Archive for the ‘ Fruit ’ Category

How Irrationality Got Me a Great Dog

The whole notion of low hanging fruit is predicated on there being some easily obtainable yet still unexploited gains from trade to be had. A few months ago, I unwittingly took advantage of what I can only call a massive asymmetry in the market for rescued dogs. The irrationality of perspective dog rescuers has granted a considerable amount of leverage to anyone willing to consider a breed with “bull” in the name. Quite predictably, (free) media sensationalism of very low-probability events has led the public to forego obtaining any further information on pit bulls (at cost) and rendered them systematically biased against associated breeds to an extreme degree. Happily, readers of this blog seeking to rescue an active, medium-sized dog will have considerably greater selection power than other market players.

First, observe the strength of the bias against Staffordshire/Bully breeds. As my aim is only to show that prospective adopters have a huge systematic bias against bully breeds, I’ll keep the comparison limited to German Shepherds and Rottweilers; Unfortunately, there’s no case to be made for pit bulls being less dangerous than beagles or golden retrievers. Still, given your confidence in your ability to select an individual dog, you may be able to get the best value of any dog in a breed that’s being adopted at a rate 11 to 13 times lower than breeds equivalent, if not more costly, in terms of expected damages due to aggression.

How equivalent? Very. On a sliding scale, this study gave pit bulls a 9 for stranger (human) aggression and a 29 for dog aggression. German Shepherds registered 13 and 26 with Rottweilers scoring 10 and 16. For reference, the most docile breeds scored 1-3 for stranger aggression and 3-6 for dog aggression. Consider further that injuries to humans are much more socially and materially costly than damages to canines and that studies (including that cited) consistently show that there is high intrabreed variance in aggression. This latter fact means greater selection power on the part of the adopter greatly increases the chances of finding a dog below its breed’s aggression mean. And, given the growing number of pit bulls immediately euthanized on arrival to shelters, much of that work has already been done for you by more knowledgeable parties. If people were only selecting against aggression, we’d have ample reason to expect pit bulls being adopted at higher rates than Shepherds and Rottweilers.

There are obviously other explanations for peoples’ resistance to adopting bully breeds, but these are mostly matters of taste and, of course, I can’t argue about taste. I can, however, argue about highly irregular distributions of taste not being about taste at all, but rather about irrationality.

Speaking of which, anyone not compelled by the data can indulge themselves in this:



My Ride-sharing Soap Box

This post will be longer than I’d like most of my posts to be, but I imagine this is one cost of starting a blog a considerable amount of time after being born. I have a lot of old ideas I would like to reference, but since I’ve developed them elsewhere, the blog needs to be caught up before I can develop them in a more incremental, and palatable, manner.

Last month, Bryan Caplan posed this question to his students: how to spend a billion dollars most efficiently and, alternatively, in such a way as to maximize utility. Part of the idea was undoubtedly to distinguish utility from efficiency by proposing plans that differed in their espousal of each principle. Another part was to see whether students could actually identify any low hanging fruit: is there something we obviously could and should do that we haven’t already done?

My proposal would be that we pay ~8 million Americans to become casual cab drivers. Here it is important to qualify “casual;” no one is leaving his job to drive people around, in fact, they’ll be asked to go no more than a few minutes out of their way. Smartphones have made it incredibly easy for people to impersonally coordinate time and locations, why not pay someone a trifling amount to take you to work when you’re right on their way?

There are two general reasons why it wouldn’t work, one preferential, the other structural. The preferential case revolves around general insularity (stranger danger, social awkwardness, misanthropy) and is more difficult to predict as it’s unclear at what prices people would be willing to shrug it off. The structural case is simply that a widespread and reliable ride-sharing network needs a big push and it’s hard to see how you could get the gains from such a scheme to accrue to the source of the push. I think a billion dollars simultaneously remitted to to a few million active participants in a handful of major cities would be sufficient to get the project off the ground and its evident convenience would sustain it from there.

There are a litany of potential benefits. Most immediate is the relief of traffic jams in major cities, perhaps to the extent that it eliminates the need for congestion charges which are a net social gain, but privately costly in terms of time/money/planning. Secondly, its short term labor labor-saving effects would free up cabbies and public transportation workers/dollars while providing people with a lower cost service across the board (t/m/p). The long-term gains are by far the largest. Ownership of private passenger vehicles is the second largest expense most Westerns face in their lives, in many cases approaching one third of the median income (consider these figures for the US). If there were a convenient, reliable way for commuters travelling in the same direction to ride-share, I’m keen to entertain estimates of how many active vehicles could be taken off the road in 15 years.

In the interest of quantifying the efficiency gains, I’ll say that we can reduce the number of passenger vehicles by 15% in 15 years (Ceteris Paribus). There were 255 million registered passenger vehicles in the US in 2007. In the interest of conservatism, I’ll suggest 265 million as a current figure. At roughly $9,000 per car per year the annual expenditure hits $2.835 trillion. 15% of that?  ~$358 billion per year. Say my plan has a low probability of success, or the reduction in car ownership wouldn’t be so great, how low would it have to be and how small would the change have to be? Even if there were only a 3.5% chance of the strategy reducing the number of cars on the road by 1%, you’d still get your principle back in savings every year.

While it may not be the most efficient way to spend $1 billion, it is perhaps the one most highly relevant to the everyday lives of Americans. Also it’s much more easily quantifiable than the promotion of natalism, unspecified scientific breakthroughs, or immigration.

I think I’ve done enough to answer Bryan’s prompt and don’t wish to make this post any longer, though I have much more to say on the topic. For those curious about how far private money has gotten in this endeavor, look in to Lyft and Uber, two services being developed while I’ve been day dreaming.

My Advice For WordPress

This blog is getting off the ground some 18 hours later than it might have because of what can only be labeled perverse and widespread risk-aversion on the part free blog-hosting websites.

At one point or another, almost everyone reading this will have considered starting their own blog and indeed many have followed through. My question is: how far have you followed through? In my cursory and completely anecdotal searches last night and this morning I stumbled upon 8-10 blogs with very compelling addresses (like, none of which would be considered by a reasonable observer to be “active” or even valued by their authors. Untouched for years and only barley touched in their time, these blogs waste away, unread and unattended. Their addresses, however, remain very desirable and many a new blogger is forced to settle for their second, third or fourth choice, if not some horrible bastardization of their first choice. From the very reasonable conjectures that this situation creates a marginal disincentive to launching new blogs and lowers the profile of blogs with less than optimally resounding addresses, we can say that this arrangement is quite undesirable for hosts as well as bloggers.

The rub is then of course why this circumstance persists throughout the blogosphere. Certainly, it would be in the interests of hosts to define the conditions of dormancy for users from the outset and enforce accordingly by freeing up addresses to those who will use them more fruitfully. Imagine how much revenue a free bloghosting site is leaving on the table by unconditionally leasing its hosting rights to whatever addresses are available in perpetuity, rather than enacting a Coasian solution for blogs that have breached a dormancy clause…

I’ll pay $5 for the rights to I doubt we’ll get a bid from our man inside the IT industry.

With such a low-cost solution on hand, the only explanation I can give for why we see such a disparate status quo is unwarranted risk-aversion on the part of hosts and users alike. Either users would be blunted by the notion of potentially of losing their [abandoned] home on the internet, or hosts are concerned about the repercussions of long-term absentee bloggers returning to find their address is no longer theirs and, in their disgust, sewing movement to other sites and negative reputation effects for the site that “wronged” them.

Naturally, I find these effects negligible against value of matching valued addresses to valued users.