connect JFK Airport to Manhattan.  Whether such a "Shanghai" type project could be implemented in NYC is an open question.  What I like about his piece is his open mind for re-inventing the NYC region.  Real estate economists would push for tall buildings to be built on the "liberated land".

Thursday, May 07, 2015

Does the State of Texas Pay Public Employees Equally for Equal Work? (Yes, it does)

I was once a labor economist.  At the University of Chicago, my thesis committee consisted of Sherwin Rosen, Gary Becker and Bob Willis.  I have published a piece in the Journal of Labor Economics.  I was Jacob Mincer's colleague at Columbia for several years.  To convince you that I still remember the basics,  I'd like to show you some new facts I've learned about the public sector labor market in the great state of Texas.  For the universe of 149,160 state workers , the Texas Tribune posts some of their demographics, what agency they work for, their job title and their annual salary.  Here are the data.  I have written this blog post to highlight what cool data the Texas Tribune has collected through a freedom of information act request.  This newspaper should be congratulated for its effort.


Some definitions


annual = $ of salary
Male = 1 if the employee is a man, 0 otherwise
Asian = 1 if the employee is Asian
Black =1 if the employee is black
Hispanic =1 if the employee is Hispanic
other = 1 if the person is not white and all of these other categories for race and ethnicity = 0
hireyear = the year the worker was hired by the stateNote that the average worker earns $44,002 dollars.  48% of the workers are Black or Hispanic.  

Here are the summary statistics

. summ annual Male Asian Black Hispanic other hireyear

    Variable |       Obs        Mean    Std. Dev.       Min        Max
-------------+--------------------------------------------------------
      annual |    149164    44002.17    21902.74        300     540000
        Male |    149164    .4324703    .4954204          0          1
       Asian |    149164    .0221233     .147085          0          1
       Black |    149164    .2292376    .4203438          0          1
    Hispanic |    149164    .2524939    .4344445          0          1
-------------+--------------------------------------------------------
       other |    149164    .0050682    .0710112          0          1
    hireyear |    149164    2007.356    6.738477       1964       2015


In case, you haven't studied the linear regression model then you should click here.
Fact #1  In Texas, the average male public employee earns $6852 more dollars per year than the average woman.  Discrimination? Or do they work in different jobs?  Let's see. 

. reg annual Male

      Source |       SS       df       MS              Number of obs =  149164
-------------+------------------------------           F(  1,149162) = 3672.23
       Model |  1.7194e+12     1  1.7194e+12           Prob > F      =  0.0000
    Residual |  6.9839e+13149162   468206617           R-squared     =  0.0240
-------------+------------------------------           Adj R-squared =  0.0240
       Total |  7.1558e+13149163   479730226           Root MSE      =   21638

------------------------------------------------------------------------------
      annual |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
        Male |    6852.98   113.0874    60.60   0.000     6631.331    7074.629
       _cons |   41038.46   74.36909   551.82   0.000      40892.7    41184.23
------------------------------------------------------------------------------

Controlling for gender and tenure on the job, the average Asian earns roughly $16,500 more per year than the average Black in the Texas public sector.  Is this discrimination?  Let's see.  For every year of public sector tenure, the average worker earns $849 more dollars per year.



. reg annual Male Asian Black Hispanic other hireyear

      Source |       SS       df       MS              Number of obs =  149164
-------------+------------------------------           F(  6,149157) = 3924.04
       Model |  9.7555e+12     6  1.6259e+12           Prob > F      =  0.0000
    Residual |  6.1803e+13149157   414345617           R-squared     =  0.1363
-------------+------------------------------           Adj R-squared =  0.1363
       Total |  7.1558e+13149163   479730226           Root MSE      =   20355

------------------------------------------------------------------------------
      annual |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
        Male |   4783.087   107.6172    44.45   0.000     4572.159    4994.014
       Asian |   7323.259   362.4389    20.21   0.000     6612.886    8033.632
       Black |  -9197.407   134.7434   -68.26   0.000    -9461.502   -8933.313
    Hispanic |  -6792.392   129.4467   -52.47   0.000    -7046.105   -6538.679
       other |  -3815.451   744.1512    -5.13   0.000    -5273.972   -2356.929
    hireyear |  -849.9424   7.876481  -107.91   0.000    -865.3801   -834.5047
       _cons |    1751751    15811.1   110.79   0.000      1720762     1782741
------------------------------------------------------------------------------

. gen lannual=log(annual)


Now I switch the dependent variable to the natural log of salary.  Men earn 9% more than women.  Blacks earn 17% less than Whites. The returns to a year of public sector tenure = 1.9% per year.


. reg lannual Male Asian Black Hispanic other hireyear

      Source |       SS       df       MS              Number of obs =  149164
-------------+------------------------------           F(  6,149157) = 4902.40
       Model |  4182.00948     6   697.00158           Prob > F      =  0.0000
    Residual |   21206.471149157  .142175499           R-squared     =  0.1647
-------------+------------------------------           Adj R-squared =  0.1647
       Total |  25388.4804149163  .170206287           Root MSE      =  .37706

------------------------------------------------------------------------------
     lannual |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
        Male |   .0932735   .0019935    46.79   0.000     .0893663    .0971807
       Asian |   .1300037   .0067138    19.36   0.000     .1168449    .1431626
       Black |  -.1701422    .002496   -68.17   0.000    -.1750342   -.1652501
    Hispanic |  -.1244257   .0023979   -51.89   0.000    -.1291255    -.119726
       other |  -.0634405   .0137845    -4.60   0.000     -.090458   -.0364231
    hireyear |  -.0189166   .0001459  -129.65   0.000    -.0192026   -.0186307
       _cons |   48.60093   .2928822   165.94   0.000     48.02688    49.17497
------------------------------------------------------------------------------

Now I include agency fixed effects. The black wage gap shrinks from -.17 to -.11. This suggests that on average that blacks work for lower paying agencies in Texas government.


. areg lannual Male Asian Black Hispanic other hireyear , absorb(agency)

Linear regression, absorbing indicators           Number of obs   =     149164
                                                  F(   6, 149042) =    3609.49
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.3574
                                                  Adj R-squared   =     0.3569
                                                  Root MSE        =     0.3308

------------------------------------------------------------------------------
     lannual |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
        Male |   .0883419   .0018595    47.51   0.000     .0846974    .0919865
       Asian |   .0923301   .0059193    15.60   0.000     .0807284    .1039318
       Black |  -.1101269    .002251   -48.92   0.000    -.1145388    -.105715
    Hispanic |  -.1158061   .0021364   -54.21   0.000    -.1199935   -.1116188
       other |  -.0440313   .0121027    -3.64   0.000    -.0677524   -.0203103
    hireyear |  -.0156331   .0001332  -117.36   0.000    -.0158941    -.015372
       _cons |   41.99657   .2673651   157.08   0.000     41.47254     42.5206
-------------+----------------------------------------------------------------
      agency |    F(115, 149042) =    388.679   0.000         (116 categories)


Now I include fixed effects for government agency and for job title.  Note that all of the demographic coefficients go to zero!  Controlling for agency and job title, there are no serious differentials across demographics groups in state government in Texas.  Asians, men and whites tend to work in the higher paying agencies and occupations.


. areg lannual Male Asian Black Hispanic other hireyear , absorb(cat)

Linear regression, absorbing indicators           Number of obs   =     149164
                                                  F(   6, 142197) =     647.88
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.9073
                                                  Adj R-squared   =     0.9028
                                                  Root MSE        =     0.1287

------------------------------------------------------------------------------
     lannual |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
        Male |   .0038785   .0008292     4.68   0.000     .0022532    .0055038
       Asian |  -.0005317   .0024335    -0.22   0.827    -.0053013    .0042379
       Black |   -.004298   .0009229    -4.66   0.000    -.0061069   -.0024891
    Hispanic |  -.0045414   .0008728    -5.20   0.000    -.0062521   -.0028307
       other |    .004608   .0048288     0.95   0.340    -.0048563    .0140723
    hireyear |  -.0041577   .0000671   -61.98   0.000    -.0042892   -.0040262
       _cons |   18.94742   .1346603   140.71   0.000     18.68349    19.21135
-------------+----------------------------------------------------------------
         cat |   F(6960, 142197) =    163.660   0.000        (6961 categories)

. areg annual Male Asian Black Hispanic other hireyear , absorb(cat)

Linear regression, absorbing indicators           Number of obs   =     149164
                                                  F(   6, 142197) =     817.34
                                                  Prob > F        =     0.0000
                                                  R-squared       =     0.9472
                                                  Adj R-squared   =     0.9446
                                                  Root MSE        =  5154.9455

------------------------------------------------------------------------------
      annual |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
        Male |   204.5159   33.22721     6.16   0.000     139.3912    269.6406
       Asian |  -20.35034   97.50772    -0.21   0.835    -211.4636    170.7629
       Black |  -249.1773   36.98002    -6.74   0.000    -321.6574   -176.6972
    Hispanic |  -284.3171   34.97338    -8.13   0.000    -352.8643     -215.77
       other |   148.7667   193.4853     0.77   0.442    -230.4607    527.9941
    hireyear |  -186.3085      2.688   -69.31   0.000    -191.5769   -181.0401
       _cons |   418029.7   5395.722    77.47   0.000     407454.2    428605.3
-------------+----------------------------------------------------------------
         cat |   F(6960, 142197) =    313.725   0.000        (6961 categories)