Archive for August, 2009


August 26, 2009

1 kg Parrys white sugar: Printed price Rs 32.95  The shopkeeper took Rs 33.


Data gaps

August 24, 2009

The great Charlie Munger (and here I wish I could give you the link to the piece) once compared the stock market to a system of horse racing called pari mutuel.  I don’t know much about racing, but from what I gather, the bookies can fix any race.  But we know that in every race, exactly one horse must win.  If the bookie is to make a profit (apart from his vigorish, which is very like brokers commissions) the amount he pays out to winners must be less than the amount he gains from losing bets.  Here is where his skill in quoting odds comes in.  The odds are the only data we have.  If we knew the volumes of bets at the various odds, we could work out (easily I believe, though I’m too lazy to try) which winning outcome would maximize the profits for the bookie sector as a whole.  An excel spreadsheet would do the trick.  But that would take the fun out of gambling, or would it.

Similarly, at my brief stint at a large investment bank, I thought it’s cool that we have all this fx price data.  Tick by tick and all that.  But do you have tick by tick volume data?  Can we have a look at both?

And then we could draw a curve of volumes transacted against price, not to be confused with an ordinary demand curve, and see if lot size is inversely related to transaction price, or some such other hypothesis.

Always remember there is a market maker in stocks.


August 24, 2009

It’s one thing to have access to a great deal of data.  It’s quite another thing to validate that data.  How do we know that the data is correct?  (Must read 1984 again.)  I think the best way would be, as a first approximation, to bring the micro and macro number into coinsistency.  Start bottom up.  Start with, say, the population.  How do we know that number is correct?  And to top it all, there is genuine error and deceit.  Even here I know this because I saw it.  My own experience I can vouch for.  Maybe evil and error partake of a common nature.  My experience tells me that evil is often an error in the long run, and about error being evil-I don’t think we can go that far.

Cui Bono?

August 23, 2009

The emphasis on the accounts of the household sector is precisely to answer the question of who benefits the most from the status quo.  Without which knowledge we will never understand politics.


August 23, 2009

What I would really like to see is a table with columns Households, General Government and would there be a rest of the world? Not necessarily, because we would be doing accounts for the planet.

The private nonfinancial, and private financial sectors would be pushed back to the Household sector.  And the Household sector is a consolidation of the accounts of households.

A problem in public finance

August 20, 2009

The problem with pushing all ownership back into the household sector, is that there remains a large chunk of socially useful capital unclaimed.  This is the government.  Let’s call it the public sector.

The public sector is largest single economic entity-this without looking at the data.  In terms of net worth.  Just a guess.

Who does the public sector belong to?  Does it equally belong to each caput?  Prima facie, no.

More on demand

August 20, 2009

Can the possesion of property constitute an attribute which neatly puts a red or white hat on every head?

For instance, in Himachal, we have our neighbours in Bagora.  There Soma owns land in the village and owns her own house.  Yet she has to work in our house just to make some money.  She would be what any right minded person would call poor.  Yet she owns property.  At the other end of the scale, we have Calpers, which owns a large part of the American stock market.  The savings of working class folk have been amassed into a fund which ends up owning a sizeable chunk of corporate America.  No: property as a criterion will just not do.  There are a number of urban landlords, who when measured on an income scale, would just be middle class.

But I would still like to have assets as a column in any database of humans.  And a breakdown of those assets.

What I am really interested in is movements of net worth.  The rate of increase of net worth of nations.  Or as an altogether wiser Adam put it before me: The Wealth of Nations.


August 15, 2009

Aggregate demand depends on income distribution.  Profits, interest, rent and wages & salaries.  Profits can be retained for further investment, or distributed as dividends.  The wage earning sector (and here I include myself) consumes a fixed basket of goods.  (How far apart is my budget from the budget of my auto driver?).  But, doubtless, issues of quality supervene.

Assume that what is distributed as dividends is consumed away or on lent.  If lent, it can safely be considered as investment.  (Where are the figures for these magnitudes?)  (In the flow of funds accounts.)

So there is a stream of spending debouching on the market for goods and services, which can be usefully dichotomised into  basic and non basic goods and services.  There is a sector of the labour force engaged in the production of the basket of basic goods.  (Or do we, as in Leontief, consider all goods as basic?).  No, we cleave to the distinction between basic and non basic.  Then there is the sector of the population engaged in the production of non basic goods and services.  So the basic goods sector turns out the bundle of consumption needed to sustain the entire labour force.  And the non basic workers receive incomes from the spending of dividend income.  Thus can one separate the population into two classes-one of workers, and one of entrpreneurs.  The entrpreneurs can be distinguished by the fact that they have property.  Property to defend.  For which they employ henchmen.  And more non basic services.

The financial lexicon

August 12, 2009

Yield to Maturity


August 7, 2009

Havn’t been updating the blog because of updates at work.  Fascinating work. 

Have to learn about databases.

This database consists of time series.  Or more precisely, realizations of time series.  Time is the primary key.  This is true if as Aditya told me last night over dinner, a primary key is the heading of the column which orders the rows in a table.  Time partitions into frequencies.  Annual, quarterly, monthly, weekly (5 and 7), up until the most granular level.  Thus a single time point has a row of geographies, concepts under that geography, sources and such like.

I am beginning to get an idea of the database.