Unnatural deaths and apartment prices

If you were looking to buy a house, and you found out that someone had recently died an unnatural death in the house, would that affect what you are willing to pay for the house? I suspect it would for many people, being a negative characteristic of the house. Based on hedonic demand theory, which my ECONS102 class briefly covered this week, negative characteristics reduce the overall price of the good. 

How does that work? Hedonic demand theory recognises that when you buy some (or most?) goods you aren't so much buying a single item but really a bundle of characteristics, and each of those characteristics has value. The value of the whole product is the sum of the value of the characteristics that make it up. For example, when you buy a house, you are buying its characteristics (number of bedrooms, number of bathrooms, floor area, land area, location, etc.). You are also buying the characteristic of whether the house has had a recent unnatural death, or not. When you bundle all of the characteristics' values together, you get the value of the house.

So, does an unnatural death really reduce house prices? That is the research question addressed in this 2018 article by Zheng Chang (City University of Hong Kong) and Jing Li (Singapore Management University), published in the journal Regional Science and Urban Economics (ungated earlier version here). They use housing unit (mostly apartments, I guess) sales data from Hong Kong housing estates between 2001 and 2015, and first note that:

Influenced by Taoism, traditional Chinese believe that people who died as a result of violence or unnatural events can become “ghosts” who can disturb successive occupants through various means... Housing units in which unnatural deaths have occurred are called “haunted units,” which are regarded as bad Feng Shui, and are unsuitable for habitation...

So, the expectation is that a recent unnatural death will reduce house prices. Comparing houses with and without a recent unnatural death (and controlling for housing unit characteristics), Chang and Li find that:

...housing values drop about 25% for units with deaths, 4.5% for other units on the same floor, 2.6% for other floor units in the same building, and 1% for units in other buildings of the same estate. However, the average house price of units in other estates within 300m increases 0.5%. For units with deaths, the price decline is sustained across the whole study period. The price impact on units in other geographic scopes follows a U shape and starts to reverse after 4–5 years of a death.

In other words, there is evidence for a sustained negative impact of an unnatural death on housing unit prices, as well as a shorter-term impact on surrounding housing units. Even having a housing unit on the same floor, or in the same building, as the unit where there was a recent unnatural death, is enough to lower the housing unit's price. Sometimes, superstition matters for consumer preferences.

Retail marijuana stores and house prices

Does having nearby retail stores affect the value of homes? Surely it does, but just as surely it depends on the type of retail stores. If your neighbourhood has a lot of payday lenders, pawn shops, and discount liquor stores, that is quite different from a neighbourhood that has health stores, pet stores, and premium wine shops. So, one way of measuring whether a community prefers to have more (or less) of particular retail stores is to measure the effect on house prices. If, when a new retail store opens, local house prices increase, then the community believes that store is a good thing. If, on the other hand, local house prices decrease, then the community believes that the store is a bad thing.

This relies on hedonic demand theory (or hedonic pricing), which recognises that when you buy some (or most?) goods you aren't so much buying a single item but really a bundle of characteristics, and each of those characteristics has value. The value of the whole product is the sum of the value of the characteristics that make it up. For example, when you buy a house, you are buying its characteristics (number of bedrooms, number of bathrooms, floor area, land area, location, etc.). As part of the location, you are buying the fact that there are a number of retail stores of different types in the local neighbourhood. So, controlling for all of the other characteristics of houses, comparing the price of houses in areas with some types of retail stores with similar houses in neighbourhoods without those same types of retail stores, provides one way of determining how the community views those retail stores.

As part of my ongoing research on alcohol outlets, I have toyed with the idea of performing this analysis for house prices and alcohol outlet locations in New Zealand. However, there are a bunch of other outlet types that we may be interested in, like vape stores. And, if New Zealand ever gets around to legalising marijuana, it would be interesting to see the effect of retail marijuana stores on house prices.

That's more or less exactly what this 2020 article by James Conklin (University of Georgia), Moussa Diop (University of Wisconsin-Madison), and Herman Li (California State University), published in the journal Real Estate Economics (ungated earlier version here, and research brief here), did. Conklin et al. look at what happened to house prices in Denver when Colorado legalised retail marijuana on 1 January 2014. Interestingly, the Colorado policy change allowed existing medical marijuana stores to become retail stores. So, the analysis doesn't compare retail store with no store, so much as what happens when an existing medical marijuana store becomes a retail store. However, this is important, as:

...since only existing medical marijuana stores were allowed to conduct recreational sales, we avoid the potential endogeneity of store location. Given the opportunity, retail marijuana stores would likely choose to locate in certain areas based on neighborhood characteristics that would also affect house prices. However, since only existing medical stores were allowed to sell retail marijuana, the siting decision was made before implementation of [legalized recreational marijuana].

Conklin et al. use a difference-in-differences approach, comparing the difference in house prices (controlling for various house characteristics) between single-family houses within 0.1 miles of at least one retail marijuana store and single-family houses between 0.1 and 0.25 miles of at least one retail marijuana store, before and after the legalisation of retail marijuana. Their initial analysis, based on properties sold in 2013 (before the law change) and 2014 (after the law change) finds that:

...being located near at least one medical facility that converts to retail in 2014 is associated with an 8.4% increase in sale price after the retail conversion occurs.

You may be worried that the number of retail marijuana stores in the neighbourhood matter, but Conklin et al. also show that the results are similar for houses where exactly one store became a retail marijuana store in the local neighbourhood (in fact, a slightly larger 11.4% increase). The results are also robust to alternative definitions of the control groups, but expanding the treatment group to include houses further away from retail marijuana conversions reduces the effect (which should be no surprise - those houses were in the control group in the original analysis).

Overall, we can conclude that local communities in Denver like retail marijuana stores. Or, more correctly, local communities in Denver prefer retail marijuana stores to medical marijuana stores (since we don't know how they feel about the medical marijuana stores, and house prices went up after those stores convert to retail). This is consistent with other evidence (as I discussed in this 2018 post), showing that legalising marijuana increased house prices across all of Colorado (rather than just Denver) by about 6 percent. The downside of this, of course, is that if you are in favour of legalising marijuana in New Zealand, one trade-off may be that house prices go up even further than they already have.

Read more:

• Legalised marijuana sales and house prices

More on the value of words in wine descriptions

I posted last month about consumers' willingness to pay for wine bullshit. That was based on research by Kevin Capehart, which was published in the Journal of Wine Economics. Now, Capehart has another new article published in the same journal on a related topic (sorry, I don't see an ungated version online). In this new article, Capehart follows up on earlier research by Coco Krumme (described here). Essentially, Capehart uses a dataset of 120,000 wine descriptions, and classifies them into 'high price' (over US$50) or 'low price' (under US$15). He then trains a Naive Bayesian Classifier to predict which wines belong to the low-price category and which belong to the high-price category, based on the words in their descriptions.

Capehart is mostly able to reproduce very similar results to the earlier Krumme work, but perhaps more interestingly:

...I find for the dataset studied here that there do seem to be two mostly distinct vocabularies for high- and low-priced wines. Out of the roughly 20,000 unique words used to describe the over-$50 and/or under-$15 wines, only 42% of those words overlap by being used to describe wines in both price categories. The remaining 58% of words are non-overlapping.

In other words, the descriptions of high-priced wines use very different words than the descriptions of low-priced wines. You might argue that is because high-priced wines have different characteristics than low-priced wines, their descriptions should include different vocabularies. However, the key question that isn't answered (and which Capehart alludes to in his conclusion) is, does the vocabulary relate to the quality of the wine, is it simply a signal of the price? In other words, do those who are writing the descriptions choose their vocabulary based on the price of the wine, or based on the quality of the wine? We'd need more research in order to answer that question, and more like Capehart's earlier work.

Read more:

• The willingness to pay for wine bullshit

The willingness to pay for wine bullshit

Consumers often can't tell the difference between two similar substitute products. I've blogged previously about bottled water, but people can't even tell the difference between dog food and pâté (ungated earlier version here). In the bottled water research, people couldn't match different bottled waters to their descriptions. That may be because water descriptions are mostly bullshit - after all, this article by Richard Quandt (Princeton University) notes that wine descriptions are mostly bullshit too.

That brings me to this recent article by Kevin Capehart (California State University), published in the Journal of Wine Economics (sorry, I don't see an ungated version online). Capehart looks at the bullshit wine descriptors that were described as bullshit in Quandt's article (descriptors such as 'silky tannins', 'velvety tannins', 'brawny', and a flavour of 'smoked game'). He then uses various methods to estimate consumers' willingness to pay for bullshit. Specifically, Capehart employs three methods:

I start by using a hedonic regression similar to regressions used by previous studies on wine prices and descriptions...

The second method uses the same dataset used for my hedonic regression, but I draw on approaches for matching rich texts... in order to obtain matching estimates of consumers’ MWTP [marginal willingness-to-pay] for select descriptors. My third method is a stated-preference survey in which I directly ask approximately 500 wine consumers about their MWTP for select descriptors.

For the hedonic regression model, Capehart draws on the descriptions in various online wine catalogues, leading to a dataset of over 51,000 wines. Looking at the effect of different descriptors on wine prices, he finds that:

Despite their joint significance, many of the descriptors have effects that are not statistically significantly different from zero at conventional levels. Examples of descriptors with statistically insignificant effects include “silky” and “silky tannins.”...

Some descriptors do have effects that are statistically different from zero. Of the 106 descriptors, 43 have effects that are significantly significant at the 10% level. Yet, some of those statistically significant effects are not substantively significant. For example, the effect of “velvety tannins” is statistically different from zero (p-value = 0.03), but it is arguably small at only 2.5% (se = 1.1%). A 2.5% change is less than a $1 change for any bottle under $40.

So, some descriptors are associated with higher priced wines, and some with lower price wines. However, mostly the effects are small. Unfortunately, overall the hedonic regression model poses more questions than it answers. As Capehart notes:

If those hedonic results are momentarily accepted, there is much to puzzle over. Why would consumers be willing to pay so much more for wine if an expert described it in terms of the smoked game? How much more or less would they be willing to pay if the expert described the game as being prepared differently, such as by roasting, steaming, or boiling? And what if the expert described the game as a specific type of animal such as a pheasant, boar, deer, squirrel, or some elusive or imaginary creature that few if any have tasted? Questions abound.

Capehart then moves onto using a text-matching estimator:

Any matching estimator tries to match subjects who have received a treatment to control subjects who are as similar as possible, except they did not receive the treatment. After matching, the effect of the treatment on an outcome of interest can be estimated by comparing the outcomes of the matched subjects. Here, the “subjects” are wines, the “treatment” is whether a given Quandt descriptor appears in a wine’s description, and the outcome of interest is the wine’s price.

He essentially uses the 'bag-of-words' approach, which really means noting whether each description contains one or more words (that is, the actual context of their use is ignored). This analysis basically compares wines with very similar descriptions, one of which contains the particular descriptor and one of which does not. In this analysis, Capehart finds that the results are:

...generally consistent or not inconsistent with my hedonic estimates.

Ok, so again wine consumers appear to be willing to pay for some descriptors. Why not ask them about it? That's what the final analysis does, based on a stated preference survey of 469 US wine consumers, conducted online. Essentially, each research participant was asked to choose between two wines, with different prices and descriptions. Based on those hypothetical (stated preference) choices, Capehart finds that:

...most consumers have a zero or near-zero MWTP for velvety rather than silky tannins; that would be consistent with “velvety” and “silky” being synonyms and not inconsistent with my hedonic and matching estimates that suggested at most a small price premium for velvety over silky tannins.

Overall, across the three methods, Capehart concludes that (emphasis is his):

One conclusion is that most consumers seem to have little if any MWTP for wines described by most of the Quandt descriptors. My hedonic approach suggested the majority of the descriptors have a price premium of zero or near-zero. My matching and survey estimates were generally consistent or not inconsistent with my hedonic estimates, at least for the select descriptors considered.

The other conclusion is that some consumers have a non-zero MWTP for wines described by some of the Quandt descriptors. The hedonic approach suggested some descriptors have price premiums (or discounts) that are significant in the statistical sense and arguably significant in the substantive sense. My matching approach suggested the same. And my survey approach suggested some expert and novice wine consumers are willing to pay more than nothing for some descriptors.

In other words, most wine consumers are not willing to pay anything for wine bullshit, but some consumers are. If wine descriptors are mostly bullshit (as Quandt claimed), why would any consumers be willing to pay for wines that have those descriptors? That is the question that Capehart doesn't answer. Perhaps those consumers that are willing to pay a positive amount for a particular descriptor, are willing to do so simply because they feel better for knowing that they are consuming something that has been described as having 'velvety tannins' or the flavour of 'smoked game'? We don't know, so more research will be required in order to uncover the answer to the question of why.

Sea level rise, coastal flooding, and house prices

In my ECONS102 class last week, one of the things we discussed was hedonic pricing - the idea that the price of some goods (such as houses or land) reflects the sum of the values of all of the characteristics of the good. In the case of property, if the property includes a dwelling, the price will reflect the quality and size of the dwelling, number of bedrooms, bathrooms, whether it has off-street parking, and so on. But the price also reflects the access of the property to local amenities, such as good schools, public transport, and so on (for example, see this post from 2017), as well as the property's risks of damage due to environmental disasters such as earthquakes or floods.

In the case of risk, properties that have a higher risk profile should have lower prices - a higher risk profile is a negative characteristic for a property. Two new research articles provide some relevant evidence.

First, this article by Allan Beltran, David Maddison, and Robert Elliott (all University of Birmingham) published in the Journal of Environmental Economics and Management (sorry I don't see an ungated version), looked at the impact of floods on property prices in the UK. They used data on over 12 million property transactions and nearly 5 million properties over the period from 1995 to 2014. Interestingly, their method looked at 'repeat sales'. That means that they essentially looked at property's prices before, and after, a flood event. Some properties were directly affected by flooding, while others weren't. They found that:

...in the immediate aftermath of inland flooding the average price of property in a postcode entirely inundated is 24.9% lower. For incidents of coastal flooding the corresponding figure is 21.1%. These results moreover emerge from a comparison of inundated and non-inundated properties all within the floodplain. Such discounts are however short-lived; property affected by inland flooding typically recovers after 5 years and in just 4 years for coastal properties. The time for price recovery differs markedly for properties in different price-quartiles. For properties affected by coastal flooding in the highest price-quartile, the property price discount disappears after only 1 year whereas for properties in the lowest price-quartile the discount remains statistically significant for up to 6-7 years.

So, floods reduced house prices, but the prices rebounded so that there was no net negative effect within several years. Interestingly, the effect was slightly lower for coastal flooding, and disappeared quicker. That is, people were quick to return to demanding coastal property soon after coastal flooding. That should be a bit of a worry to us, given that sea level rise is likely to be one of the enduring effects of future climate change.

Which brings me to the second article, by Asaf Bernstein (University of Colorado at Boulder), Matthew Gustafson (Pennsylvania State University), and Ryan Lewis (University of Colorado at Boulder), published in the Journal of Financial Economics (ungated earlier version here). This article provides more direct evidence on the effect of sea level rise on house prices, using data from over 460,000 property transactions of properties in the US that would "be inundated following a 1-6 foot increase in average global ocean level". Their analysis is not based on repeat sales, and neither is it based on actual sea level rise (it is projected future sea level rise). The latter point means that, if there are negative impacts on property prices, then buyers are factoring in future sea level rise in their decisions about buying. They find that:

...SLR exposed properties trade at a 6.6% discount relative to comparable unexposed properties. We further break this into exposure buckets, with properties that will be inundated after one foot of global average SLR trading at a 14.7% discount, properties inundated with 2-3 feet of SLR trading at a 13.8% discount, and properties inundated with 4-5 and six feet of SLR trading at 7.8% and 4.4% discounts, respectively.

Interestingly, it is non-owner-occupiers would are more likely to apply a discount to the property:

We find that the SLR exposure discount is concentrated in the non-owner occupied segment of the market. On average, exposed non-owner occupied properties trade at a 10% discount, relative to comparable non-exposed proper- ties, while exposed and unexposed owner occupied properties trade at similar prices.

In other words, owner-occupiers likely underestimate the negative impacts of sea level rise on their homes. They also found that owner-occupiers with stronger beliefs regarding climate change did apply a discount in buying coastal property.

These two papers, taken together at face value, should probably worry anyone who is concerned about the future impact of sea level rise and coastal flooding on people living near the coast. Coastal property is at risk in many (perhaps most) areas. Holding all other factors constant (such as the quality of housing, access to amenities and services, etc.), the value of these properties should be decreasing relative to less vulnerable property (or at least, not rising as quickly). It appears that is not the case, and in fact following flood events (which should make it abundantly clear to potential purchasers that these properties are vulnerable to coastal flooding and sea level rise), property prices are rebounding quickly to their previous levels. On top of that, it appears that it is owner-occupiers (and in particular climate-change-naive owner-occupiers) who will face the brunt of these future impacts.

I don't know that this leads to a strong case for regulation of coastal property in some way, but at least it suggests that coastal property owners (and potential buyers of coastal property) must become better informed about the risks. The specific vulnerability of coastal property to inundation and flood events probably needs to be communicated to potential buyers for every coastal property transaction.

Red light districts and house prices

Following on from Monday's post on legalised marijuana sales and house prices, it is reasonable to consider whether other legal or quasi-legal but controversial activities also have impacts that can be picked up in house prices. A recent working paper by Erasmo Giambona (Syracuse University) and Rafael Ribas (University of Amsterdam) considers the case of red light districts (RLDs) in the Netherlands.

This is a nice paper, and they exploit a change in city policy in 2007 that aimed to reduce the number of red light windows, as well as considering the geography of Amsterdam, where canals provide plausible breaks in the geography that can be used to identify the effects of red light windows, where prices change abruptly. This can be seen in the maps below, where the background colours represent house prices, the black dashed lines are the edges of the two main RLDs in Amsterdam, and the thick blue lines are the canal borders of those districts.

The maps seem to suggest that house prices are higher outside the borders of the RLDs up to 2006 (on the left), but less so from 2007 onwards (on the right). The comparison between the period before the change in red light windows and the period after is important. It means that the analysis isn't confounded by the access of properties to other amenities (that didn't change between the period up to 2006 and the period from 2007 onwards). Indeed, the authors show that the distribution of restaurants, bars, and coffeeshops did not change appreciably between those two time periods.

In their key results for Amsterdam, Giambona and Ribas find that:

...homes next to prostitution windows are sold at a discount as high as 24%, compared to similar properties outside the RLD...

They find similar results using alternative methods, and similar results for Utrecht, where the red light districts were closed entirely in 2013. Specifically, for Utrecht:

 ...we find that households paid up to 1.5% of their property value to be 100 meters further away from the RLDs.

What is the mechanism that underlies the negative impact of red light districts on house prices? Giambona and Ribas find that around half or more of the price difference relates to crime:

To understand the type of nuisance related to prostitution, we also investigate the change in crime rates after the downsizing of RLDs in both cities. In Amsterdam, the crime rate in the RLD declined by 18% relative to other parts of the city. Yet half of the house price discontinuity remains unexplained after controlling for all forms of reported crime and misbehavior. In Utrecht, the crime rate near the RLDs declined by 11%, which represents more than 300 crimes per year. While property crimes and violence can explain up to a third of the price effect in Utrecht, changes in drug-related crimes and minor nuisances explain almost all variation in house prices triggered by the end of the RLDs.

The results are clear - red light districts impose negative externalities on surrounding homeowners, and those externalities can be measured in terms of their effects on house prices. Crime is not the only negative externality that is present (at least for Amsterdam), so presumably red light districts also create other disamenities for their neighbourhood. As recent experience in New Zealand has suggested, for home-based brothels.

[HT: Eric Crampton at Offsetting Behaviour, back in January]

Legalised marijuana sales and house prices

Does the community believe that legalising marijuana sales a good thing? There are both benefits and costs associated with legalising marijuana sales. Benefits might include easier access for recreational users (and higher consumer surplus), job opportunities in the marijuana sector, savings on policing and justice costs, and increased tax revenues (if marijuana sales or profits are taxed). Costs might include adverse impacts on public health, decreased productivity, and increases in crime. How do we weigh up these benefits and costs?

Hedonic demand theory (or hedonic pricing) recognises that when you buy some (or most?) goods you aren't so much buying a single item but really a bundle of characteristics, and each of those characteristics has value. The value of the whole product is the sum of the value of the characteristics that make it up. For example, when you buy a house, you are buying its characteristics (number of bedrooms, number of bathrooms, floor area, land area, location, etc.). You are also buying the bundle of local regulations for the area the house in located in, which includes whether marijuana sales are legal. So, controlling for all of the other characteristics of houses, comparing the price of houses in areas where marijuana sales are legal with the price of houses in areas where marijuana sales are not legal, provides one way of determining how the community views the balance of benefits and costs of marijuana sales.

And that is exactly what a new paper by Cheng Cheng, Walter Mayer (both University of Mississippi), and Yanling Mayer (FNC Inc.), published in the journal Economic Inquiry (sorry I don't see an ungated version), does. Overall, I like the approach of evaluating through the effects on house prices. I've blogged before on its use in terms of the effects of radiation following the Fukushima nuclear disaster, sunshine, and proximity to strip clubs.

In this case, Cheng et al. outline the logic for why differences in marijuana regulations would affect house prices:

As home buyers and sellers respond to changes in local amenities and disamenities... the associated benefits and costs of the public programs, such as legalizing retail marijuana, are capitalized into housing values. However, the net effect on housing values is ambiguous ex ante given the opposing effects of the benefits and costs. For example, on the one hand, the benefits of retail marijuana legalization potentially raise housing values by either increasing housing demand (e.g., attracting more home buyers) or decreasing housing supply (e.g., discouraging homeowners from selling their properties and moving). On the other hand, the costs have the opposite effects on demand and supply and, therefore, potentially lower housing values. Thus, this paper estimates the net effect of legalizing retail marijuana on housing values, which reflects the net capitalization of the benefits and costs by the housing market.

They use data from 91,943 house sales in Colorado over the period 2010 to 2015. Over this period, 46 out of 271 municipalities in Colorado chose to adopt legalisation, while the others did not. They find that:

...on average legalizing retail marijuana in Colorado increases housing values by approximately 6%, or $15,600 per property, which can explain about 27% of the overall housing price appreciation in adopting municipalities during the examination period.

Importantly, their results demonstrate that the change in prices occurred right after marijuana was legalised, so it is unlikely that other contemporaneous changes in the housing market or regulatory environment explain the results (especially since different municipalities adopted legalisation at different times).

So it appears that, in Colorado at least, the benefits of legalisation of marijuana sales exceed the costs, because house prices in those areas adopting legalisation moved higher in response to legalisation. However, some caution is warranted before over-interpreting these results. While the benefit-cost evaluation seems to come out in favour of benefits, the size of the change in house prices may not be replicable everywhere. It is notable that Colorado was one of the first two states in the U.S. to adopt legalisation, so to the extent that some people would move to areas where marijuana sales are legal, the demand side of the market has already corrected. States that were later to adopt legalised marijuana would be unlikely to see jumps in house prices to the same extent, because any pent-up demand for living in an area with legalised marijuana sales has already been satisfied.

Radiation and house prices after the Fukushima nuclear disaster

How much less would you be willing to pay for a house in an area affected by radiation significantly above background levels, compared with an otherwise-identical house that is unaffected by radiation? It's not a crazy question. In the U.S., hundreds of millions of people (including the populations of 26 of the 100 most populous cities) live within 50 miles of a nuclear reactor. Worldwide, there are 21 nuclear plants that each have more than one million people living within 30 kilometres of them.

Of course, nuclear accidents are thankfully rare. But the risk is not zero, and when an accident does occur, as happened in Fukushima in 2011, people can be understandably reluctant to live in the affected areas due to the risks to their health and wellbeing. Obviously, that has a flow-on impact on house prices, even outside the most heavily affected areas.

Hedonic demand theory (or hedonic pricing), which we discussed in my ECONS102 class last week, recognises that when you buy some (or most?) goods you aren't so much buying a single item but really a bundle of characteristics, and each of those characteristics has value. The value of the whole product is the sum of the value of the characteristics that make it up. For example, when you buy a house, you are buying its characteristics (number of bedrooms, number of bathrooms, floor area, land area, location, etc.). When you buy land, you are buying land area, soil quality, slope, location and access to amenities, etc. You are also buying the exposure to current levels of radiation, as well as the risk of future exposures to radiation in the event of a nuclear accident. If each of those characteristics can be separately valued, then you can place a value on how much people are willing to pay to avoid radiation (or alternatively, how much they are willing to accept to live in a radiation-affected area).

In a new paper in the Journal of Regional Science (sorry I don't see an ungated version anywhere online), Alistair Munro (National Graduate Institute for Policy Studies, Japan) looks at the impact of the Fukushima disaster on house prices in Fukushima and Miyagi prefectures, using data from 2009 to 2017. Fukushima prefecture was most affected by radiation as well as the tsunami that led to the nuclear disaster, while neighbouring Miyagi prefecture was only affected by the tsunami. So, differences between the two in terms of changes in house prices can be attributed to differences in radiation levels (once you control for other characteristics of the properties, of course). He finds that:

...across the subsample of noncondominium residence types a 1 percent rise in radiation leads to a 0.051 percent drop in values, while for condominiums treated separately the elasticity is also 0.051. For housing land the elasticity is 0.044, and 0.032 for land with existing buildings if the age of the building is controlled for.

In other words, areas more affected by radiation have lower house prices. How much lower? Munro reports that:

...using a variety of methods... the impact of radiation translates into a one to two million Yen (US$10,000–20,000)... reduction in housing prices for average residential properties.

That is quite substantial, but is not terribly surprising. However, the next part of the paper is very cool. Having established how much less people are willing to pay for living in an area with more radiation, Munro then uses that information plus information on the risk of cancer arising from environmental radiation, to estimate the value of a statistical life (or VSL).

As I will discuss with my ECONS102 class later this semester, the VSL can be estimated by taking the willingness-to-pay for a small reduction in risk of death, and extrapolating that to estimate the willingness-to-pay for a 100% reduction in the risk of death, which can be interpreted as the implicit value of a life. Munro estimates VSL to be in the region of US$4.5-6.4 million, which is similar to VSL estimated in other studies (and other risk contexts). An additional take-away from that analysis is that there isn't a particularly high element of dread associated with avoiding death from radiation (otherwise, people would be willing to pay more to avoid it, and the estimated VSL would be much higher).

Next we really need to know whether the Fukushima disaster affected people's perceptions of nuclear risk in other areas that are near nuclear plants but which weren't affected by the disaster. That would be much more difficult to establish, but potentially much more interesting.

World has just destroyed any premium value for their 'Made in New Zealand' clothing

Following the fallout from the World Made in New Zealand saga (where the New Zealand fashion label World was caught out selling t-shirts with "Fabriqué en Nouvelle Zélande" labels when those shirts were actually made in Bangladesh and Hong Kong), the New Zealand Herald ran a good story on why we're willing to pay for 'Made in New Zealand':

But why would anyone pay $99 for a T-shirt which, it turned out, was materially no different from one sold for a fraction of that price?

When we pay a premium for retail items it's because the branding for that item convinces us it contains some intangible benefit, says Dr Sommer Kapitan, a senior marketing lecturer at Auckland University of Technology.

"Before we knew there was a question about that brand we had this quirky, artistic premium New Zealand fashion brand."

A World shirt was not just a shirt for someone who values artistry or quirkiness, but an expression of those values, Kapitan said.

And the target market for World and most other designer labels were people willing to pay a premium to stand out.

"I [a designer label fan] might not want to see myself as someone who wears $30 stuff. Whether you can tell or not, I want to know that I wear a $100 T-shirt," Kapitan said.

World's decision to write Made in New Zealand in French on its swing tags was an example of the way brands would use symbolic cues to communicate worth to consumers, Kapitan said.

This story, and the explanation, calls to mind two important (and related) concepts from economics. The first concept is hedonic demand theory (or hedonic pricing, which I have written about earlier here in the context of education). Hedonic pricing recognises that when you buy some (or most?) goods you aren't so much buying a single item but really a bundle of characteristics, and each of those characteristics has value. The value of the whole product is the sum of the value of the characteristics that make it up. In the case of a t-shirt from World, you are not only buying a t-shirt, but you are buying something else. As well as the garment, you are buying: (1) the 'warm glow' feeling that you are supporting New Zealand clothing manufacturers; and/or (2) an image that wearing that t-shirt allows you to project to the world ('Look at me! I'm wearing this t-shirt that was made in New Zealand. Aren't I a great person?'). So, people are willing to pay a premium for a World t-shirt that has a 'Made in New Zealand' tag for one or both of those reasons.

That brings me to the second concept - conspicuous consumption (which I've written about earlier here). People engage in conspicuous consumption as a form of signalling - they want to signal to other people the type of person that they are (or the type of person that they want other people to think they are). A signal is only effective if it has two characteristics: (1) it is costly; and (2) it is costly in a way that makes it unattractive for those with 'low-quality' attributes to attempt (in this case, it would have to be unattractive for people who aren't the type of person who buys New Zealand-made to pretend to be that type of person).

With a $100 World t-shirt (with a 'Made in New Zealand' tag), the first characteristic is assured. There is a premium for the 'Made in New Zealand' label, which makes the t-shirt more expensive than a regular t-shirt made in Bangladesh. What about the second characteristic? Assume that there are two types of people: (1) those who really do buy New Zealand-made because they want to support local industry; and (2) those who buy things with 'Made in New Zealand' labels because they want to be associated the first group, even though they don't feel that strongly about it. The first group is (probably) willing to pay a greater premium for 'Made in New Zealand' than the second group, but this can't be assured. Even if the extra cost of buying 'real New Zealand-made' clothing is high, at least some of the second group will not drop out of the market. The signalling value of 'Made in New Zealand' is therefore pretty weak.

The weakness of the signal (and therefore the conspicuous consumption value) of buying 'Made in New Zealand' is also clear because the 'Made in New Zealand' tag is hidden inside the t-shirt. If you wanted people to know that you are the type of person that buys New Zealand-made, you'd want to project that to the world, which is difficult to do if the tag is hidden. So the only way you could present that signal is therefore to buy from a designer where all of their clothing is New Zealand-made (then you don't have to show the label). And this is where World has clearly gotten things wrong. Any signalling (or faux-signalling, given the weakness of the signal) value is going to be lost from World clothing, now that we know they aren't really selling New Zealand-made t-shirts.

Although, that does still leave the 'warm glow' from buying New Zealand-made. But in order to be willing to pay a premium for the 'warm glow', customers have to believe that the 'Made in New Zealand' tag is credible. And World's credibility has surely taken a huge hit. Why would you believe that their clothing is New Zealand-made based on the tag, when the tag has been shown to be worthless in this case? And this is a seriously weak response from World:

But World co-owner Dame Denise L'Estrange-Corbet told Newstalk ZB tags sewn into the garments said "Made in Bangladesh", and stated they were sourced from AS Colour, so it was not misleading customers.

L'Estrange-Corbet said only a small percentage of her products were manufactured overseas.

She said: "99 per cent of our clothing is made here."

So, if the signalling value (which was limited anyway) of World t-shirts has declined, and the 'warm glow' value has declined due to a loss of credibility, that leaves World in a seriously difficult position.

School zone is not the whole story of house price differences

From Friday's New Zealand Herald:

Property buyers seeking houses in zones for certain public schools can expect to pay a premium up to 90.5 per cent on homes in the wider area, new data shows.

Homes.co.nz released data comparing the median regional price to the median price for different school zones to find a "school price premium". The prices are the estimated value calculated by Homes.co.nz and are updated monthly.

The median estimate price for property in the Auckland region is $940,610, according to Homes.co.nz.

Houses zoned for Epsom Girls' Grammar School have a median price of $1.79 million, an increase of 90.5 per cent on the Auckland average.

Now, I don't doubt that the price of homes in the Epsom Girls' Grammar School zone are 90.5 percent higher than the Auckland average. However, it would be wrong to describe this as a "school price premium", because school zone is not the only difference between those houses and the median house in the Auckland region. Houses in good school zones might also be in nicer neighbourhoods, closer to amenities (like the CBD), and with better access to services. All of these neighbourhood-level variables are important for house prices - they also have value for home buyers and home owners. Those variables also vary systematically by neighbourhood and are likely to be correlated with the location of good school zones.

On top of that, houses in those areas might also be larger, have more bedrooms and bathrooms, have better views, and differ on many other house-level dimensions that contribute to house prices. Or they may not. The point is that none of those potential differences are controlled for by the Homes.co.nz analysis.

So, while there is almost certainly a price premium for good school zones (for example, see this 2014 working paper by my colleagues John Gibson and Geua Boe-Gibson), it almost certainly isn't as much as 90.5 percent for the Epsom Girls' Grammar School zone. Once your subtract the value of the good neighbourhood, access to amenities, etc., then the marginal value of the school zone is likely to be much less than that. But at least Homes.co.nz didn't claim that the whole price of the house was the price of living in a good school zone!

Sunshine, the value of housing and compensation for externalities

In ECON110 today, we discussed hedonic demand theory (or hedonic pricing). Hedonic pricing recognises that when you buy some (or most?) goods you aren't so much buying a single item but really a bundle of characteristics, and each of those characteristics has value. The value of the whole product is the sum of the value of the characteristics that make it up. For example, when you buy a house, you are buying its characteristics (number of bedrooms, number of bathrooms, floor area, land area, location, etc.). When you buy land, you are buying land area, soil quality, slope, location and access to amenities, etc.

In a new Motu working paper, David Fleming, Arthur Grimes, Laurent Lebreton, Dave Maré, and Peter Nunns show that sunshine is one of the important characteristics that contributes to house values. The New Zealand Herald reported a couple of weeks ago:

Motu Economic and Public Policy Research Trust has released what it calls the first research carried out anywhere in the world to specifically evaluate the extra value house buyers put on extra sunshine hours.

Arthur Grimes, a senior fellow at Motu and co-author of the study, said there was a direct correlation between more sunshine and higher values and the study was precise about how much extra value is added.

"Direct sunlight exposure is a valued attribute for residential property buyers, perhaps especially in a cool-climate city such as Wellington. However, natural and man-made features may block sunlight for some houses, leading to a loss in value for those dwellings," the study said.

The effect is quite large. Quoting from the paper:

...each additional hour of direct sunlight exposure for a house per day (on average across the year) adds 2.4% to a dwelling’s market value.

The paper also has some interesting implications in terms of negative externalities. If a high-rise apartment development will block the sunlight from nearby houses, then it will reduce the value of those houses. This constitutes a negative externality imposed on the affected homeowners. Fleming et al. note that these externalities could be dealt with through compensation:

At a policy level, our estimates may be used to facilitate price-based instruments rather than regulatory restrictions to deal with overshadowing caused by new developments. For instance, consider a new multi-storey development that will block three hours of direct sunlight exposure per day (on average across the year) on two houses, each valued at $1,000,000. The resulting loss in value to the house owners is in the order of $144,000. Instead of regulating building heights or the site envelope for the new development, the developer could be required to reimburse each house owner $72,000. In return, the developer would be otherwise unrestricted (for sunlight purposes) in the nature of development. If the development cannot bear the $144,000 then the efficient outcome is that the development does not proceed. Conversely, if the development can bear that sum, then the socially optimal outcome is for the development to occur and, from an equity perspective, the neighbours are compensated for their loss of sunlight exposure.

The idea that compensation can be used to deal with externalities relies on the Coase Theorem - the idea that, if private parties can bargain without cost over the allocation of resources, they can solve the problem of externalities on their own (i.e. without government intervention). In the case of a bargaining solution to an externality based on the Coase Theorem, the solution depends crucially on the distribution of entitlements (property rights and liability rules). In this case, the homeowners have existing rights to sunlight and because an apartment development would infringe on those rights, the developer would be expected to pay compensation to the affected homeowners. This will only be viable if the total amount of compensation paid to affected homeowners is not so great that it makes the development unprofitable.

The study was based on data from Wellington. Given that development in Auckland is happening faster and involves increasing density and greater numbers of taller mixed-use buildings, it would be interesting to see if the results hold there as well. As noted in the New Zealand Herald story:

"For places other than Wellington, the value of sunshine hours may be higher or lower depending on factors such as climate, topography, city size and incomes. Nevertheless, our approach can be replicated in studies for other cities to help price the value of sunlight in those settings," Grimes said. 

So the approach is transferable, even if the results are not. It's almost certainly extendable to considering the value of volcanic viewshafts in Auckland, and hopefully someone is already thinking about undertaking that work.

Strip clubs, externalities, and property values in Seattle

Property values tend to reflect not only the characteristics of the property itself, but also the neighbourhood that the property is located in. This is hedonic pricing - the price of a property reflects the sum of the values of all of the characteristics of the property. If the property includes a dwelling, the price reflects the quality and size of the dwelling, number of bedrooms, bathrooms, whether it has off-street parking, and so on. But the price also reflects the access of the property to local amenities, such as good schools, public transport, and so on (for example, see this post from earlier this year).

But not all local amenities are positive. Some features of the neighbourhood might create disamenity, reducing property prices. One example may be strip clubs. If a strip club attracts unsavoury people and petty (and not-so-petty) crimes, then fewer people will want to live in that neighbourhood, reducing demand for properties in that area and consequently reducing property prices. Another way of thinking about this is that the strip club creates a negative externality on local property owners (an externality is the uncompensated impact of the actions of one party - in this case the strip club locating in a particular neighbourhood - on others, in this case the local property owners).

There is evidence to suggest that some facilities do create disamenities that negatively affect property prices, including meth labs and toxin-emitting industrial plants. But what about strip clubs? A recent working paper by Taggert Brooks (University of Wisconsin - La Crosse), Brad Humphreys and Adam Nowak (both West Virginia University) looks at relevant data for Seattle.

Specifically, Brooks et al. looked at repeated property sales (where the same property was sold multiple times) over the period 2000-2013, a period during which a moratorium on new strip clubs in King County (which includes Seattle [*]) was removed. Using repeated property sales gets around the problem of accounting for the different quality of different properties (provided you assume that property quality doesn't markedly change between sales). Their dataset included over 317,000 property sales, of which about 5,400 were within 2000 feet of a strip club.

What did they find? A whole lot of nothing. In their preferred specification of the mode, the results:

...indicate that the presence of an operating strip club is not associated with any differential in residential property prices over this period. These results indicate price dynamics for those properties within K of an operating strip club are no different from price dynamics for properties between K and 1 [mile] of a strip club.

There did appear to be some weaker evidence that condominium prices were lower when a strip club was nearby though:

However, the results using the condominium sub-sample, and the single family home sub-sample, provide weak evidence that strip clubs are associated with residential property price differentials in some cases... condominiums located within 1000 feet of a strip club have transactions prices about 5.5% lower than condominiums located farther from operating strip clubs. Some weak evidence also suggests that condominiums within 500 feet also sell for lower prices...

These results are interesting, but are based on only a small amount of variation in the sample. If I read the paper correctly, there were only 370 properties that were sold multiple times, where there was a nearby strip club at the time of one of the sales and no nearby strip club at the time of the other sale. So, given the small number of 'identifying observations', I'd be much more cautious than the authors about interpreting the lack of statistical significance here as suggesting that strip clubs have no effect on property values. I would be more inclined to say that they may have an effect, but this study didn't have sufficient statistical power to detect the effect. Although statistically insignificant, the point estimate of the effects from their preferred specification suggests that property prices are 6.5 percent lower when there is a strip club within 500 feet, 2.9 percent lower within 1000 feet, and 1.6 percent lower within 2000 feet. That is quite a large effect.

It would also be interesting to see if similar results obtain for other cities in the U.S. and elsewhere. It also suggests to me that we could use a similar approach to evaluate the negative effects of alcohol outlets in New Zealand. Something to follow up later.

[HT: Marginal Revolution last year]

*****

[*] Yes, as I mentioned in an earlier post I was in Seattle a couple of weeks ago and no, it wasn't to collect observational data on strip clubs. My wife was with me and can attest to the lack of strip clubs in our itinerary.

The price tag for a top decile Auckland education is not $2m+

I meant to write a post on this a while back, when Corazon Miller wrote in the New Zealand Herald:

Parents wanting to send their children to some of Auckland's prestigious state schools may be forced to fork out more than $2m to buy a house in the zone, or more than $700 a week in rent - double the cost ten years ago.

Figures from property analysis site Relab.co.nz which analysed 33 Auckland school zones showed buying a home was more costly close to higher decile schools.

Topping the list were two decile nine school zones, Auckland Grammar and Epsom Girls Grammar - with median values topping $2m last year...

Relab marketing director Bill Ma said the figures showed parents what they should be budgeting for.

"The higher decile schools definitely come with a premium price."

Yes, homes in good school zones do have a premium for the school zone. But no, the premium is not the whole price. The reason is simple - if you weren't living in a house in the double Grammar zone, you'd have to be living somewhere else, and presumably that somewhere else is not free. To find out the actual premium for school zoning, you have to look at two otherwise-identical houses: one in the zone; and one outside the zone.

This relies on the concept of hedonic pricing - an idea that was first introduced by Irma Adelman, who sadly passed away in February. Hedonic pricing recognises that when you buy some (or most?) goods you aren't so much buying a single item but really a bundle of characteristics, and each of those characteristics has value. The value of the whole product is the sum of the value of the characteristics that make it up. For example, when you buy a house, you are buying its characteristics (number of bedrooms, number of bathrooms, floor area, land area, location, etc.). When you buy land, you are buying land area, soil quality, slope, location and access to amenities, etc.

In the case of Auckland, school zone is only one of many characteristics that make up the value of the house. And it is possible to separately identify and value those characteristics if you have good data, as my colleague John Gibson and his wife Geua Boe-Gibson did in this 2014 working paper using data on 8000 houses in Christchurch. They didn't look at school zones, but looked at the value of school outcomes (measured by NCEA pass rates), and found that a standard deviation increase in school performance raises house prices by 6.4%.

So overall, the 'price tag for a top decile Auckland education' won't be $2m or more. If we (generously) assumed that the premium was the difference between the $2.09 million median in the Auckland Grammar school zone mentioned in the article above, and the $1.05 million median across Auckland as a whole (as of today), then the premium is a little over $1 million. However, that wouldn't be correct, since the median house in the Auckland Grammar zone is clearly not the same as the median house across all of Auckland, as well as having access to different amenities (besides the school zone). So the premium for the Auckland Grammar zone is likely to be substantially lower than $1 million.