Statistical Magic and its uses

It seems these days that we all need to be professional statisticians, the real statisticians having long since drop any pretensions of integrity. I refer the reader to this mail post by Kris Sayce, an investment advisor, with several investment newsletters (readers note: I subscribe to his Australian small-Cap Investigator). Watch Mr. Sayce dispel the magic:

CBA’s Chart of Income to House price

The same chart using the proper numbers, the previous chart used numbers from two different sources one for australia one for the rest of the world.

Now for ANZ banks chart. You can see the  rather prominent upsurge of population relative to housing from 2004 onwards.

Here is Sayce’s own graph with the scale adjusted.


Why is the change in scale important? What Mr. Street did is move the scale of the annual population growth to be 2.5 times the number of dwelling starts. For example, 200,000 on the right hand scale is 2.5 times greater than 80,000 on the left hand scale.

Chronic oversupply

This number represents the approximate number of people per dwelling.

In other words, if the average household size is 2.5 people, then you arguably need one dwelling per 2.5 people.

So, if the population increases by 200,000 you could expect to need 80,000 dwellings.

As this chart shows, for most of the 1980s, 1990s and early 2000s the ratio was closer to one new dwelling per 1.5 people. Or to put it another way – an oversupply of housing.

In fact, a chronic oversupply of housing.

Most likely as a result of negative gearing rules that encouraged investors to build housing at a loss. So what this chart shows is far from there being a chronic undersupply of housing, there has been a chronic oversupply of housing for all but the last two years!

 These are Australia’s largest banks massaging statistics. Statistics used by people; entering into home loans, leveraging on investment property,by journalists and no doubt government bureaucrats. Policy will be decided on figures such as these. This is neither the first time or second, but a continual trend. As I have pointed out in a previous post, what the public took for granted was the implicit notion of integrity in statistical models. They cannot assume this today, even with the simple charts as above. To argue that these were unintended mistakes, errors in the system, is very naive. One of my areas of training was statistics and econometrics specifically. I know that you cannot make mistakes like the above, they were deliberate attempts to mislead. Many institutions take a data sample and then perform the statistical methods that render the correlation/result that they wish, this is considered heresy by professional  honest statisticians. Take a moment for that to sink in. Remember such memes as, ‘latest study says that eating too much fats raise your chance of cancer 30%‘. Or ‘oceans will rise by 1 meter by 2050 if current trends in CO2 emissions remain’.

How does it go. ‘Theres lies, damn lies and statistics’. The beauty of massaging statistics is that an important decision maker no longer is required to make decisions, ‘the graphs show that, ….there’s a strong correlation between..’. Hence, politicians can fall back on experts. It’s far worse than you think, the technique was developed in the civil service bureaucracy of western governments, slowly the methods and mentality has corrupted large corporations, banks, and regulators. Yes minister, which Pavel Jacko and I will do further posts on this brilliant series, illustrates this so clearly

The government bureaucracy isn’t a sealed and self-contained part of  society, as one tends to think. As government subsidies and welfare grow, the bureaucracy of government begins to invade the private sphere. Now libertarians will say, of course Mr. Worden, obviously! What they don’t understand is that the private individuals and institutions who must deal with government bureaucrats, and who hasn’t, learn the corrupting power techniques of the bureaucracy. The bureaucracy itself extends from government departments into large corporations through contracts. The profit motive tends to keep this cancer at bay in corporations, however, the larger they are, and the more contact there is with government and its regulators the more insidious the bureaucracy. Work is not done in the government bureaucracy, to be sure, the aim however is to secure larger budgets and employ more personal. This requires the creation of ‘needs’, or demand if you want to be economical. Of course, questions will be asked. So the fudged statistics provide the convenient cover for further ‘new’ policy. This is what consultation is all about. It is merely the private sector using bureaucratic methods. Read a prepared report by a consultant, a majority of which is mere filler, with critical numbers and figures manipulated.

With this in mind, one can now understand the seemingly unstoppable arrogance of the anthropogenic global warming brigade, oh sorry I meant climate change brigade. How could we forget that little Orwellian hedge. A majority of scientists especially ones concerned with climate change are employed by the government, if not directly, through grants, subsidies, tenure. For them to have a raison d’être, there must be a constant search for and study of climate change. The research always conveniently turns up results for climate change, which of course, demands further detailed research, and so on etc.

From a commercial perspective, which is strangely more illuminating, the scientists are creating a market for themselves; or they’re creating a brand and telling people, politicians and the voting public, to buy it.

Similarly we return to our banks CBA and ANZ, who are incredibly leveraged. Mr. Sayce again provides us some interesting charts.

ANZ as of now:

The light blue area is ANZ Bank’s exposure to the residential mortgage market.  59% of ANZ’s loan book is exposed to residential mortgages.  [cough]

The dark blue area represents “institutional” lending, the medium blue “commercial”, and the olive colour “other retail & wealth”.

In other words, just 17% of ANZ’s lending is directed towards commercial enterprise.  The majority of its lending goes towards propping up the housing bubble.

Yet if you listen to most mainstream commentators, they’ll tell you how important a strong banking sector is… and how if it wasn’t for Australia’s strong banks, the economy would be in a dire condition…

…The fact is, the banks are false prophets.  They do nothing more than gain central bank and government support for their activities, and then once they’re established they ensure continued support by warning of the consequences of withdrawing the support.

It’s similar to US Treasury Secretary, Timothy Geithner’s series of letters to the US Congress.  He warns about the dire consequences if the US government is denied the ability to go further into debt!

In reality, banks aren’t the economic saviours they’re made out to be.

The banks and the people who run them are just bean-counters and pen-pushers… they’re long on bureaucracy, but short on initiative…

And why do banks favour housing?  Because it’s backstopped by the government of course.  Unlike businesses which can be much harder to stop from collapsing.

Trouble is because the banks know this, it concentrates their loan book.  Rather than diversifying their exposure to naturally lessen risk, the banks take advantage of their special gift from the government and central banks – that is to take as much risk as they can reasonably get away with.  Because the taxpayer will ultimately provide a bail out.

The concentration of risk has resulted in the current house price bubble.  If they hadn’t shifted so much of their lending towards housing the bubble wouldn’t have been created and the collapse of the property market needn’t happen…

…Now take a look at this interesting snapshot.  It’s from the 1978 ANZ Bank annual report:

You’ll notice that just one-quarter of the trading bank lending in Australia was to persons – we’ll make a guess this includes mortgages.  In New Zealand it was even less – about one-sixth.

But look at the rest of the numbers: manufacturing alone accounted for 17% of the bank’s lending.  Today the bank’s entire commercial loan book accounts for 17% of all loans made by the bank!

In 1978, total lending to the business sector made up over half of all the bank’s lending.  Yet today it’s a pathetic 17%.

It’s a perfect illustration of how the banking sector has all its eggs in one basket – the house price basket.  No wonder they’re so keen to downplay the idea of a housing bubble.

 So this is why the banks conjure up fraudulent statistics, they have created a market of home loans for themselves. It is important for them to maintain the mirage of demand. The statistics created the need and then are used to further support it. Just like our government bureaucrats and scientists. If one now approaches society from this perspective we can see all sorts of mirages. The French philosopher Bruadrillard refers to this as Simulacra and Simulation. There are now two realities, or a reality and its simulacra; a property bubble caused by inflation, and a property market with low supply supporting a strong valuation. Upon close expectation one of these realities disintegrates, and we are left with the actual reality; there is a property bubble cause by inflation with no underlying fundamentals.

Of course, the public find this all too confusing, and simply leave it to the experts. For a disturbing mental thrill consider the entire society and modern culture as a giant bureaucracy in which you are a statistically insignificant cipher.

About Avadoro Worden

This entry was posted in Australia, Climate Change, Computer Models, Finance and Economics, Mathematics, Socialism, Uncategorized and tagged , , , , . Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>