In New York City, Hong Kong, Shanghai and London and Paris,  rich people live in apartments.  President Trump owns several apartments in Manhattan.  Poor people also live in apartments.  These apartments are of different sizes, located in different areas and in different buildings and apartments are of different quality levels as well.  In the United States, the vast majority of people live in spread out single family homes.

The majority's choice to not live in multi-family apartment buildings has social consequences.  If more of us lived in apartments in multi-family buildings, then we would walk more and use public transit more and our most productive cities would offer greater economic gains for more people. 

In this blog post, I want to sketch out a thought experiment.  For every American adult, let's consider what his/her willingness to pay for living in a house versus an apartment.  What elements of the structural utility function differ between the two?   What attributes are bundled into a house versus an apartment and why do the private attributes of a house offer greater utility than if the same attributes (such as a swimming pool) are part of an apartment complex where you share this club good with your neighbors.

1. Homes have private backyards  --- While this is true, the market indicates that people do not greatly value their backyard. When you run a hedonic pricing regression of;
home price =  controls + b1*indoor space  +  b2*outdoor space  + U  , where indoor space and outdoor space are measured in square footage;  b1 is much greater than b2.  This indicates that people do not pay much for outdoor space and that in the absence of zoning that a developer would convert the outdoor space into more indoor space.  There is an arbitrage opportunity that zoning does not allow to be engaged in.  

I am sure that in hot places that home buyers are willing to pay more for a backyard if they have a swimming pool there. There are scale economies with respect to such pools. If you trust your neighbors, you can have a common pool that you all enjoy together.  As crime falls in cities, more people (see central park in NYC) are willing to have their kids play in public rather than in private (i.e your own backyard).

So, note my key point here --- the same housing attribute (having a swimming pool) --- yields different marginal utility depending on whether you trust your neighbors or not.  If you trust them less, then your demand for your private pool is higher.  Structural IO economists do not explicitly model this "marginal utility" parameter.   They use revealed preference methods to estimate a single marginal utility parameter and perhaps allow it to vary by demographic groups.  But,  if you trust your neighbors then a public pool is close to a perfect substitute for a private pool and you will be more willing to live in an apartment complex where you share the pool with your neighbors.  Trust is an attribute that the home buyer knows he/she has but the econometrician does not observe this (so this is a case of Jim Heckman's essential heterogeneity).  Heckman's work on essential heterogeneity has not been incorporated into the Sherwin Rosen 2-step hedonic research.  I recognize that a BLP style researcher could estimate a "random effects" parameter for the home's pool attribute but this approach would not yield a microfoundation for why the marginal utility distribution changes over time.  My "model" sketched here predicts that the mean marginal utility and the standard deviation of this marginal utility from having your own pool shrinks as trust in neighbors increases.

2.  Homes have more privacy  ---  In apartment buildings, one may be subject to sounds from neighbors and smells of smoke and cooking.  An interesting question is how much progress has been made in designing windows, and walls that limit these negative interaction.  Of course, there are negative externalities associated with having neighbors but market products can offset these.

3.  No collective action issues arise with a single family home --- If you live in a multi--family apartment, you are either a renter or a member of a collective that has to make some group decisions and if people differ in their preferences for local public goods (such as what plants to plant around the property or whether to replace the building's boiler)--- some members of the group will not be happy with the majority decision.   I'm not sure that this is a big deal.

4.  Homes are larger than apartments   --- This is an arbitrage again.  If you want a large apartment, can you buy two adjacent smaller apartments and blow out some walls and turn two 2 bedroom apartments into a 4 bedroom apartment?

My point here is that I believe that for more and more Americans that apartments and single family homes are becoming close substitutes.  But, there is a supply side.  Since the housing supply is durable and we have built perhaps 80 million or more single family homes that each might last for 60+ years , the transition to an apartment complex cities will take a very long time.

When real estate developers build durable structures, they are locking society into a pattern of land use that remains for decades.


So, in a society featuring lower crime and trust between strangers --- I believe that apartments and single family homes become close substitutes and then the supply side (zoning and durable capital) together determine how our cities actually look in terms of urban form.

The protests in France over raising gasoline taxes there highlights that middle class people understand that higher carbon taxes have income effects.  If you drive 15,000 miles a year and if your vehicle achieves 30 miles per gallon and if the price of gasoline increases from $4 to $4.40 due to a 10% increase in the gas tax, then your disposable income declines each year by (15000/30)*.4 = $200.

Economists celebrate the substitution effects induced by the carbon tax --- that people who drive will demand more fuel efficient vehicles and drive them less.  On the supply side, the tax will nudge firms such as Tesla to engage in induced innovation to create even more fuel efficient vehicles.

Since voters are smart and do not want to be poorer (as their purchasing power declines due to the tax),
economists have pondered how to offset the income effect through policies such as "tax and dividend" or by lowering income taxes and raising carbon taxes (see Gib Metcalf's 2007 Hamilton Project paper).

A deep issue arises here.  Who has the property rights to pollute?  If the incumbent polluters have this right, then the designed policy must fully offset the negative income effect I sketched above. Recall that in the 1990 Clean Air Amendments that created the so2 sulfur dioxide market that utilities received free allotments of permits. This meant that they had the property right to pollute and this must have angered some environmental groups. But, the tight cap on total emissions and the incentive effect of being able to sell unused permits created an incentive for these polluters to reduce their emissions.

In my work with Jonathan Eyer,  (see our 2017 paper)   ,  we explore how states and local governments have tried to protect their coal interests in the face of increased federal government regulation and market conditions favoring using natural gas for generating electricity.  On some level, this is a battle over property rights.

Do fossil fuel consumers and producers  have the property rights to engage in this activity?  If they do, then those who seek to mitigate the challenge of climate change must compensate them for their income loss associated with carbon pricing.  Are progressives willing to identify themselves and pay for this property?    If these polluters do not have this property right, then they will suffer an income loss from this new well intended policy and they will use their full arsenal of strategies (including protests) to oppose a change from the status quo.

Given that every American differs with respect to her current production/consumption of fossil fuels, how does  a smart public finance economist design a carbon tax and refund policy that induces the substitution effect of carbon pricing without the income effect?

The political economy of climate change mitigation and adaptation has not been fully explored by academic environmental economists who in recent years have focused on creating computable general equilibrium IAM models (see Nordhaus) or on reduced form empirical studies examining   the "cause and effect" relationship between climate effects and economic outcomes.  Such reduced form "cause and effect" studies should play a key role in determining which voters support carbon taxes. For example, if my home will be flooded because of climate change then I have strong asset protection incentives to vote in favor of a carbon tax.   The role of self interest (beyond ideology) in spurring support for carbon taxes should be explored more in new research.

What else do we know about the political support for carbon pricing?   Riley Dunlap has been the leader in environmental sociology studying long run trends in support among republicans and democrats.
Michael Greenstone released an optimistic contingent valuation study a few years ago.  I tend to be skeptical about such survey evidence.   I wish that his survey is right.  My results in my 2013 paper on the voting on the Waxman-Markey Carbon Tax bill in Congress and my 2015 paper  on California's voting on introducing carbon pricing tell a different story.   High carbon area voters oppose such taxes.  This dovetails with this blog post's main theme.

Soren Anderson has new research on this subject;  Here is his preliminary paper.   Read the abstract and you will see that his paper's findings are consistent with this blog post's main themes and with my past research findings. In studying recent voting on Washington state's proposed carbon tax he finds;

" Support (for carbon taxes) is weaker in precincts with larger shares of car commuters, bigger homes, and workers in carbon-intensive industries and stronger in precincts with larger shares of young people, racial and ethnic minorities, college educated adults, and voters that are ideologically aligned with the left’s broader policy agenda."

This is the challenge that we environmental economists face as we try to implement incentives to combat climate change.  Let the competition to design a proposal that induces substitution effects without negative income effects begin!

UPDATE;    A fundamental question in microeconomics asks;  "who is at the margin?" In the case of supporting carbon pricing a given person will support the policy if her expected present discounted value of benefits from the policy exceeds the expected present discounted value of the costs she will incur from the policy.

In an economy where people differ on many attributes such as location, asset ownership, industry, education -- it is difficult to quantify these factors and include them in a voting regression. After all, we do not observe how individuals vote on election day; instead we rely on precinct level data and face the ecological regression fallacy.


This is a long winded way of saying that if the costs faced by suburbanites for voting in favor carbon taxes decline then more suburbanites will vote for carbon taxes and support Representatives who vote in favor of these policies. Our 2017 paper explored how the private choice of buying solar panels bundled with electric vehicles could flip some suburban voters toward supporting carbon pricing because the income effect they would face would shrink to zero.

The New York Times published an interesting piece about how U.S energy infrastructure might be affected by climate change.   At the end of the piece, there is a nuanced quote about whether the energy companies will experience large profit losses due to climate change.  For some, this would be a delicious irony but Sarah Ladislaw argues that this is wishful thinking.  Here is her quote:

"Yet energy analysts cautioned against expectations that the effects of climate change will cause irreparable harm to the fossil fuel industry or make oil, gas and coal production fundamentally unattractive to investors. Sarah Ladislaw, an energy analyst at the Center for Strategic and International Studies, noted that the oil and gas sector has a long history of managing risks, including figuring out how to operate in politically unstable countries and prodding governments to loosen regulations they find too burdensome.