Another amazing insight from fooled by randomness. The essence of this is that if you are a passive investor, the more often you track your portfolio, the more your headache. Suppose you have invested in a portfolio where the expected annual return is R%, and the volatility is V%. The insight is that the more often you track your portfolio, the likelihood of the portfolio delivering a positive return between two observations falls extremely quickly.
Category: fundaes
Predictably irrational traffic
There have been several experiments where they test the behaviour of people with respect to fairness. A popular test is that person A is given 100 rupees to divide among himself and person B. Now, B can either accept this division, in which case both get the amounts that A proposed, or he can reject it, in which case both get nothing.
Logically, even if B is paid 1 rupee, he should accept it since he’s better off with the rupee rather than without. However, numerous studies have shown that B rejects the division when he thinks it is “unfair”.
Saw a version of this yesterday. Typically at busy intersections, you would do well to follow the rules and wait for the green light. The cost of the wait is typically less than the cost of a potential jam (there’s a limiting case wrt traffic, of course; hence you see that there are more signal jumps when traffic is thin). However, it seems like fairness is important.
It was fantastic to watch, and even participate in it though I knew what I was doing was wrong. West had waited for five minutes at the intersection (a long time by that intersection standards) and when the policeman decided to ignore them to give Green for a second consecutive time to south, the barrier breached. As if acting on a cue, everyone on the west side breached the signal and caused a traffic jam.
It was fantastic. People were really patient for the first few minutes of the signal. No one even attempted to jump it. The policeman, they thought, was doing his job and they did theirs. After the jam been caused, once again, people were remarkably behaved and listened to the policeman as he helped clear the jam. It was only when the policeman acted “unfairly” that they didn’t behave. It was as if at that moment the crowd had lost trust in the signal system and decided to breach.
I ultimately ended up spending some 20 minutes at the intersection (i was part of the side that was “unfairly treated”).
Let me know if you have observed anything else that is similar.
Teaching History
About a week back I finished reading Guns, Germs and Steel by Jared Diamond. It is a mish-mash of history/anthropology/biology/linguistics and basically tries to explain why different civilizations have developed differently, and stuff like why the Europeans were able to capture most of America and Africa, etc. It doesn’t delve much into modern or medieval history, but basically uses the period in history when human beings started getting “civilized” to explain the possible causes.
Calendar
Contingent upon a favourable position of the moon, a unique alignment is going to happen this friday. Holi, Good Friday and Id Milad all occur on the same day. I’m not able to recall the last occasion when festivals of three very different religions, which follow vastly different calendars occur on the same day.
I’m curious to know that kind of calendar Christians follow. I know it has to be Gregorian, since Gregory was a Pope, but how is the occrrence of Good Friday/Easter determined? Clearly, the date varies every year due ot the day of week constraint. And I haven’t really observed any “nth friday/nth sunday” of the year kind of a situation also. There has to be some other funda, given that it’s clearly not lunar also.
On dumping tomatoes, burning wheat and leaving stands unsold
About a month back, I’d written that farmers in Karnataka, when faced with a glut in the tomato crop, elect to throw sack loads of tomatoes on the highways, rather than selling them. During the great depression in America, sack loads of wheat were burnt in order to prevent wheat prices from falling. During the India-Pakistan test match in Bangalore 2 months back, an entire stand (south east i think) was left completely unsold. All these have a common thread of logic – artificially restrict supply so that prices don’t crash, and you make more money.
Yes, I understand this is counterintuitive. How can you expect to make more by selling less rather than selling more? How can you expect to make more money by destroying what you’ve produced after investing thousands of rupees? Here is my take on the same. I’ll start with the necessary conditions for this kind of a situation, and then proceed to try and explain why this works.
1. Monopoly: A monopoly is essential for implementation of this kind of a situation. It is easy to understand why. Suppose there are multiple independent suppliers. Who is going to dump their stock? What is the incentive for you to dump your stock? You would rather that your neighbor dump his stock which is going to increase your profits. The only way out of this is in collusion. All producers get together and decide to dump stocks. Which effectively creates a cartel, and thus a monopoly.
2. Inelastic demand: For dumping to work, the additional revenue we make out of the un-dumped stocks should be more than the revenue we would’ve made from the dumped stock if we hadn’t dumped it. So basically the demand needs to be inelastic – around the region where we are going to dump. What i’m saying is that for a small change in quantity supplied, the price should increase by a large amount. As long as this keeps happening we can dump.
Going back to textbook monopoly economics, what we do to price is to maximize quantity * price. In other words, we supply the quantity where the total revenues are maximized. And it usually happens that this particular level is below the total amount we have produced. So we introduce into the market only as much produce that will maximize our revenues.
But what about the effort that has gone into production of this excess? Just look at the examples that I’ve mentioned. In all of them, you have already spent whatever amount that you had to spend. The costs have already been sunk. Apart from a couple of minor expenses (transportation, facilities, etc.) all expenses have been incurred before we made this decision. In other words Revenues are almost equal to profits. So we maximize revenues, not profits.
Now, taking the case of tomatoes, what do we do with the stock that we don’t want to sell? One option is to store it. That again, we’ll need to do based on how much the stored tomatoes will fetch us in the future, costs of storage et al. Given the facilities in India, it usually turns out that the costs of storage would be much higher than the expected revenues from it. So we only lose money by doing so. So what do we do? Dump them on the highways. Or if they take my suggestion, organize a Tomatina.
The other thing with tomatoes is that farmers don’t cooperate when they are making the decision regarding what to plant. If they did back then, some land that would’ve otherwise been used to sow tomatoes would be diverted to some other crop, which on the margin would yield more. Interestingly, the farmers seem to come together in a cartel only after the tomatoes have been produced!
So what are the policy implications from this? Firstly, infrastructure has to be improved. We need to be able to make storage of tomatoes cheaper, so as to encourage storage rather than throwing away. We need to encourage building of cold storages, and refrigerated transport systems. We need more investments in warehouses. Intuitively, it may appear as if these warehouses are just going to add to the cost of production, and thus push up inflation. If you see the larger picture, they are effectively encouraging efficient usage of land – which in my opinion is the most precious resource.
Second, the farmer needs to be able to easily estimate the revenues he will get by storing his goods. More importantly, he should be able to have a good idea about the revenues he will get from each crop even before he sows. And should be able to lock in the revenues before sowing.
We need to extend futures markets into all agricultural commodities. And keep the lot size reasonable so that it is accessible to small farmers. It is not as if the farmers won’t be able to use technology. Make it accessible to them, and they’ll easily take to it. The cell phone revolution is proof of that. Yes, small lot size could be a problem when it comes to settlement. Cash settled futures need to be explored.
Throwing tomatoes on the highway may be economically efficient when looked at in isolation. Looking at the larger picture, it only points to certain amounts of land and water and other inputs that have been wasted. That have been wasted growing tomatoes which no one needs, when they could’ve been used to grow something else. Agricultural commodity prices have been going up all over the world. Agricultural land and water are precious inputs, and need to be utilized judiciously if we have to continue feeding everyone. Futures markets help us allocating these resources efficiently.
Cross posted at the Indian Economy Blog
Bus to airport
The Bangalore International Airport is scheduled to open two months hence, and the question on everyone’s minds is to how one would get there. This airport is some 30 km outside the city, and though the road is part of the north-south corridor, it’s narrow in places and commute time to the airport will be massive. Plans to have a high speed rail link from the airport to the city center have remained just that – plans. Note the plural. There are way too many plans, and it’s unlikely we’ll have a rail line for another six years.
The BMTC has decided to do its bit by having “airport special” buses. It has commisioned some 40 Volvo buses which will run exclusively ot the new airport from various points in the city. Though the BMTC claims that the frequency of these buses will be once in 10 minutes or 15 minutes depending upon the route, a back of the hand calculation based on number of buses shows it won’t be more frequent than one every half hour. That too in the best case.
There are various other problems associated with this kind of a system. For one, passengers are likely to carry a huge amount of luggage. There needs to be an efficient system to make good use of buses so that a large number of passengers as well as their luggage can be carried. Due to the long journey, standing may not be an option. Another problem with the long commute is that the variance in travel time is going to be huge. To account for this, passengers will have to leave home insanely early.
Here is a better solution. The buses should be run by airlines. Let’s say Kingfisher takes control of some empty site level place somewhere in the middle of the city. Now, there will be dedicated buses for each flight that is going to take off (we might require some 3/4 buses for each flight). I know that this might result in inefficient usage of road space and buses, but wait.
The key to this is that passengers should be able to check in at the kingfisher bus stand! Using a Roving Agent kind of technology. That way, luggage can be managed easily (passengers dont’ need to take care of it till the airport). This means that the luggage can be easily taken care of in the boot of the bus or in a separate truck, and passengers need not be worried. Also, the fact that the luggage is taken care of by the airline means passengers dont’ need to unload the stuff at airports, go through the check in, etc.
So basically what I’m proposing is two or three “starting points” within the city (not more, it’ll be tough to predict, etc.) where people can board the bus after checking in their luggage. Apart from this, people without check in luggage should be able to board the bus at various points on the way to the airport! Each bus will have a “conductor” who has the roving agent with him and is capable of issuing boarding passes.
This is going to be the hard bit, but now that the airline itself is taking responsibiilty of the passengers’ transit to the airport, they can probably delay the flights till all the buses carrying their passnegers have arrived. Of course, they’ll time the departure from the bus stands with sufficient and more time for all buses to reach the airport, but in case of unforseen traffic jams, etc. planes could be held back.
And yeah – once a bus reaches the airport, remember that the checkin is already over and people can just go through security check (maybe there could be a few special security counters for those coming by these buses) and board the aircraft. The process gets much more streamlined!
Of course this is going to cost the airline quite a bit. This can be initially made complimentary, and then maybe added to the price of the ticket. Uniformly. Wthether a passenger opts for it or not. This will encourage more passengers to use this facility rather than driving up to the airport by themselves.
I think this is a far superior situation to having people traveling by public transport to reach teh airport.
Trade-offs and city structures
Of late, i’ve been reading this blog on urban issues written by this chap called Sandy Ikeda. One thing i noticed in the blog (don’t remember exact post) is that earlier, the poor used to live in the city centers and the rich used to live in the suburbs. And that this is changing in most places, with the rich trying to move closer to the city and the poor are being forced to move out to the suburbs.
I think it’s due to the steady change in costs. The “rich” are rich when it comes to money and poor when it comes to time. On the other hand, the “poor” may be poor in money, but they are usually rich when it comes to time. Earlier, public transport was scarce, while population density was much less. This meant that cost of commuting from the suburbs was quite high in terms of money, though not much in terms of time (the lack of congestion meant that travel was quick).
What has led to the change has basically been congestion. Due to migration, cities have become larger, and the roads more congested. And hence time – that oh-so-scarce resource for the “rich” – to travel to the city center from the suburbs has shot up, and thus the rich have tried to make do by moving closer to the center.
The poor have been priced out and forced to move away, and this has also led to development of things such as public transport, etc. And I think this trend is only going to continue. There is no chance of a reversal here. Unless of course – a high class premium “first class” public transport system is developed. (note that the last sentence was in passive; mainly because i’m not sure whose responsibility it is. though i think it’s the city governments).
What a premium public transport system does is to increase costs of transports in terms of money and reduces it in terms of time. And makes it easier for the rich to move away to the suburbs.
Of course, this whole model is based on the “traditional” city model where the office space is in the center and the living areas around. It fails miserably in the case of places like Bangalore where most of the office space coming up is located outside the city.
Are
I’m halfway through this excellent article. I highly recommend you to read if you’re even remotely interested in social networks. Duncan Watts (prof of sociology at Columbia; now on sabbatical at Yahoo!),? whose book six degrees was instrumental in introducing me to the concept of social networks, tries to refute some of the ideas talked about by Malcolm Gladwell in his book The Tipping Point. Watts, studying email networks, seems to have concluded that there is no such thing as “influential people”. That everyone else is as influential as everyone else when it comes to propagating stuff.
The point of that verbose first paragraph was that this particular article has reminded me of the word “are”. Not “are” as in plural of “is”, but more in the IIMB context, where it refers to something really good. For example, “livejournal are” translates to “livejournal is really good” in normal english. Or sometime, when someone tells you something which you find to be really great, you can just reply saying “are”. An extremely simple but extremely powerful usage of an extremely simple and common word.
It all started on the first day of Saarang 2006. I don’t remember too well but at least all of the following people were present at the dinner at Eden (at besantnagar) that evening – samanth, vinod, tchami, kodhi, woreshtmax visnu and i. The key characters were samanth and vinod. Theyy had got pained by the IITian usage of “are there” for “is there” and were responding to every sentence with “are” (they were too lazy to say “are there” i guess), and occasionally with “are not”.? Five days later, as we were on the bus to Bangalore, “are” had come to mean what it does now. It’s antonym was “are not”.
Now, Kodhi and I, who claim credit for this particular usage of this word, were extremely active on BRacket, the IIMB internal message board. So what happened was that the probability that someone read a word written by us was much higher than someone reading a word written by the average IIMB bloke. In that sense, yes, we were influential. The next “convert”, I think was
![[info]](http://stat.livejournal.com/img/userinfo.gif){kind=small}
, who is always on the lookout for new lingo, so much so that he uses more IITM lingo than the average guy from IITM.
I don’t recall it being a very conscious attempt. However, we were unwittingly using the word more often in our conversation. We used to respond to posts on BRacket with a simple “are”, indicating our appreciation to the preceding argument. Consequently, we received queries to explain our wonderful responses, and this gave the word fresh footage. Some of the people who thus asked us immediately liked the word and coopted that into their language. I remember
?and
?being especially early adopters.
It was a matter of time before it had unofficially entered the official dictionary.
So coming back to the whole process – it was originated and propagated by a bunch of people who were more active in the community than others. However, most of us wouldn’t by any standards qualify as any kind of thought or opinion leaders. For all you care, we were just a bunch of random guys, only slightly more visible than the rest. One thing is there – all of us were much more receptive to fresh ideas (and fresh words) than the average populace. And our assimilation was quick – the word didn’t take much time in entering our normal conversations.
Now what does this mean in terms of Gladwell’s and Watts’s frameworks? On one hand, yes, we were a set of “influentials”. However, that we were influential in this matter (creating a new word) had nothing to do with us being influential in any other sense. My take on the subject based on only this particular case study – yes, influentials matter. You do need people with a higher-than-average visibility in order to propagate the concept. On the other hand, targeting influentials is harder than most people (admen, etc.) think. The same people who were influential with hush puppies shoes weren’t influential with say popularizing *insert random music genre here*. It is important for marketers to somehow identify who are the kind of people who will be more influential in the current context, and target them.
Are
Coffee pricing at Chalukya
The pricing of coffee at the Samrat restaurant (part of Hotel Chalukya on Race Course Road) is interesting. This is a popular old restaurant, and being in an area full of government offices, is perennially crowded (despite its large size). It is a sit-down kind of restaurant, though you might have to share a table with strangers if you’ve gone in a group of less than four.
Improving umpiring
Just Outsource the front foot no ball
Takes a lot of pressure away from the umpire. Right now he has to keep looking down till the bowler bowls and then suddenly jerk his neck up and decide on lbws and caught behinds and bat pads. This jerk of the neck causes considerable strain, and can also present the umpire with a not-so-good angle and the probability of error goes up.
One reason us people in front of the TV get more LBWs right than the umpire is that we are not looking down at the popping crease waiting for no balls. Another thing is that there have been a number of cases where the front foot no ball hasn’t been given (notably in the barbados test in 1997, old trafford in 2001 when saqlain consistently overstepped, and last week when jaffer got out to lee) and outsourcing the no ball decision, either to the third umpire or to a computer would help immensely.
We can have special cameras installed in the ground which only provide a view of the popping crease at both ends. And while the ball is being delivered it’s the job of the third umpire to decide on front foot no balls. He anyways has? no work to do unless he is referred by the on field umpires.
That said and done, we need to develop technology for this soon enough. I suspect some variation of the cyclops used in tennis might come in useful here.
Anyways, I’m reminded of this quiz question in the Shaastra Main Quiz in 2002, where the question was about the inventor of cyclops sitting in his home and watching saqlain destroy England bowling no balls and wondering whether his invention would be useful.