Thursday, July 18, 2019

What is it about LA?

Cities will make the following deal with developers: we'll give you more density, if you provide units that are deed restricted to be "permanently affordable," meaning will have rents below some ceiling that is tied to area median income.

I have now talked to developers who say the deal works in Seattle and New York, but not in Los Angeles, because city governments around LA specify so much of what must be done during the development process.

This argues (again) for the need for performance based regulation.  For instance, when the EPA requires auto makers to have a minimum MPG for their fleet, it doesn't tell the auto makers how to get there--it just says, "get there."  

Friday, July 12, 2019

Housing really is harder to afford

I have enjoyed reading Michael Kinsley's stuff for as long as I can remember.  The problem is that he inspired a group of young people to value cleverness and contrariness, and so we get stuff like Kevin Drum in Mother Jones saying that housing costs are not rising at an alarming rate.

Allow me to present two pictures (and again, let me acknowledge the people at IPUMS for making it easy to use census and American Community Survey Data).  The first is median rent to median renter income for about 225 MSAs in 2000.

The next is the same picture, but for 2016.

Notice how in 2000,  in the vast majority of MSAs, the median renter would spend less than 30 percent of income on the median rental unit.  By 2016, that had reversed: the median income renter pays more than 30 percent of income in rent in the majority of cities.  And while the 30 percent number is somewhat arbitrary, that fact that rents relative to incomes rose nearly everywhere is not.  It is easy to see why renters are upset.

Thursday, July 11, 2019

Joel Kotkin, Housing Affordability, and the Fringe

Joel Kotkin notes quite correctly that housing is expensive relative to incomes in California--he is certainly not alone in this view.  His cure for this particular illness--make it easier to develop on California's metropolitan fringe.

At first glance, he has a point.  The median price of a house in Banning, California (which I think counts as the fringe), is $257,000, which seems pretty affordable.  Here is where Banning is:

There is not a lot of, um, public transit available in Banning.  The nearest job centers are in Riverside and Palm Springs, which are 31 and 23 miles away.  Riverside has about 170,000 jobs; Palm Springs has about 26,000 jobs.  So let's say average distance to jobs is about 30 miles.

What are the implications of this for affordability?  First, the cost of driving is, according to the Federal Government, about 58 cents per mile.  This means driving cost is about $35 per day.  Let's say people can work from home one day a week, so they drive to work 16 days per month.  Living in Banning adds $560 per month relative to living next to a job.

Now let's say you're a parent and need to pay for daycare.  As best as I can tell searching daycare websites, the least expensive late afternoon care costs about $8 per hour (feel free to correct me if I have this wrong).  Assuming an afternoon drive of 45 minutes to Banning, that is another $128 per month per kid.

We have now added $688 per month for a household with one young child in living costs by living at the fringe.  At a 4 percent interest rate, this "payment" translates into a $144,000 mortgage--that $260k house is similar in cost to a $400K house near jobs.

Note I haven't taken into account the opportunity cost of time, and I am only talking abou private costs.  Surely there are social costs to having people drive longer distances--particularly with respect to greenhouse gas emissions. 

Tuesday, July 09, 2019

Where are the half-million units?

One of the many useful things the people at the St. Louis Fed do is maintain the FRED page, which has a cornucopia of economic data that is easy to download and graph.   Tonight I just happened to be curious about this picture:

The red line is permitted housing units; the blue line is completed unit.  Let's assume that there is a lag of one year between permitting and completion (the length of the time series means that any reasonable lag assumption should be innocuous.  If we look at permits going back to 1967 and ending one year ago, we find that slightly more units were completed than permitted (this is OK, because small places do not necessarily report building permit data to the Commerce Department).  But since 2008, about 500,000 fewer units have been completed than permitted.  This is about a six percent melt from permitting to completion.

So the question is: why?  Very curious about this.

Thursday, July 04, 2019

E.B. White on the definition of democracy (h/t Leslie Appleton-Young)

E.B. White in the New Yorker in 1943:

We received a letter from the Writers’ War Board the other day asking for a statement on “The Meaning of Democracy.” It presumably is our duty to comply with such a request, and it is certainly our pleasure.

Surely the Board knows what democracy is. It is the line that forms on the right. It is the don’t in don’t shove. It is the hole in the stuffed shirt through which the sawdust slowly trickles; it is the dent in the high hat. Democracy is the recurrent suspicion that more than half of the people are right more than half of the time. It is the feeling of privacy in the voting booths, the feeling of communion in the libraries, the feeling of vitality everywhere. Democracy is a letter to the editor. Democracy is the score at the beginning of the ninth. It is an idea which hasn’t been disproved yet, a song the words of which have not gone bad. It’s the mustard on the hot dog and the cream in the rationed coffee. Democracy is a request from a War Board, in the middle of a morning in the middle of a war, wanting to know what democracy is.

Tuesday, July 02, 2019

Lots of folks over 65 are spending a lot on housing.

The Census has a nice tool that allows one to map American Community Survey data by counties (at least counties with sufficient population to develop estimates based on samples).  I drew two today.  This first one is the share of those renters over the age of 65 who pay more than 30 percent of their income on rent.

The second one is the share of those owners over the age of 65 who spend more than 30 percent of their income on homeowning.

The picture for elderly renters is pretty grim: in most counties that are mapped 45 percent or more are spending more than 30 percent of their income on rent.  So one may take a small amount of comfort in the fact that the 78 percent of those over 65 are homeowners.  But in the median county mapped here,  a quarter of those over the age of 65 pay more than 30 percent on housing cost.  This is because such people have either not paid off their mortgage, or have high property taxes.

Altogether, about 55 percent of those 22 percent of the 65+ households who rent pay more than 30 percent of their income in rent, and 26 percent of the 78 percent who own pay more.  This means nearly 1/3 of those over the age of 65 live in an unaffordable house.

It is hard to imagine things getting better.  The old population is growing quickly--the very old population is growing even more quickly.  As people age, their incomes fall, so absent falling housing costs, the share that will be pressured by housing costs will increase in the years to come.     

Sunday, June 30, 2019

Is it reasonable for Angelinos to expect a shorter commute?

I have lived in LA for 11 years now, and I think it is safe to say that among the top topics of conversation is traffic and the misery of commuting.  And in a sense, people are right about
the fact that they have long commutes:  LA-OC Metro 2017 American Community self-reported average travel time to work ranks 246 among the 260 largest Metropolitan Areas, and share of people with a commute of an hour or longer ranks 231 (we will get to who does worse in a bit).  Thanks to Steven Ruggles, Sarah Flood, Ronald Goeken, Josiah Grover, Erin Meyer, Jose Pacas and Matthew Sobek, IPUMS USA: Version 9.0 [dataset]. Minneapolis, MN: IPUMS, 2019, who make it easy to download various data sets and play with them.

But should people reasonably expect LA to do any better than this?  Consider this plot of number of commuters in an MSA against average travel time (I looked to see if someone else had done this particular plot, and couldn't find anyone who has.  If I have failed to cite work already done on this, I apologize).

Two things are pretty clear: larger cities have worse commuting times, and Los Angeles has shorter commutes than New York, but also shorter commutes than some places with smaller numbers of commuters.

How does it do in context?  If we run a simple bivariate regression of number of commuters explaining travel time, we get this:

That is a lot of explaining for one variable--the t-statistic is 11 and the R2 is .32.  And I don't worry too much about getting the direction of causality wrong here.  The regression predicts that LA's commute should, given its size, be six minutes, or about 20 percent, longer than it actually is.  So it appears that people who live in Los Angeles and Orange Counties have shorter commutes than they should expect, given the very large size of their employment market (and there are large benefits to living in such large employment markets).  

I should note that the scatter plot above looks quadratic, and so I also did the quadratic regression:

The addition of the squared term improves the fit, and so now we can explain almost 38 percent of the variation in commuting time with one variable.  It doesn't change the basic result, however, that LA's commute times are considerably shorter than its size would predict (although its commutes are five, rather than six, minutes shorter than the model would forecast).

So we who live in LA really have nothing to complain about with respect to congestion--we get agglomeration benefits from living in a very large city (Ed Glaeser, Enrico Moretti, Vernon Henderson, Stuart Rosenthal, Will Strange, Jane Jacobs, Alfred Marshall, and many, many others have written about the productivity and consumption benefits of living in a large MSA), and the congestion costs, while real, are relatively low.   

But....this doesn't tell the whole story.  Look again at the scatterplot above, and one sees that the MSA with the longest commute is the relatively small Easton, Pennsylvania.  What is going on here?  Easton is a town immediately to the west of the Delaware River from New Jersey.  It is on Interstate 78, 68 miles to the west of the Holland Tunnel.  It is an inexpensive housing market.  And people commute from there to metro New York.

And so it is with Los Angeles. One place with fewer commuters than metro LA, but a longer commute, is Riverside-San Bernardino, the Inland Empire.  The Census counts people who live there as living in a separate metropolitan area, but they are in fact very much part of the Los Angeles regional economy.  And the people who live there have commutes that are more than 20 percent longer than the model predicts.  People who live out east may get some of the production externalities of the region, but I am not sure, after their exhausting commutes, that they benefit that much from the consumption externalities.

So I, and my neighbors, who live in Pasadena have no reason to complain about our commutes (cost of housing is a different issue).  Those who live in Ontario, on the other hand, have a point.