Monday, February 24, 2014
"Lights Go Out on Dismal Science"? A Long Time Ago!
Illustration of the Gaussian Copula Formula used by economists to justify credit default swaps.
"How many economists does it take to change a light bulb? One! When the one they used in graduate school goes out, they sit in the dark."
The above joke, as cited by Neolib hack Robert Samuelson, is supposed to hint at why economists are so pathetic at making accurate predictions. Unlike astronomers, who can accurately predict the position of Jupiter or Mars in 2050 or the next lunar eclipse or occultation of a star, the economists can't even predict simple stuff in their immediate domain - say like forecasting the growth would be 3.2 % in 2011 when it was only 1.7%
Of course, the reasons for that failure are now well known. Unlike astronomers and astrophysicists who - when they make predictions - must take into account all relevant information, the economists tend to leave out what they dismiss as irrelevant. In the case of the growth forecast, it was simply that the economic stimulus passed in 2009 wasn't large enough to promote the growth expected. As Nobel winning economist Paul Krugman noted it needed to be at least a trillion dollars more, given the recession was much deeper than the economists originally thought.
On April 26, 2010, a letter of mine published in The Financial Times, drew attention to the abysmal failure of modern macro-economics (touted as a “science”) to predict the 2008 market meltdown and financial crisis. I pointedly noted that economics had too many “externalities” which included aspects like environmental costs and resources that were omitted from its models, as well as lacking any consistent empirical basis analogous to physics.
In the particular case of the failure in predicting the credit crisis and meltdown, academic economics erred by assuming a putatively poorly regulated system was capable of sustaining massive risk entrenched in obscure, poorly understood credit derivatives created by Wall Street “quants” , most of whom had forsaken bright careers in science or mathematics to invent these devious financial instruments for investment banks. The assumption saw all the hinges come loose when the credit default swaps were immersed in securities purported to be safe, since they were given AAA ratings by credit agencies like Moody’s and Standard and Poor’s.
Thus, the forecast failure was predicated on a three-way collapse of the paradigm: 1) commercial banks taking on the risk of investment banks by leveraging their assets to preposterous ratios (sometimes as high as 33:1), 2) credit derivatives designed using the Gaussian Copula Formula which enabled them to be sliced, and spread throughout ordinary securities such as collateralized mortgage obligations, and 3) a failure of the credit rating agencies to take proper note of (2) and in effect, be blinded while assigning bond ratings the securities didn’t deserve.
Since then, it’s become ever more evident why economics is a failure and can’t even be regarded as a science like Physics, or even in a “pre-scientific phase” as one respondent to my FT article (Sanjay Bissessur) claimed. But economists' failure to even appreciate or use the Gaussian Copula formula correctly shows that they aren't even in the pre-scientific phase.
In the case of the Gaussian copula, invented by David X. Li while working at JP Morgan Chase and articulated in his (2000) paper: ‘On Default Correlation: A Copula Function Approach” it isn't a true mathematically sound equation analogous to those used in physics or celestial mechanics.. It is more an intellectual Frankenstein monster that never should have seen the light of day any more than a four-headed baby with a pointed tail. For example, Li's misuse of the distribution functions (FA(1)) and (FB(1)) would appall any genuine mathematician or physicist. Each is actually based upon significant uncertainties via survival law distributions which can vary enormously. There is no way to normalize any probability based on (TA, TB so there is no way to equate Pr[TA, TB] to anything on the left side. The equal sign is dangerous recklessness masquerading as math. Were economists or "quants" know any of this when they cranked out credit default swaps? The evidence of failure to predict the 2008 credit crash shows they didn't.
Back to Robert Samuelson who, in his recent piece on the Dismal Science (2/ 21, Denver Post), attempts to blame Keynesianism (the theory of John Maynard Keynes that when aggregate demand is low government must offset the decrease via stimulus and deficit spending.) He asserts "the U.S. DID respond aggressively to the financial crisis" - but misses the point (as was missed by most Pareto- compromised economists) that in reality the stimulus and deficit spending weren't large enough! This was only evident in retrospect when it was learned the negative growth actually hit -1.6 % in the first quarter in which the stimulus was initiated, in 2009. Hence, Krugman's insight that the stimulus ought to have been nearly $1.7 trillion instead of $787 billion.
Samuelson also makes the error of most Pareto -distribution based economists, in referring to the "Fed pouring $3.2 trillion into the economy since 2008 to keep interest rates low and accelerate economic growth". He adds that "so much money pumped out so quickly should have spawned higher inflation by monetarist reasoning, but it hasn't."
But one wonders what planet he's living on. In fact, inflation has been going up and rapidly - more than 16% a year in the health care costs domain, and 5% a year in food costs. Don't believe me? Check meat , egg, dairy costs at different intervals. The problem isn't that there is no inflation, but that the Federal Reserve selectively ignores those costs -expenses it doesn't want on its cost of living assessments. The problem is that the Fed leaves these costs, as well as for fuel, out of its CPI or Consumer Price index.
According to former Fed honcho Ben Bernanke (WSJ, Feb. 17, 2012,, p. B1, 'Consumers Price in Real Cost of Living'), all such costs as food, housing - even exploding rent, as well as meds - are designated as "shelter costs". Hence, they are, get this: "essentially made up numbers" . Bottom line, according to Bernanke's statistics, higher costs of gasoline, groceries, rent and so forth - while not directly in our heads- are of no concern to him as regards inflation. After all, the CPI or consumer price index, is gamed to lowball the impact of these shelter costs so that even if inflation is really going up by about 2% a month, to Bernanke and his Pareto-stats it is really ZERO since his statistics (like the 'owner's equivalent rent") don't show up anything worth fretting over.
And economists wonder why they can't even get basic predictions correct?
The other aspect, is that although Bernanke pumped in $3.2 trillion - via his "quantitative easing" - it largely benefited only Wall Street and the stock pushers, traders by inundating them with cheap money. Bernanke likened the Fed’s $85b a month bond buying to giving the economy small, needed doses of medicine - but it was more like crack to the financial markets that quickly became hooked on it. This is the main reason a giant stock bubble has been produced which has no relation to actual corporate performance, i.e. corporations are still sitting on over $1.7 trillion in cash - one reason job creation is so sluggish.
Bernanke's pumping up of the economy via QE I and II certainly hasn't helped seniors on fixed incomes who depend on the interest earned from fixed income investments for extra money to spend. (They lack the time horizon to be dabbling in stocks, or even mutual funds, as they'd never have the time to make up the losses like a young Turk.) It hasn't helped most college students whose loan costs have grown higher, and it hasn't helped minimum wage workers who've now seen their food stamps benefits cut!
In fact, a much superior alternative, if the economists had any sense and weren't tied to their delusions (and the Tea Party idiots in the House wouldn't have opposed it), would have been for much greater food stamp benefits - no cuts -and similarly extending unemployment benefits especially for the long term unemployed. These simple measures would have instantly - relatively so - spurred aggregate demand and a healthier economy. The reason? People receiving that money would have gone out to buy food, pay car or home loans, or buy fuel and meds. As opposed to corporations using the money to buy back their own stocks creating a false jump in stock prices, or traders jacking up the P/E ratios to make stocks look better and enhance their own commissions.
The sad fact is the entire U.S. economy is 'upside down' - in so many ways it makes a genuine scientist's head spin. In any rational real world economy, for example, the country would be creating jobs to repair its crumbling infrastructure - thereby killing two 'birds' with one stone (or rather stimulus cash infusion). It would also long since have pulled out of Afghanistan, realizing that quagmire, like Iraq, is a losing wicket and merely a money drain. All the money saved from the last two years in Afghanistan alone could have funded a major infrastructure repair program. But we're a nation too deluded by external "threats" and security while ignoring our true domestic security.
Incredibly, the dismal science economists even have much of the nation dumbing down - going gah-gah over their stocks or mutual funds' performance, but forgetting that doesn't mean diddly if they can't get water because their water mains are rusted through, or their toilets back up because the sewer lines are corroded.
The very condition of our nation, with so many serious, unmet domestic needs, shows that the economics adhered to is more in the way of magical thinking than any "science"- even a dismal one.