Garbage in, garbage out – Management by Numbers

Since the seminal book of Oskar Morgenstern[1] on economic statistics, On the Accuracy of Economic Observations, which addressed the issue of the quality of National Accounts Data, the topic has been with us for a long time. Management by numbers has been a concept to introduce science into management, i.e. to become scientific management.[2]  Neil Postman[3] has in his book, Technopoly: The Surrender of Culture to Technology[4], pointed out that its origins has emerged in the military to organize the complex organization of an army. The principle introduced was reports writing, all the essential information to make decisions has to be included in reports. What was not written in reports did not matter. Human communication was considered to be unreliable, biased by personal interests. What was written in a report was objective in particular when the report contained lots of numbers and statistics. It led to a new kind of technocracy which substituted human verbal communications and debates by numerology.[5] As soon there are numbers, they seemed to have an objective character. Of cause manipulation of numbers and statistics happened, but there was strong believes that such things could be eradicated sooner or later.

With advent of the computer this became even more popular with Management Information System (MIS).[6] As Bill Gates once termed it, information at your fingertips[7], would be instantly available everywhere to everyone. There was a hope that all necessary information to make decisions would be available with the emergence of the Internet, Search Engines, etc. in the near future. The World Brain Project[8] envisioned by H. G. Wells[9] which would give all people access to the whole knowledge of the world. However, this kind of universal democratic access towards human knowledge is becoming more and more a chimera.

With the Internet becoming commercialized and more and more controlled by governments, the level playing field is biased more on more towards those who have access to privileged knowledge. The latest dramatic revelation was the uncovering of the NSA[10] espionage of the total global digital communication content via a system named Prism[11]. By this one could access all communication content of all users indiscriminately even if considered private.

It is not a surprise that George Orwell’s[12] negative vision of the Big Brother is Watching You[13], has become a popular term again. Asymmetric information[14], i.e. knowledge of a few opposite the rest, has become a strategic weapon in social interactions, as cyber wars[15], as cyber espionage[16], and ordinary economic interactions of markets.

In particular in financial markets privileged access to information often in the form of data play an essential role in the distribution of wealth, power and income. Often top decision makers feel like masters of the universe. However, they face a vulnerability of systemic failures. As the recent near breakdown of the global financial market system revealed painfully, the information created by models and data proved to be fallible. Wrong model assumption, i.e. model errors, made the applied model to predict or forecast events inapt. False data did the same. Garbage in, garbage out is the obvious outcome. Academics of all kind had a lot to explain why the allmighty systems failed so miserably.

One source of error is the intentional manipulation of data and statistics to influence decision making based on such data. As Charles Goodhart has pointed out: “Any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes.”[17] This fact, called Goodhart’s Law, invalidated the whole idea of scientific management. Since those who produce data as information for others to make decisions on it, they could use their control on the database to manipulate the data such that they could influence the decisions taken. The evidence that such strategic data manipulation has become endemic in accounting practices of firms, rating practices of rating agencies[18], market prices like the Libor interest rate[19] or exchange rates[20] and stock prices[21], more and more it becomes obvious that data are not what they intended to be, objective reliable information.

Furthermore Anasthasios Ophanides[22], currently governor of the Bank of Cyprus, pointed out that due to inaccuracies and delay in the availability of many macroeconomic statistics decision making failed to take the right decision due to misinformation.[23] In particular National Accounts are often miserably comparable between countries due to an improper implementation of Accounting Standards and/ or due to manipulation for political puposes.[24]

This kind of critical consciousness that data are often not just correct representations of the real world is rarely told to students at universities and the general public. Instead a naïve perception that data are more or less correct and can be used without reflections on their quality and potential manipulations through agencies with a high formal reputation is not part of their training. Instead academics focus all their effort to learn the last up-to-date statistical and econometric methods to foster their academic careers. This has become a major problem. Without any understanding about the implications of bad data quality on their research and results, they just ignore the problem. It is just a not-in-my-backyard, NIMBY, attitude. Nobody could be blamed for mistakes he are not directly responsible for. However, this leads to a catastrophic disconnect between data producers and data users.

After all this growing misdirection of the division of labor in the design of quantitative decision making  between those who make decisions, those who create models to make decisions, and those who produce data, it is time for a debate how to overcome these problems.

The growing complexities have led to an increasing vulnerability of complex systems.[25] Often people lack the common sense to see the obvious just right in front of their eyes. They have not learned to use simple heuristics to check the validity of results to match their guts knowledge. Gerd Gigerenzer is one of those who have challenged this seemingly inevitable trend to ever increasing complex decision making systems. Instead he is pleading for a more heuristic based approach to decision making.[26] Probably we are reaching a time were a new paradigm is taking hold. Instead of the belief in complex systems which are more and more unreliable, common sense, human communication between mortals are getting more important again. The technocratic approach towards scientific management and decision making becomes obsolete. In particular in decision making in uncertain environments risk-averse common rules are a better hedging against fat tale risks.[27] The more and more fragile global environment needs new strategies of committing towards antifragile solutions.[28] They may be suboptimal from a mathematical point of view, but they are more robust to obtain a sustainable antifragility of systems. Ancient architects could not calculate the stability of their buildings so they used safeguards to protect them from collapsing. Therefore they lasted for centuries even millenia. Nowadays this is not anymore the case. Miscalculations lead to collapsed buildings around the globale. Look e.g. at China or India and you know what I mean.

Smaller, less complex, simplified environments are more robust and easier to manage. Less dependency creates less vulnerability. If such an attitude would become more popular in our societies, we will probably end in a better world. Less stress, less burn outs, less possibilities of disinformation because one does not need it. Simplify your personal environment to make your life and that of all others easier. Furthermore it is a new way to regain more individual freedom. New regionalism and less complex life-styles are already trend setters. Humanize a society which has become more and more inhumane and technocratic.

[8] H.G. Wells (1938): World Brain – H.G. Wells on the Future of World Education, Adamantine Classics for the 21st Century, 1994, UK.

[23] Orphanides Athanasios (2001) ‚Monetary policy rules based on real-time data‘, „American Economic Review“, 91(4), September 2001, pp. 964-985

[24] Brian Sturgess (2010): Greek Economic Statistics: A Decade of Deceit – SO how come the rating agencies missed it again?, in: World Economics, Vol. 11, No. 2, pp.67-99.

[26] Gigerenzer, G. (2008).Rationality for mortals: How people cope with uncertainty.New York: Oxford University Press.

16 Gedanken zu „Garbage in, garbage out – Management by Numbers

Kommentar verfassen

Trage deine Daten unten ein oder klicke ein Icon um dich einzuloggen:

Du kommentierst mit Deinem Abmelden / Ändern )


Du kommentierst mit Deinem Twitter-Konto. Abmelden / Ändern )


Du kommentierst mit Deinem Facebook-Konto. Abmelden / Ändern )

Google+ Foto

Du kommentierst mit Deinem Google+-Konto. Abmelden / Ändern )

Verbinde mit %s