The 2003 Nobel Prize in economics

Two economists who have developed statistical techniques to track economic trends and to measure investment risk - Clive Granger and Robert Engle - were this year awarded the Nobel Prize in economics. In this article I will be focusing on the contribution of Clive Granger to economics. According to the Nobel committee Granger's important contribution to economic science is his discovery of a phenomenon called cointegration. This discovery, so it is held, enables economists to accurately validate relationships among various economic data. So what is it all about?

Often we observe that two pieces of data, which are not supposed to have any relationship, appear to have very high visual correlation. For instance, we may discover a strong correlation between the intensity of dog barking and movements in stock prices. One is then tempted to take advantage of this discovery in order to make money in the stock market. In reality however, both the barking of the dog and movements of stock indexes have nothing to do with each other.

What makes the apparent good correlation is that they are both influenced by an upward long-term trend. Also, fluctuations of these data don't seem to converge around the trend but just seem to move in an upward direction. These types of data statisticians label non-stationary. (Granger, C.W.J. and Newbold, P. (1974) Spurious Regressions in Econometrics, Journal of Econometrics, Vol. 2, pp 111-20.)

In contrast, data that converges around a fixed value is labelled stationary. Data that is stationary implies an unchanged structure, something that is stable and hence one can make sense of it, whereas non-stationary data is associated with irregular fluctuations, which of course makes it very difficult to make any sense of. Thus if something drifts aimlessly it is not possible to say much about its future course.

Now, if one tries to make sense from data that is irregular obviously one will not get very far. This, however, creates a major problem for economists since it is held that most of the data that economists and financial analysts are employing are not stationary. Consequently, incorporating these types of data into economic analyses leads to misleading results.

For instance, an economist wants to establish the importance of changes in production on people's consumption. The common procedure for this is to apply statistical methods on consumption and production data in order to establish their interrelationship. In short, by means of a statistical technique, also known as regression analysis, one establishes how consumption and production are quantitatively connected to each other. Let us assume that an economist has found that the relationship between consumption and production is summarised by the following mathematical expression:

Armed with this finding the economist can now tell us the direction of consumption if there is a change in production. Thus if production is 100 then consumption will be 60 (because 10 + 0.5 x 100 = 60). Economists label the numbers 10 and 0.5 parameters. Observe that the knowledge regarding the size of these parameters i.e. whether they are 10 and 0.5 or something else is obtained by means of the regression technique.

Note that 10 and 0.5, which have been generated by regression method, are estimates of true parameters in the real world, or so it is held. It is also held that on average these estimates are very close to the true parameters. In short, it is held that any conclusions derived from the equation regarding the relationship between consumption and production are a reflection of reality. Granger, however, contests this.

He argues that no meaningful conclusions can be drawn from the above equation if the data employed in establishing this equation are non-stationary. In plain English, it is counter-productive to establish meaningful conclusions from data that drifts aimlessly. The parameters that one will get from such data will be erroneous and hence the outcome of the analysis will be meaningless. So how does one overcome the problem?

Now, if one could establish a common factor that influences both consumption and production then these two time-series are said to be connected, or cointegrated. Granger and others have shown by means of mathematical and statistical methods that the existence of the common factor makes the interrelationship between non-stationary time series stationary.

Thus consumption and production can be observed separately as a non-stationary time series. Therefore if one tries to establish economic relationships between them one will get misleading answers. However, if one were to suggest that both consumption and production have a common factor then one could infer that over time both consumption and production must move together. This common or co-integrating factor could be that people's well being requires consumption and production.

Moreover, that without production there cannot be consumption and without consumption no production is possible. Another example is an identical good, which is trading in different locations. The day to day fluctuations in prices may appear to be random in various locations and therefore most likely will not correspond to each other. However, the existence of arbitrage and the law of supply and demand will make sure that over time prices in various locations will move close to each other.

Now, instead of trying to find out what the co-integrating factor is Granger and others have produced a mechanised framework, which enables economists to establish whether data complies with co-integration i.e. whether the relationship between the data makes sense so to speak.

Once it is established that the data is cointegrated it can then be incorporated by means of a certain mathematical procedure to establish the correct parameters. (Granger, C.W.J. and Weiss, A. A. 1983, Time series analysis of error-correction models in S.Karlin, T. Amemiya and L.A. Goodman, Studies in Econometrics, Time series and Multivariate Statistics, in Honor of T.W. Anderson, Academic Press, San Diego, pp 255-278.)

Various statistical results that are produced by means of Granger's framework are therefore regarded as valid since they have been applied on co-integrated data. Granger's discovery raises serious doubts about conclusions regarding economic interrelationships which are reached by means of the old techniques. It also provides a criticism of the popular usage of correlations without attempting to make sense of the relationships. Granger's framework can be seen as a preventative in minimising the use of meaningless correlations.

For instance, the Granger framework will indicate that movements in the stock market and the intensity of the dog barking cannot be co-integrated and therefore the use of these relationships to make money in the stock market could prove to be a very expensive exercise. In this respect it could be regarded as bringing back the validity of fundamental analysis. This must be contrasted with the popular way of thinking that fundamental analysis is of little help since the data is of a random i.e. non-stationary, nature. So it seems that the Granger's framework is a great tool in furthering our understanding of the economic universe. But is it?

The major issue that Granger hasn't addressed is not whether the old techniques have been generating valid parameters estimates, but whether such parameters exist at all. In the natural sciences, the employment of mathematics enables scientists to formulate the essential nature of objects. Consequently, within given conditions, the same response will be obtained time and again. The same approach, however, is not valid in economics. For economics is supposed to deal with human beings and not objects. According to Mises,

"The experience with which the sciences of human action have to deal is always an experience of complex phenomena. No laboratory experiments can be performed with regard to human action". (Ludwig von Mises, Human Action (1963), p 31.)

In short, people have the freedom of choice to change their minds and pursue actions that are contrary to what was observed in the past. As a result of the unique nature of human beings, analyses in economics can only be qualitative. In other words there are no parameters in the human universe. Thus Mises wrote,

"There are, in the field of economics, no constant relations, and consequently no measurement is possible." (Human Action p 55.)

The popular view that human economic activity can be captured by mathematical formulae expressed through fixed parameters implies that human beings are operating like machines. For instance, contrary to the mathematical way of thinking, individual outlays on goods are not "caused" by income as such. In his own context, every individual decides how much of a given income will be used for consumption and how much for savings.

While it is true that people respond to changes in their incomes, the response is not automatic, and it cannot be captured by a mathematical formula. For instance, an increase in an individual's income doesn't automatically imply that his consumption expenditure will follow suit. In short, every individual assesses the increase in income against the goals he wants to achieve. Thus he might decide that it is more beneficial for him to raise his savings rather than raise his consumption.

At the best mathematical formulations can be seen as a technique to provide a snap shot at a given point in time of various economic data. In this sense it can be seen as a particular form of presenting historical data. These types of presentations, however, can tell us nothing about the driving causes of human economic activity.

What's more the employment of established historical relations to assess the impact of changes in government policies will produce misleading results notwithstanding Granger's framework. After all, to assume that a change in government policy will leave the structure of the equations intact would mean that individuals in the economy ceased to be alive and were, in fact, frozen. In this regard Mises wrote,

"As a method of economic analysis econometrics is a childish play with figures that does not contribute anything to the elucidation of the problems of economic reality." (Ludwig von Mises, The Ultimate Foundation of Economic Science (1962), p 63.)

but I smell a philosophical rat.

Shostak's or Granger's?

Granger, however, contests [that any conclusions derived from the equation regarding the relationship between consumption and production are a reflection of reality.]
He argues that no meaningful conclusions can be drawn ... if the data ... are non-stationary. In plain English, it is counter-productive to establish meaningful conclusions from data that drifts aimlessly. The parameters that one will get from such data will be erroneous and hence the outcome of the analysis will be meaningless. So how does one overcome the problem?
Granger and others have shown by means of mathematical and statistical methods that the existence of the common factor makes the interrelationship between non-stationary time series stationary.
Now, instead of trying to find out what the co-integrating factor is Granger and others have produced a mechanised framework, which enables economists to establish whether data complies with co-integration i.e. whether the relationship between the data makes sense so to speak.
Once it is established that the data is cointegrated it can then be incorporated by means of a certain mathematical procedure to establish the correct parameters.
Various statistical results that are produced by means of Granger's framework are therefore regarded as valid since they have been applied on co-integrated data. Granger's discovery raises serious doubts about conclusions regarding economic interrelationships which are reached by means of the old techniques. It also provides a criticism of the popular usage of correlations without attempting to make sense of the relationships. Granger's framework can be seen as a preventative in minimising the use of meaningless correlations.
In this respect it could be regarded as bringing back the validity of fundamental analysis. This must be contrasted with the popular way of thinking that fundamental analysis is of little help since the data is of a random i.e. non-stationary, nature. So it seems that the Granger's framework is a great tool in furthering our understanding of the economic universe. But is it?
The popular view that human economic activity can be captured by mathematical formulae expressed through fixed parameters implies that human beings are operating like machines.

Granger is essentially only 'proving' a new kind of test to apply to previously correlated data, which test demonstrates that many or most past correlations were invalid.

Shostak goes the next step (as a good Austrian) to argue then while Granger's work is valid in eliminating old falsehoods, carried to a logical conclusion, no seemingly related phenomena not based on actual fundamental correlations will in the future pass Granger's new test, because human behaviour cannot be parameterized.

All "cause" is, is a high and repeatable correlation.

The presumption (indicated by your quotes, I presume) is whether a true cause can be known.

I suspect the truth is somewhere in between. That there are causal econometrics (based on largely repeatable and more or less predictable human behaviours - I give you the VIX, VXN as contrarian indicators for example) for which reasonably practical parameters can be discovered, approximated, and refined. Perhaps the ensuing discussion might be to suggest other 'causal' human economic behaviours that may be quantifiable.

Shostak's other point is that when a policy change is made, the causal parameters may have to be rediscovered all over again seem plausible. That seems a corrolary to our ealier agreement that having the Fed publish and follow 'rules' as opposed to allowing Greenspan undisclosed 'discretion'. Deviation from the 'rules' aids in discovery of the new causes/relationship.

If it's an empirical fact that dogs bark more as stocks rise, then so be it. The universe doesn't need to explain itself to scientistic rationalists who reject correlations for which they can't devise "causal" "mechanisms".

Granger's argument, I presume, might be:

If the data were collected during a period were the dog had fleas, then there is no co-integration and no true "causal mechanism" - it was mere (albeit obscure) coincidence. Granger's new framework, presumably detects and exposes this case.

However, if the dog is barking because it isn't fed or given attention, because it's owner is daytrading a rising market, then when the market falls, the owner is no longer daytrading and feeding the dog instead, the data collected should pass Granger's co-integration test and thus establish a basis for an econmetric parameter of barks/S&P-increment.

Shostak would argue, Granger's 'contribution' has only eliminated prior false correlations, and opened the door to new false ones, that being that unfed dogs bark when markets go up.

I would argue Grangers work helps us narrow our focus on those remaining seemingly causal mechanisms for subsequent fundamental evaluation for reasonablness.

So, while it may be an empirical fact that unfed dogs bark more as stocks rise, I would not run my company, economy, or fund based on that empirical observation, no matter how factual, because someone could starve dogs in my neigborhood trying to induce me to throw long for 'distribution' purposes.

Being right for the wrong reason is just as bad as being wrong.

Granger's work does not help in this regard. Granger's work conflates neither validity nor utility. It only eliminates invalid (though seemingly useful) prior correlations. Which elimination only leads to the additional discovery of the dog not being fed by its day trading owner as the causal mechanism.

Granger's work is pre-requsite, but not proof of causation.

Shostak argues, causation can't be known for aggregate human economic behaviour. Were that true, Technical Analysis would seldom work. But because TA is believed and used by so many, it has become in some sense a self-fulfilling prophecy, or in Granger's terminology: TA exhibits 'co-integration of a stationary relationship'.

Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.