Which Cities Are The Most Generous?
Recently I decided to come up with a rough measure of generosity across different communities. Which communities have a culture of giving away unused items as opposed to trying to sell them? And why?
One of the ideas was to use Craigslist as a rough measure. Looking at the 23 major cities on Craigslist, we took the number of free items being given away in one week and divided it by the number of items being sold in the furniture category, as a quick index of generosity. In a nutshell: for every 100 items of furniture being sold, how many items are being given away for free?
As you can see, Portland, SF Bay, Boston, and Seattle come up on top in terms of this measurement, with Miami, Phoenix, and Houston on the bottom.
Is this a good metric for measuring generosity across communities?
The Honest Truth About Dishonesty: How We Lie to Everyone - Especially Ourselves
I think it’s quite a good idea to get a first impression but I am not sure whether it serves as a strong indicator. For example people in Houston might simply use another way for giving away furniture for free, for example a local non-profit organization or even a communal institution…
I agree with Nils. And the reason people in Miami, Phoenix, and Houston don’t use Craigslist to give items away is fear. It is safer to call one of the recognized charities to come and haul stuff away. In Portland, they trust everyone. They have all bonded over the amount of rain they get, their love of lattes, and locally sourced food. Mr. Ariely, your measure of generosity is silly but I may buy your book just for fun.
As you have indicated this is far from a perfect measure — the question is if there is any validity and information in this measure
http://dynamicservices.wordpress.com/2011/02/11/t-systems-introduces-cloud-computing-in-south-africa/
I think it captures some of it but maybe a small %. In Bk where I live, you find people giving away things all the time by just putting it on their stoops or in the streets, I reckon more people do that than put it on craigslist (it’s almost a cultural thing). Finally I don’t know that giving stuff away is a sign of generosity, it may just mean you no longer have use for something and you don’t want to horde.
But if a guy looks at the couch he doesn’t want anymore and says to himself, “I wonder how much I can get for this on Craigslist,” isn’t he being greedy? And if greedy is the opposite of generous, he is being generous by default if he is not being greedy. ?
Few comments:
1. is the price of furniture is same in all the areas considered?
2. Type of giving away- is it with every piece of furniture sold/ is there any minimum commitment (no. of pieces/ price) to buy against which ‘free’ stuff is given/ is the ‘free’ stuff given to all types (commercial instituion/ welfare instituion/ a retail customer).
All above details will help in determning the right ranking.
Hope this helps!
Is it not possible that ‘givers’ have a higher propensity to use craigslist than ‘sellers’ in Portland, SF Bay, Boston, and vice versa in Miami, Phoenix, and Houston?
I agree with both Nils and ebon, people might have different ways of disposing of old things. There might be monthly / yearly days when the city takes away whatever people pile in front of their houses. It would be interesting to correlate this with the income and or average age in those cities. I expect a city like SF might be more generous because of many young people coming and going, because of the vibrant tech scene. Also, a city as big as NY is surely going to give more things for free than Austin, just by being 20 times bigger.
The interesting thing is that bigger often doesn’t mean more. For example, Nanowrimo: NYC has more people than Seattle, and had more overall participants in 2010, but produced fewer total words or finished projects. Communities are measurably motivated differently, and to different outcomes, that do not reflect sheer population. It’s an interesting subject.
It may also just be a function of Ave per capita income. For a higher income person, it may not be worth the time to go through the negotiation and hassle factor. They may just want the used furniture out the house fast. The income generated from the furniture would be very small relative to their total income.
I don’t see these stats as an indication of generosity at all. People simply have too much stuff and in today’s economic climate it’s tough to sell it.
But this would not explain the differences between places. Your argument is about the overall level of giving.
Would this have any thing to do with housing markets and average income? Without looking at data, it looks like the bottom cities also are experiencing bad housing markets.
I should point out that the Act applies to private party garage/yard sales and organizations such as the Salvation Army, not just manufacturers.
we already have! here is a better measure and it ranks us as the 2nd most charitable city in america!
http://www.charitynavigator.org/index.cfm?bay=content.view&cpid=1073
Unfortunately there’s a new website phenomenon to confound using Craigslist: Freecycle. Freecycle is sort of like Craigslist, in that it links “buyers” and “sellers” of home goods and assorted sundries in the same geographic area. However, Freecycle items are all free. It’s the first place I would go (other than the thrift store) to give away anything. Perhaps looking at some combination of sales at thrift stores, Freecycle sales, and Craigslist free sales would be a better approximation?
Apparently blog moderation removed my comment containing a URL so here it is without the link.
The Consumer Product Safety Improvement Act went into effect in Feb 2009. By June of 09 it was reported by many mainstream news outlets. No doubt as the news spread it had an effect on the public’s willingness to sell/donate used items.
The Act “increases civil penalties for failure to report possible product hazards to the CPSC in a timely manner from $5,000 per violation (with a cap of $1,825,000) to $100,000 per violation (with a cap of $15 million)” and “increases criminal penalties for various prohibited acts to include forfeiture of assets and imprisonment for up to five years.”
I must agree with DavidK. While this is interesting, I think it’s more likely an indicator of which cities have higher per capita incomes, and/or disposable income. If you make $100,000/year, I’d imagine you’re far less likely to bother selling your old sofa for $50 on Craiglist. If you make $30,000/year, you might be more used to actually budgeting and trying not to waste money. (I freely admit this is a generalization, of course.)
What would be interesting would be to compare this list to the list of cities whose residents tip the best. For example — if I remember correctly — Philadelphia is the poorest of the 10 largest U.S. cities, but its residents average the highest restaurant tips at more than 20%. Granted, you’re still limiting yourself to examining people who can at least afford to go out to dinner, but it might add a new dimension to the analysis.
I don’t think it tells anything, you would need a lot more info to figure out if that statement were true.
What about the cost to have your old things carted away to a landfill or to a thrift store. People may give stuff away to avoid that charge, especially if they don’t have a vehicle to cart that stuff around. Then, you wouldn’t necessarily call them generous.
Or, maybe the lower ranking cities-folk like having yard sales and don’t put the low ticket items on craigslist, becasue they think yard sales are fun.
or people try to have only serious people come for their stuff (tell someone there’s free stuff and all sorts come out of he woodwork – and don’t show up when they say they will)…
As a few of the others have noted there is likely a large X-factor related to the geography and what is “standard” in those communities.
As a quick example – if I lived in a non-walking city (e.g. not New York where very few people own cars) and was trying to give away a large piece of furniture I might be more likely to try and sell it (since actually conveying the property will itself incur some kind of cost on my (or the recipient’s) part due to car rentals etc.). In a city where it’s assumed most people own a car (and in some cases a large vehicle (truck/SUV) I’d be more likely to give away since the actual process of giving it away would be more seamless.
Re the Act, mentioned by Joseph, I have not seen any substantial impact on the local second-hand store stream. They’ve all posted warning notices of Buyer Beware, although the children’s toy after-market was affected for a few months while they figure it out.
Re the original question, I’d say it’s a pretty non-useful measure. You really need to control for the length of time CraigsList has been available in that town.
Then, look at the competition for “stuff”–towns with very active Habitat for Humanity chapters WORK the donation / collection of used stuff for resale. Areass with less active Habitat / AMVets / Goodwill / etc chapters may find more free stuff on Craigslist. People may prefer to donate to the “deserving” poor served by Habitat and Goodwill than self-serving poor, looking for freebies.
In some places, people simply don’t want anyone coming into their house to pick up “free” stuff, so you’d have to control for recent burglary reports. It’s a pretty good way to case the joint. I have friends who gave up on holding yard sales because they didn’t want strangers seeing the quality of the stuff they were getting rid of.
In other words, nice try, but try again.
I will
I would like to know what are the percentages of Democrats vs. Republicans in these cities. Would there be a correlation?
Thank you for th belly laugh Juimmy! AFter seeing responses on many political sites, I am thinking perhaps they should rename your parties the selfish and unselfish parties.
Furniture in Boston/Cambridge, where I’ve lived for 10+ years, is often given away when 25% of the population who are students must leave and are desperate to rid themselves of bulky items. Used furniture is often sold at steep discount to the original price paid, so the marginal dollars a seller might receive is usually not perceived to be worth the extra hassle and time it takes to cull through interested buyers.
I would re-run this experiment with items that retain more of their values (e.g. electronics, etc) and are not dependent on volatility in transient citizens.
I think you may have just discovered the cities with the most junk that no one would dare pay for…
Fun post — congrats on being Freshly Pressed!
There are at least two confounding factors:
Like someone pointed out, it first depends on how tech savvy the city is. And the second factor is the size of the student population. These actually go hand in hand. Students are more likely to take to the internet for all of their day-to-day needs and wants because it’s part of the culture of being young. At the same time, students often have to uproot quickly and may not have the time or energy to deal with the hassle of selling things. And let’s be honest, most students get hand-me-downs anyway so the stuff they’re giving away is not exactly quality.
You need to control for quality of merchandise and likelihood of posting on craigslist. You didn’t mention during what time period these data were collected. If it was from the end of last fall semester, students might have been desperate to get rid of some of their crap before moving to a new dorm or apartment, which would at least account for Boston and NY being near the top of the list.
Best of luck, Dan! I’m a big fan of your work and I hope you figure out more about the behavioral economics of generosity. And I hope we can all try to be more generous.
Nara
look at it again in the summer and see if it holds up…8 of the top 9 cities are enveloped in winter.
This isn’t about generousity – it is an indication of how lazy a city is when it comes to throwing things about. People don’t give away furniture because they think that someone else will like it, they give it away so that they don’t ahve to haul it to the dump themselves!
I agree with Mitch. This seems more a measure of laziness than generosity. Measure instead the quantity of donations to thrift shops, homeless shelters and charities. Even then there’s still an element of self interest (tax deduction) over pure generosity.
i love checking the free sections on craigslist. im from chicago and the list isnt so generous =[ but if im ever in portland or SF area ill make sure to check that section out. good post!
Fantastic comments…several about perspectives that had not occurred to me. Joanne’s for example might explain upticks in donations/sales that occur as students move in/out each semester.
Karen, posted warning signs aren’t enough. Those businesses are still liable. Onus is on the business not the buyer. It seems the impact hasn’t been felt perhaps because they’re not taking the new legislation seriously. And it still doesn’t address the impact on private party sales. Nonetheless, I was a bit surprised to learn that the impact in your local area was low.
Read Mitch Leuraner’s comment above. It is the truth.
CL’s freebies list is a good measure some ways. But I believe it’s also a function of economics — can people afford to just give away their furniture? Or do their circumstances require that they receive payment. Some of the top cities are the richest ones the country. But then again, their our outliers like Detroit.
The internet is never a good determinant of any social phenomenon. Firstly, despite large growth in access, daily/habitual usage is still limited to select demographics.
Secondly, you’re using one website, and one that’s not particularly user friendly. Sure, it gets a lot of hits, but that’s no reason to assume that it’s representative. You’re lack of study into the history of craigslist also belies the top contendors for “generosity:” Portland, SF Bay, Boston, and Seattle – All areas where craigslist has been a long-time major website. Compare that to New Orleans or “Jersey Shore (i think that’s what they’re callng it)” and you’ll see significantly less traffic and even less representative user base.
Like really? This is a middle-school level “project.”
What a great idea to measure through craigslist. I am not sure if it is the most scientific measure, but since I California came in second, I think it is right on the money. -Thank you.
Furniture alone might not be a good measure, but i can testify to the generosity of Queens, NYC.
I had a great bar-b-que last summer, and though we didn’t have seating for it at first, a quick check of “Free” in CL had us headed out in each compass direction to pick up free bits of furniture from friendly, generous people, until my backyard was the biggest/best living room/dance floor of all time.
That night, in a drunken wildly smiling haze, I dragged it all back out to the curb, slapped a crude post on CL with my camera phone and my broken english, and by the morning sunrise the stuff was gone…
Gotta love it when a plan comes together – especially when it’s made possible by the awesomeness of a community @ large!
~J
I noticed is some of the top cities (San Francisco, Seattle) tend to be hilly, while some at the bottom of the list (Houston, Las Vegas, Phoenix) are flat. Maybe people living in the former, where moving furniture is a bigger hassle, are more likely to choose the “It’s yours if you haul it away” option.
What about a premium on space? Most of the cities that show up higher on the list are known for having smaller living spaces. If you get a new piece of furniture you might simply not have the luxury of keeping the old piece around while you wait to find a buyer, weighing the ability to move around your studio without crawling over an extra table the twenty dollars you could get for it.
Cool post!
I think it has to do more with culture than generosity. I think someone already pointed out the political divide between those cities at the top (tend to be liberal) versus those at the bottom (tend to be conservative). However, I’m guessing there are other cultural issues here. “giving” here means stuff like Freecycle and Craig’s list where as “giving” in a place like Houston might mean participating in a neighborhood organization or tithing at church. There’s also a high degree of disposable income in some of these areas and a concentration of young, highly educated people which could also contribute to the differences.
In other words, I don’t believe this is a good indicator of generosity, only a good indicator of how likely people are to give away their furniture rather than sell it.
It is not a measure of generosity.
The list merely correlates to the ubiquity of internet usage.
Craigslist is internet based.
Around here, people just put free stuff at the curb. It is gone in days. No need for internet. This happens more in neighborhoods with high income diversity and those that are more densely populated.
your comment form needs to be redesigned. The comment box is BELOW the “your comment”, while the name box is ABOVE “name (required)”.
Consistency creates clarity, please.
Being from San Francisco, I can attest that San Franciscans are very friendly, haha. So, from my not-so-scientific personal experience, I concur with this survey.
Damn, I’m from Phoenix and I was going to look on Craig’s List, rats!! Ha, ha!!
I wonder how those numbers correlate with with reported cases of bedbugs.
Someone may have pointed this out, but the communities at the top all seem to have active craigslist user bases. The communities at the bottom have smaller ones. People that have free furniture to give away in those places may not know that craislist exist as an option to do that.
I live in a rural area that has had a Freecycle list since 2005, but a local Craigslist for only two. People are much more likely to have heard of Freecycle and use it to offer their unwanted furniture. The local Craigslist has a lot more spam and a lot fewer users.
I would have used Freecycle.com based on a number of listings being posted.
I also considered this but if you use Freecycle.com, you don’t have any number to compare it to — it is just the level of usage of Freecycle.com. here we have both the free and a non-free category so we can measure the partial effect of free. Anyway, we will try more ways.
The coining of this indicator is quite questionable so will the usage of this ratio. So far I can only view it as a measure of proportion of freebies listed on CL to the # of furniture selling on CL. Other than that I can’t see how far one can stretch it beyond what it is or towards a more general measure of genorosity.
Like many has pointed out first the give-aways could go through multiple channels such as donation, curb dump, pass-along, etc. This implies the numerator is severely underestimated, biased downwards. CL listing represents a fraction of the overall volumes of freebies. Maybe the first step is to identify the overall volume of give-aways and establish a relationship between that to the CL listing. This will greatly enhance the precision of measurement.
On the other hand whether the volume is a good measure remains question as well. The value of freebies could be to the point of not worth the effort of selling. How should the value of items be factored into this equation would be another black box to crack open before the answer to “generosity” can be properly attained.
It sounds like an interesting research question
. What about monetary donations? So many aspects seem to be missing. Going backwards maybe the original question is too broad in nature and needs to be narrowed to be remotely measurable!
1 four-legged yellow wooden stool – Porto, Portugal.
=) jk. People are generally quite generous around here.
I Like the idea but I think it might be a little to rough, being that Houston was just named as one of the top 5 giving cities in America, like Nils said we just don’t use Craigslist, instead we have many places that accept donated furniture and they are always packed with new items weekly, The Salvation Army, Habitat for Humanity Store, and the furniture bank just to name a few.
Couches can be hard to sell and a lot of people just give them away for free because it’s easier than trying to figure out how to get rid of it. Kind of like mattresses.
to be more specific Houston is currently ranked #2
http://www.charitynavigator.org/index.cfm?bay=content.view&cpid=1073
I crunched some numbers and found an interesting relationship. I looked at the correlation between Household income and % free furniture and found an amazingly strong positive correlation… R-Squared = 0.77 (after taking out 2 outliers of BOS and Portland). So this somewhat supports my earlier hypothesis that with a higher income person, it may not be worth the time to go through the negotiation and hassle factor. They may just want the used furniture out the house fast. The small amount of money they would get for the furniture is minuscule compared to their income.
Naturally, its always shakey ground to distinguish correlation from causation.
I’d send the analysis, but don’t have an ability to send file on this forum
These numbers could be compared to the rate of bed bugs each city has.
uTKRaV gkbkwbbqxrju
bew2Ll olotyzdtwuvc
I’m not a statistician and can’t comment much on the metrics. But as far as topic of generosity goes this is an interesting post. To give a bit more insight into the different aspects of generosity I would suggest reading the book Rambam’s Ladder: A Meditation On Generosity And Why It Is Important To Give by Julie Salamon. It may shed some light on the influence that culture and socioeconomic status have on generosity. Just a thought.
I feel like you need to also bring into the list how many people give money to charities or give away things to salvation army or the like. There are a lot of different ways to give.
DavidK, the high income also means they buy a lot, and therefor have a lot more to get rid of.
Also need to take into account the ages and generations in the area. My own family is just now starting to give furniture away to strangers; up until the past couple years it’s mostly been handed down to the next generation.
Craigslist numbers are going to be skewed by recent crime rate, particularly in situations where a highly publicized crime has been linked to the service.
Interesting idea, but there are many other measures of generosity. Charitable contributions and volunteer work are also important indications. Furthermore, some generous people wish to remain unknown.
Thanks for the comments. I realized that this is not a perfect measure (far from it) but I did not consider all the problems that people have suggested — so this was useful and educational for me.
It might also suggest that I should stick to things i know more about (experiments)
best
Dan
Portland surprised me, but still great. Houston step up your game!
San Francisco Bay Area REPRESENT!!
Another reason why I am proud of being a born and bred Bay Area chick.
Great post! Congratulations.
People are highly educated in these areas, live in smaller spaces, and are more likely to be transient. Therefore, they are more tech savvy, live with less and are happy to give it away in order to move to another location. Generosity = convenience.
Being a born Houstonian, I definitely don’t think this is a good representation of my people.
There are many God loving people in Houston and more than likely they either donated it to a local goodwill/purple heart/shelter/church. Not that God loving people don’t exist anywhere else, just that Houstonians may not be the most courteous drivers but they are nice people. Many people will even put old furniture on the street outside their house with a “Free” sign on it and someone who really needs it will likely pick it up. I wouldn’t doubt it if there were some paranoid people in Houston that are cautious of another “Craigslist-Killer.”
Have you ever tried scouting for curb alerts in the dead of summer? I think people in the hotter south are just too afraid to do any heavy lifting perchance they die of heatstroke in a matter of minutes.
hmmm.. interesting approach. Just a suggestion, maybe you can count the number of non-profit organizations and divide it by the population or size of city, because they exist only if there are enough people participating in the program.
My immediate thought was that in places like the South, preference would be toward giving to church, charity, goodwill over craigs– a practice with a long history, coupled with short craigslist timeline. So would expect the gradient would be west to east and north to south, so to speak.
Yet Los Angeles is at the bottom! Also places like Raleigh & Atlanta have long had large craigs communities (& certainly Raleigh was an early adopter, yes?) And how is it NY & DC is so much higher than Chicago?
I think this does give an interesting snap shot– since it catches how a community is using craigslist to move things around (don’t forget that there’s a barter section, too! which adds a pretty wrinkle!).
Now the income correlation might be sticky: could track higher end products- antiques or tvs? or car makes? to compare user incomes at different craigslists? Figuring the free/sale ratio for more craigs sites might make for really intriguing groupings that could perhaps tell more about role of income, student population, etc.
love your work, the blog & the bottomless curiosity! keep it up!
What you do is measure the size and activity of the “hippie/social/alternative” segment (poor nomenclature, I know), plus perhaps other people who believe in free sharing of goods: big in portland and SF, small in Houston, TX (sorry if this offends or is seen as politically incorrect).
Great post Dan and congratulations on being Freshly Pressed! As a few others have mentioned there are probably some confounding factors which limit the analysis. You might be interested in looking at individual non-cash contributions from IRS data although I am unsure if you could get the data at the state or city level and that source of info also has some confounding factors attached.
Also, while charitable giving is certainly a measure of generosity I always feel that volunteerism is probably the best measure — not to be cliche but time really is the single resource we have that we can never get back. Choosing to consistently spend that resource in the service of others rather than in front of the TV, hanging with friends, or at the bar probably says the most about someone’s willingness to give of themselves. There are some interesting state level volunteerism statistics put out by the CNCS each year: http://www.volunteeringinamerica.gov/
Thanks for a very interesting post!
congrats on freshly pressed. I love your theme!
Congrats on FP~ I’d be curioust to see the political links between these cities. Republican? Democratic? Moderate? Or is there a connection? What about a religious connection? Christian? Athiest?
Keep on, keep on.
Caitlin
http://www.collectivedisclosure.wordpress.com
In Houston, I’ve never found the need to put unwanted items on Craigslist. ANYTHING I don’t want I put on the curb and it’s usually gone within the hour. Except books! There are also a LOT of garage sales here. And then there are the organizations that are always looking for donated items. I have no idea how generous Houstonians are but that’s just my 2 cents about free stuff on craigslist.
I would just like to point out to many of the commenters that suggest alternate ways of counting generosity that any method that counts donations to churches (e.g. IRS charitable donations) ought not to count because many people tithing to their own religion are not doing it out of “generosity” but other motivations (e.g. buying their way into heaven? fear of hell? just keeping their local place of worship open for their own use on Sundays?). I think a lot of the other studies that measure “charitable giving” in the US end up with huge biases towards the Bible Belt because of this.
This is a good start, but you might include freecycle.org to your sources for a quick overview. Thanks, Dan, for aking our input!
I agree with a lot of the comments above. However, I still like the direction this measurement is going in, since it’s measuring generosity in a non-monetary way. A lot of studies seem to use financial donations to judge generosity because it’s an easy metric (e.g. Conservatives donate more money to non-profit groups and are therefore more generous). But using craigslist, though imperfect, covers an extra dimension of generosity that merits more analysis. While this data set has its problems, I think it can help measure generosity along with other variables.
HOLLE, DAN,I come from China .I like your books very much,and also like behavioral economics.But my english very bad,I can’t chat with you ,and read more about your blog.But I will often come here.
I am not sure this means anything. I have lived in Houston and Massachusetts, and I feel pretty confident when I say Houstonians are no less generous than Bostonians.
This probably has more to do with the huge amount of pickup ownership in Houston, and the ease with which one can get around town.
In SF, it is probably an enormous challenge to sell furniture to all the prius drivers, especially when the value of the furniture is far less than the cost of renting a van. I would just give it away too. In Houston you will easily find people who will come to your house with their truck and buy a couch for $25.
The larger the pool of potential used furniture buyers, the greater the demand. The smaller the pool, the more people are forced to give away what they don’t want. It’s not generosity – it’s new live-in girlfriends forcing crap bachelor furniture out of the house by any means necessary.
I don’t think you’re going to find anything useful this way. Consider: the cities on the bottom half are all in much warmer weather, and it’s winter now. How would you account for the difference in results that might cause? More importantly, though, the question of whether you’d give away furniture is not a test of generosity so much as of circumstances that vary by city and region: living space, frequency of turnover in rentals, truck ownership rate, culture of curbside giveaways, popularity of Craigslist, characteristics of population that uses Craigslist, student or seasonal population, and so forth. On all of these factors, the top and bottom ranked cities have obvious differences that will confuse any results you hope to find with this method.
Hmmmm … cites in blue states at the top, cities in red “Christian” states at the bottom. Just sayin.
This may have been pointed out already, but the top 5 cities have large populations of young, educated, people (i.e., heavy internet users) that live in small spaces and move a lot. (Certainly true in my part of Seattle.) Also, I bet population density and car utilization factor into this a great deal.
Well Dan now you know where to sk a question! Who needs google or brainstorming when you have commenters of this caliber?
While I am certain you aren’t going to use your criteria to make a generosity absolute statement, I have one more thing to add. people also give away to garage sales and events that are for charity.
We also have an odd tradition in our town whereby businesses are solicited for their donation to charity events, wedding social giveaways and family days in the community. That tradition alone generates a whole lot of good will that is passed on (as there are no highly visble ads to make it a viable method of advertizing)
I am also with those who said volunteerism is a far better guage of generosity.
This is driven by age, but it’s a bit more nuanced than a lot of folks here (myself included) expected. I downloaded age distribution data from the census website (Table 22 at http://www.census.gov/compendia/statab/cats/population.html). It’s not perfect, since the demographic data pools metro areas (NYC and Northern NJ, for example). I also had to combine LA and Orange County, for this reason, so I just split the difference and gave the LA/OC area a score of 2.2.
The first thing to note is that plugging the age data into an ordinary least-squares regression yields an r^2 of 0.92, which is remarkably high. The two most notable correlations with craigslist score are size of the under-18 population (r=-.80) and the 45-64 population (r=.72).
The equation the OLS spits out is CL_Score = -67.4 – 0.08*Under18 + .96*18to44 + 1.25*45to64 + .72*65to74. OLS has plenty of flaws, but if it’s fitting the data as well as this does, it’s working pretty well.
My best guess is that this is driven by families. People in cities that have a lot of families (which will have a lot of Under 18′s) are more inclined to sell things, perhaps because of culture, perhaps just because of a need to feed the extra mouths. Cities with a lot of 18-44′s or 45-64′s, but without a lot of children mean more people who are more independent or mobile, and who may be more inclined to give stuff away in a hurry as they pick up and move somewhere else.
Whatever the case, since the relationship with age distribution is *so* strong, I think we can dismiss the notion that this has anything to do with generosity or any other value particular to the people of a given city.
Seth,
Nice analysis. Not sure if you saw my earlier post:
I looked at the correlation between Household income and % free furniture and found an amazingly strong positive correlation… R-Squared = 0.77 (after taking out 2 outliers of BOS and Portland). So this somewhat supports my earlier hypothesis that with a higher income person, it may not be worth the time to go through the negotiation and hassle factor.
So, maybe there is a bit of multicolinearity. e.g. Ave age also likely correlated with income
David,
Totally missed that, but that makes sense. I suspect that you’re right that part of it has to do with income. I will say though, I’m struck by just how strong the correlation is with the age distribution, and without having to get rid of outliers too. It’s hard to tease the two apart, but I suspect they might work together–someone who has money can avoid the hassle of negotiating, unless they have kids, in which case they might really need the money. Something like that perhaps?
Obviously I think the age is hiding some sort of underlying causal factor, which includes income, but given how much extra variance it soaks up, I suspect there’s something more going on, as well.
Thoughts as a Portlander: the ranking above appears to have a rough inverse relationship with how car-centered the cities are: the top seven are cities where many people rely on public transportation, bicycles, or walking; the bottom seven are cities where driving is essential.
Pete Cognoscenti, above, may be on to something in mentioning the role of car ownership, but he has the relationship backward.
Where neither party to a potential sale owns a vehicle suitable for moving furniture, the transaction costs will be substantial: if truck rental is part of the total cost to the buyer, he will not be willing to pay as much for the furniture itself.
On the other side, if the seller can’t haul the furniture somewhere himself, his options for disposing of it are limited; he may be glad to have it removed for free instead of paying someone to take it.
Not the only relevant factor, but I suspect that’s part of the reason for the numbers you found.
I think a lot of this may have to do with the degree to which the city sports a highly transient population.
Cities that attract highly skilled, highly educated young singles are going to feature lots of people moving in and out of the community, and moving from one apartment to another while there. I lived in DC and nearby areas for several years and this was basically my life: moving from place to place about once a year. Every time I moved, some things had to go, and other things had to be added to my collection of stuff. The more things that people are going to try and get rid of, the more things that will end up in the “free” section of craigslist, largely because there just won’t be the demand for so much used stuff in any given community.
On the other hand, cities where folks tend to settle down, buy a house, etc, are going to be places that are less mobile, less transient, and where fewer people are constantly shedding stuff on craigslist. If you look at the rankings of the various cities on your list, it pretty much comports with this analysis. Cities like NYC and Seattle are places that 23-year-olds go to find themselves after college. Cities like Houston are places where 33-year-olds go to settle down and buy a house.
You seem to presume that the alternative to given it away on Craigslist is selling it on Criagslist. But there’s another: throwing it away. Rather than generosity, this ratio might measure a community’s commitment to re-use and re-cycling.
As already noted the average income seems to be highly correlated with the craigslist numbers presented above. Hence if we consider the direction of causation (income -> waste) we will just conclude that the free/furniture indicator is a just proxy of the avg. income. Hence those numbers don’t measure generosity but income. You will definitely need to condition on income and see how it goes. (but do you know whether the craigslist popularity is uniform across the states?)
I can’t speak for the trailers (Houston, Phoenix, Miami), but one distorting effect in Philadelphia is the presence of a rival web site Philly Freecycle, dedicated solely to giving stuff away. I’ve never used craig’s list, but I’ve given away a ton of stuff (probably considerably more, in literal terms) on Philly Freecycle.
No, your method of evaluating generousity is only one out of dozens or thousands. I donate via an organized service that collects donated goods (like Goodwill). I also live in a city and if I leave desirable items in the alley, people collect them. Additionally, donations are given through formal campaigns and personal networks.
I strongly disagree with Dan’s premise. To determin the degree of philanthropy of a location using a furniture give away criteria in no way identifies how charitable that location is. Cleveland, considered on a population basis, is far more charitible than most of the cities mentioned. And this level of philanthropy extends back decades and is not simply determined during single narrow point in time.
Does that statistic really measure generosity? I suggest there may be a relationship between the numbers other data of those cities, such as unemployment rate, physical size of metropolitan area, proportion of residents who are college students etc.
See here for examining numbers and statistics critically (non-technical article): http://comopolis.blogspot.com/2011/02/how-to-examine-numbers-and-statistics.html
Dan,
Wow great post and amazing set of comments….too much to reply to but just wanted to let you know about Generosity Day (rebooted Valentine’s Day) that we launched very quickly and very successfully starting 3 days before….feels like there’s a real hunger for being more generous out there, if we just give people permission.
Here are some URLs or you can just search for #generosityday on Twitter. Would love your thoughts on it!
http://sashadichter.wordpress.com/2011/02/14/generosity-day/
http://www.fastcompany.com/1727145/re-booting-valentine-s-day-for-good
Hi Dan,
No, I do not think this is necessarily an accurate measure of generosity although it might be slightly correlated with generosity . Let me explain myself with an example. Let’s say I have a piece of furniture (inexpensive) that I could try to sell or give away. If I was to try to apply some of your research, I would say that one of the best indicators of what my choice would be, would be that which other people with my same type of item have done before. If people in my area had given that item away, I would probably give it away and on the other hand if people have a similar item posted for sale then I would post mine for sale. I would in a sense be picking the default option selected by other people. Therefore using this logic, what the indicator would really measure was if the first couple of people on craigslist in my area were generous or not which may or may not be related to my specific city being generous or not.
P.S. I am currently a student in the Duke CCMBA program. Would you please consider teaching an elective for us in the fall
?
Thanks,
Felipe
Regarding determining the most generous cities, check out the recent information from Convio. http://www.convio.com/convio/news/releases/convio-ranks-most-generous-1.html At least from an onling giving perspective, we offer some insights to which communities are giving generously.
This simple test obscures the likely real issue. That is, used furniture is likely to have a value that is less than its disposal cost. In other words, furniture has two cost components….intrinsic value and transport cost. When transport cost >= intrinsic value, the exchange price cannot be positive. So, in places where transport costs are high, you would expect more exchanges at zero price. At first blush, the cities at the top of the list are coastal cities with high costs of furniture transport. A better test of generosity would be to look at giveaways of cheaply portable goods.
As a footnote, I’d like to know more about the underlying data. These percentages are close enough that it is also that the data are insufficient to draw statistical inferences.
How do you qualify generosity? Me and couple look at the free category just to get a good laugh since most is junk, and big.. Just something that was difficult to dispose of. Wouldn’t there need to be a quality level that would define if some is actually generous?
Granting one mans junk, is another treasure.
Free is a function of the cost of disposal.
If it costs you something to dispose of an item, because the city charges your for removal or you need to use private removal services, free is the option you choose.
If there are no charges for furniture removal, because the city, for example, picks up furniture with the trash, then you charge or use the private market.
Consider costs of disposal.
After study few of the blogposts on your site since yesterday, and I truly like your style of blogging. I tag it to my favorites site list and will be checking back soon.
Is it not possible that ‘givers’ have a higher propensity to use craigslist than ‘sellers’ in Portland, SF Bay, Boston, and vice versa in Miami, Phoenix, and Houston?
Re: Dan Ariely » Blog Archive Which Cities Are The Most Generous? « thanks, I enjoyed reading it, got an insight or two!
First off I would like to say excellent blog! I had a quick
question which I’d like to ask if you don’t mind. I was curious to find out how you
center yourself and clear your thoughts before writing. I’ve had a hard time clearing my thoughts in getting my ideas out. I do enjoy writing but it just seems like the first 10 to 15 minutes are usually wasted simply just trying to figure out how to begin. Any suggestions or hints? Kudos!