Home > The Economics Profession > Rethinking the basic assumptions of economics

Rethinking the basic assumptions of economics

from Lars Syll

Economics has long had the ambition to become an “exact science”. Indeed, Walras, usually recognised as the father of modern economic theory, said in his Lettre no. 1454 to Hermann Laurent in Jaffe (1965):

“All these results are marvels of the simple application of the language of mathematics to the quantitative notion of need or utility. Refine this application as much as you will but you can be sure that the economic laws that result from it are just as rational, just as precise and just as incontrovertible as were the laws of astronomy at the end of the 17th century.”


Furthermore his successors openly declared themselves as having the same goal. However, two things raise doubts as to whether the pursuit of this ambition has achieved meaningful this phenomenon). First, as in any science, models have to be built on assumptions, and it is a standard procedure to develop those assumptions on the basis of a careful analysis of the observed empirical facts. This inductive approach, however, is not the one prevailing in economics, where widespread assumptions are based on the introspection of economists. This has been acknowledged by many distinguished economists from Pareto (1916) to Hicks (1939) to Koopmans (1957), for example. Second, and perhaps worse, the reference model in economics is one with isolated optimizing individuals. This model of “perfect competition” is considered as a useful idealization, and features such as the aggregate effects of the direct interaction between individuals are thought of as inconvenient “imperfections”. However, deviations between economic theory and reality may be of crucial importance in practice, and the consideration of the links between individuals and institutions cannot be written off as being of little relevance to the behaviour of the system as a whole. This is a lesson that is clear to all those, who are familiar with the analysis of complex systems. Given the systemic impact of certain financial instruments (such as large leverage effects, the market for credit default swaps, etc.), it would seem to be unreasonable to put too much trust in conventional economic models, in which the structure of the interactions between the participants in the system is not included in the underlying assumptions.

Alan Kirman & Dirk Helbing

  1. Lyonwiss
    November 11, 2013 at 2:22 am

    The government is ultimately responsible for the economy, because the government has the legislative and enforcement power to direct economic activities. Failure of the central planning economy has allowed markets in recent decades to play a greater economic role in resource allocation even in communist countries like China. The reason is genuine markets, not manipulated, provide important economic information which is impossible to collect by a government bureaucracy.

    The government has the power to control the extent and function of markets through regulation and law enforcement. The failure of market regulation and subsequent over-reaction of deregulation are due to appalling ignorance of how markets actually work, a situation brought about by the rhetoric of the “invisible hand” in economic education. The extreme faith of the “free market” held by the Austrian school and neoclassical school are sadly misplaced.

    Competition, between humans e.g. in sport or in markets, requires rules and referees to enforce those rules. Also, with evolving competition, rules need to be monitored, reviewed and modified. It is the role of government to referee competitive markets in an open and rational manner. It is not the role of government to be a major player in competitive markets or to side with particular competitors or to manipulate market prices, because all these actions completely defeat the purpose of markets. Yet this is what governments are doing.

    Government bureaucrats and politicians are interfering in markets, because markets are considered to have failed and their ad-hoc actions are justified by the Keynesian doctrine that the government could, and should, direct the macroeconomy. But historical failure of Keynesianism has been conveniently ignored. Central planning and Keynesian economics failures are due to the lack of adequate economic data or theory to run a complex modern economy under central command.

    The government bureaucrats and politicians in charge of creating policy are equipped with a false economic education. They see the government and the market as a false dichotomy, propagated academically by the adversarial views of market-based neoclassical school versus government-based Keynesian school. So long as this false dichotomy remains unresolved with creation of a more comprehensive synthesis, our world will remain in a state of economic schizophrenia.

    The education system has produced a set of misguided government leaders who have been divided and conquered by vested interests at the expense of ordinary people.

  2. November 11, 2013 at 9:53 am

    Economics will never be a science until it deals honestly with certain basic facts and assumptions:
    1. What is the economy for? The way economists today describe it, it is value-free, that is, devoid of concern for human beings, just for price and demand (actually effective demand, since that’s all that counts, in Economics) models.
    2. What are the factors of production? Labor, Capital and…wait for it… Land. Without the last one, it’s like talking about physics without heat or gravity. And no, Land is not a form of Capital, as it has been wrongly conflated with for over 100 years; it’s almost the opposite.
    3. What is money and who should be allowed to create it? Is a private monopoly on money-creation the best model? What about a Public Option for money, aka Sovereign Money, or debt/interest-free money?
    4. Following up on #3, who should be allowed to created credit? Just the private sector, or should there be a Public Option for credit in the form of Public Banks too?
    5. How much money is actually available? This basic accounting fact is not as straightforward as it seems. There are 10s of trillions in government assets (described in the CAFRs), yet governments say they are “broke.”
    6. What are the true costs of “externalities,” conveniently ignored by economists, but not by the planet? Who pays the cost of these?

    This is just the beginning. As Keynes noted, mathematical models based on nothing add up to nothing.

  3. davetaylor1
    November 11, 2013 at 11:36 pm

    The abstract of the Kirman-Helbing paper Lars is drawing our attention to lists “some of the standard assumptions and postulates of economic theory”:

    1. An economy is an equilibrium system. In other words, it is a system in which all markets systematically clear at each point of time, but where the equilibrium may be perturbed, from time to time by exogenous shocks.

    2. Selfish or greedy behaviour of individuals yields a result that is beneficial to society – a modern, widespread, but inaccurate reformulation of the principle of the “invisible hand”.

    3. Individuals and companies decide rationally. By this it is meant that individuals optimize under the constraints they are facing and that their choices satisfy some standard consistency axioms.

    4. The behaviour of all the agents together can be treated as corresponding to that of an average or representative individual.

    5. When the financial sector is analysed, it is assumed that financial markets are efficient. Efficiency here means that all the relevant information concerning an asset is reflected in the price of that asset.

    6. For financial markets it is assumed that they function better if their liquidity is greater.

    7. In financial markets, the more connected the network of individuals and institutions the more it reduces risks and the more stable and robust is the system.

    Let me add some even more fundamental – but UNSTATED and generally unrecognized – assumptions:

    8. Logic is the static Aristotelian one relating causes and effects, including words and meanings.

    9. Science and mathematics are quantitative [because Hume said so in 1740].

    10. Words refer only to observables or observations [which can be referred to objectively and transparently, but adverbial qualities and verbs are meaningless], while mathematical functions map quantitative inputs to quantitative outputs of quantitative dimensionality [using the quantitative arithmetic of 1740, unconsciously disregarding the coordinates and geometrical meanings of continuity, unity, differentiation and variability].

    11. Economic and political science cannot speak about invisible causes of change, only about statistical relationships between already observed concomitant events [because Hume said so in 1740].

    With unobservable Aristotelian justice not explicable as the aim of a balance of unobservable Newtonian forces, would-be scientific economists have resorted at (1) to visualising [i.e. imagining] Smith’s concept of [control by] “an invisible hand” as the equilibrium level water reaches in both sides of a U-tube, or returns to in a pond as ripples generated by dropping in a stone die away. Neo-classicals followed the fashionable (if unseen) Maxwellian explanation of constant pressure in a steam boiler as the random directions of motion of hot atoms cancelling each other out. (More acceptable than explaining how boiler pressure is maintained by increased steam usage drawing more air through the fire to make it burn more quickly, requiring the fireman to work harder)? I regret to say Kirman and Helbing are changing the fashion rather than the metaphor when pursuing Complexity Theory for an explanation of the random motions.

    The paradigm shifts required here are from physical work adjusting itself automatically to [not necessarily true] communication motivating people to adjust their own work, and from the static 300 b.c. word logic of (8) to the dynamic logic of information channels in the post-1948 form of cybernetic [steering] control logic, wherein the “invisible hand” of (2) is motivated not by greed but by PID information feedbacks about errors in direction, directing and risks of motion. Arguably, Keynes anticipated this in 1935, being “off course” creating errors in price steering for which the correct prescription was “sideways” movement to get back on course. Very confusing for economists when, at (10), their arithmetic does not provide for sideways movement – or indeed movement.

    At (3) , individuals may be constrained by their abilities and/or situation to be specialists, acting more or less rationally in pursuit of the aims of their specialism, not for that of the system as a whole but as director, course-setter, steersman, course-corrector and anticipator of danger, or corresponding physical activities of empowerment, servicing, operation, repair and maintenance. At (4), an average or representative individual is a specialist reliant on others; the very few who cross the oceans single-handedly are exceptional, and even they have different priorities at different times.

    (5) contradicts the first theorem of information theory: that you can’t put an quart into a pint pot. (6) contradicts the dynamic version of that: you can’t travel faster than the speed of light (i.e. faster than the physical system operates). (7) Financial markets thus contradict (a) the logic and (b) findings of PID error control. (a). If all the participants have the same aim of making rather than investing money, the more interconnected they are the more likely they are to go with the flow and ignore the non-monetary PID correctives, enabling prices to be pushed up where profit-generating investment is not needed and depressed where it is. (b) The faster the flow the less time there is to carry out error correction before the situation has changed. Automation of international speculative trading has defeated rational Keynesian smoothing of “swings and roundabouts”, reducing moderation to “seat of the pants” reactions which (at critical speeds) can reverse the phase of corrective feedback: driving the system, as we have seen, into the unstable stability of overload or shutdown.

  4. bruceedmonds
    November 14, 2013 at 3:15 pm

    The article is absolutely right. One of the ways economists have avoided the responsibility of making empirically adequate models/theories is by conflating predictive & explanatory models.

    (1) A predictive model can be a black box and so its workings do not have to be realistic (though if successful this suggests finding out why it predicts), but has to be tested on data that is *unknown* to the authors. Testing such models using in-sample/out-of-sample data is not good enough because the temptation to fit the out-of-sample data is too great.

    (2) An explanatory model establishes an possible explanation of some outcomes from some mechanism. Here the micro-specification has to be compatible with known micro-facts otherwise one gets an irrelevant explanation. This kind of model can be tested using in-sample/out-of-sample data.

    Economists use models whose micro-level specifications are unrealistic (their agents do not behave as humans are observed to) and hence fail (2) but also fail to predict *unknown* data (hence fails 1). Thus they fail as either explanatory or predictive models. They save face by testing their micro-unrealistic models using in-sample/out-of-sample and pretend this is good enough. This scandal of economic modelling has allowed them to fool themselves that they are making progress. Basically they are bad science dressed up as hard science.

    [Mature science connects 1 and 2, but this is not necessary for some good science]

    • davetaylor1
      November 14, 2013 at 10:12 pm

      Bruce, I’m not sure what “The article …” here refers to, and I cannot relate your (1) and (2) to what I see in Lars’ extract on Basic Assumptions or indeed anywhere in the full Kirman/Helbing paper. Is it a reaction to anything I wrote about explanation? I’d be obliged if you would give us a clearer lead, since this is potentially a constructive discussion. Are we at cross purposes on science: my interest being in theory generation and design of practical applications via paradigms of method, and yours in the testing phases?

      For my part I would like to clarify my objection to the direction of (as against much admirable and constructive detail in) the K/H paper. I am not objecting to “Rethinking Economics USING Complexity Theory” (for indeed the authors use it to characterise very helpfully some issues and possible resolutions in the current state of our economies), but to rethinking it AS a complex system. That is to see [or model] economies in general as generators of chaos, which (unhelpfully) can be true even when their proper or necessary function is coordinating the regeneration and distribution of goods and services.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.