What GDP can Tell Us (and What it Can't)

Debt/GDP Ratio, co. The Market Ticker

GDP is an oft quoted figure for conveying the economic output of economies.  The figure is useful in conveying a broad sense of how much value is being produced by an economy.  I say a broad sense, since the figure is more like gross income on your paycheck rather than net income.  And while gross income might be useful for some purposes, the money you actually take home is net.

More specifically GDP is the upper bound on the productive output of the economy.  Of course, it's very hard to actually determine exactly what the productive output of a large economy is precisely.  Furthermore, different ways of measuring GDP will actually give different values.  Again, because of it's nature as an upper bound this really isn't worrisome.  It is probably more worrisome that GDP heavily overstates it's case.  Nonetheless, because it is a best possible estimate, comparing it to the national debt (a figure that is bound to be understated) and finding national debt exceeds it, should bring some pause.

There are additional cases where using GDP as an indicator won't distort what economic facts you are trying to discern.  In the case of excess credit in the system, as has been noted repeatedly by Denninger, it is clear that since GDP is an upper bound, if what you are measuring paints a dire picture, you can be assured that the reality is far worse.  For indicators like this, sadly it also means that optimistic news is always going to be overstated.

The problem is in how GDP is computed.  GDP is a simply the product of the sum of the credit and cash in the system with the velocity--the number of times the money changes hands.  The reasons that money is being exchanged don't matter.  If you buy $1000 of capital equipment to expand your business, it affects GDP the same as if you spent $1000 on an insurance deductible, because flood destroyed your existing equipment.  It has the same mathematical effect on GDP.

It is no surprise that  this measurement of value suffers from a failure on the common-sensical level.  Clearly buying $1000 additional equipment is going actually increase the overall societal value, whereas replacing destroyed equipment has a neutral effect.  We don't have people run around and destroy things for the sake of increasing societal value.  Though, some people will do that in poor neighborhoods and split the cash received from an insurance company with the insured.

Luckily other methods have been developed for determining "real" overall societal value.  Sadly GDP cannot be viewed this way, because of the aforementioned distortions.  Using GDP per capital doesn't really convey the true contribution of each citizen.  The mean is going to be positively biased due to the nature of there being a small fraction of companies contributing to a large majority of GDP, along with it already being inherently overestimated.

GPI/GDP Divergence Over the Last 30 Years

GDP v GPI per Capita in yr 2000 USD

An indicator that better states its case is the Genuine Progress Indicator or GPI¹.  Looking at historical calculations, GDP has diverged ever more from then GPI.  It doesn't take a remarkable inductive leap to conclude that externalities, that are not accounted for by GDP, are to blame.  These observations, along with an increase in GDP per capita, are highly suggestive of a simultaneous privatizing of profit while imposing externalities on society.  Of course GPI includes societal costs, like crime, that aren't the result of conducting business.
GPI has many benefits as an indicator.  It captures a large amount of important costs borne by society like natural disasters and healthcare that GDP doesn't.  Additionally, it adds in human progress indicators like income disparity to better represent actual life quality.  Looking at the GPI per capita, it is observed that US residents have made no human progress or even negative progress since the 1970's.

The indicator is not without it's critics.  Take Landefeld's² take of GPI as too "subjective" tainting GDP.  That is of course true, as any attempt to accurately asses the value of an economy will at it's core be subjective.  GDP suffers from the same flaw, resulting in multiple ways computing it; each method giving different values.  Despite the dense and deep mathematical material, economics is a multidisciplinary field, where the rigour must eventually come to grips with human social behaviour.

It's important to have econometrics to be able to make societal and public policy decisions in the first place.  All economic indicators are going to be subjective, so it makes sense to try and come up with a more complete picture of the economic pie.  Since it includes less tangible benefits to society though volunteering as well as accounting for losses like healthcare and leisure time, the indicator draws a more complete picture of reality.  Perhaps GPI is more objective, taking into account additional factors that people consider relevant to their well being.

Further Reading & References

¹Denninger, Karl.,  The Market Ticker. (Sept 12th 2010.)  http://market-ticker.org.  Web.  12 Sept 2012.
²Talberth, John Dr. Cobb, Clifford. Slattery, Noah., The Genuine Progress Indicator 2006. A Tool for Sustainable Development. (2006) Redefining Progress Oakland, CA.  http://rprogress.org/publications/2007/GPI%202006.pdf.  Web.  12 Sept 2012
³Landefeld, Steve.,  Economist Debates: GDP Proposition: The House Believes That GDP Growth Is A Poor Measure Of Improving Living Standards.  The Opposition's Opening Remarks.  (Apr 20th 2010.)  http://www.bea.gov/about/pdf/Newsclips_GDP_group.pdf.  Web.  12 Sept 2012.