Presented at the Eastern Economic Association Meetings, March 1995




This paper derives the measure of likelihood associated with the hypothesis of God's existence. God is defined as an omnipotent, omniscient being. These words are given econometric definitions. The probability of an event is decomposed into its rationality (a measure of subjective preference) and nomicity (a measure of subjective influence). The likelihood of an event is thus a convolution of rationality and nomicity, as a function of the data. The existence of God is thus represented as the state of perfect rationality (omniscience) and perfect nomicity (omnipotence). It is shown that this is the maximum likelihood model for any data set. Hence, the most statistically sound explanation of behavior is "God wanted it that way." However, the ultimate purpose of econometrics is not explanation or prediction, it is to increase the utility of the agent using the analysis. As such, the theory that God causes all behavior may not be economically useful, and is put aside for other theories/models with less likelihood but greater nomic value. The implications for the neoclassical approach to methodology are also considered.



Why would anyone believe in the existence of God? The skeptics seem to have relegated God to the role of psychological crutch, illusionary benefactor or an opiate. The whole notion of a supreme, creator-being is presented as wishful thinking. Beyond this, the use of God within scientific explanation has traditionally been frowned upon as an unnecessary complication. However, many people still believe, with great conviction, that God exists - based on the argument that no other explanation of the origin of the universe is as compelling. I suggest there is abundant empirical evidence for God's existence. The key is developing the appropriate model that allows econometric principles, such as maximum likelihood, to be invoked.

The first step is to clarify what is meant by God, from an economic perspective. I suggest that God be conceived as a Generalized Objective Decision-maker (Axelrod, 1993). Because this is an economic paper, the ultimate force in the universe is a decision-maker. A generalized decision-maker is one capable of making any decision. An objective decision-maker is one which can make decisions using only objective information, or possibly beyond any individual subjective experience. In essence, GOD is the ideal to which economic theory aspires.

The second step is to recognize the paradoxical nature of the basis of the scientific method, the controlled experiment. Clearly, the greater the control one has over an event the less "natural" it is, and the greater a reflection of one's preferences, the observed behavior becomes. This insight is expressed in much of twentieth century physics (as in quantum physics). The controlled experiment is not to be aspired to; it is a second-best method for gaining knowledge that is used because the best method, perfect intuition is unavailable. Ultimately, we use the scientific method due to pragmatic concerns. We develop econometric models for the same reasons that DaVinci invented the telescope - to see worlds hidden to the unaided eye.

A compelling expression of this paradox is presented in an article entitled "Statistical Analysis and the Illusion of Objectivity" (Berger & Barry, 1988, American Scientist, pp. 159 - 165). Here they state that data alone is insufficient for determining the significance of results, one must know how the experiment was run. They deduce that "... If the investigator died after reporting the data but before reporting the design of the experiment, it would be impossible to calculate a P-value or other standard measure of evidence."

It is also instructive to consider an article by Kruskall (1988, JASA, 929-939) "Miracles and Statistics: The Casual Assumption of Independence." He provides a review of the use of statistical analysis with respect to religious questions, such as efficacy of prayer. He points out that many "miraculous" results are inferred when relevant correlations (similar to selection-bias) are ignored.

In statistics, likelihood has a technical meaning. It is a function of a data set, based on the probabilistic characteristics of the model being used to analyze that data. It is a measure of how well the model fits the data. The greater the likelihood of the data given the model, the better the model. Thus, the model (and, in particular, the value of its parameters) with maximum likelihood is understood as the best.

If we are to develop a testable model of God's existence, from an econometric perspective, we will need to capture God's characteristics as a decision-maker. Our next step will be to consider the most fundamental economic relationship, that which exists between persons and things. This dichotomy is the ontological basis for all of neoclassical theory.



Walras described persons as "self-conscious" and "self-determining." Things, he wrote further, are "not self-conscious" and are at the command of persons. This relationship between persons and things which underlies current economic thought (and if you don't believe this, just consider how you might explain to a freshman what a good is and what does it mean to own a good) is fundamentally a metaphysical relationship. Command, control and causality are the explanatory terms we use to motivate a meaningful understanding of economic relationships.

Persons are the forces that determine the state of things. Persons are thus hypothetical beings. We infer their existence by the observed behavior of things. To bring the point home, a dead body, for all practical purposes, is not considered a person, even though a living one is. This is due to the lack of behavior to support there being a person whose volition is expressed in the body. This is not a mere philosophical statement. Our laws will divest property rights (and responsibilities) from a dead body, i.e. it ceases to be treated as a person. On the other side of life, there is much debate over the personhood of fetuses. Further, the history of law and politics is rife with emancipation and suffrage. These are clearly changes in the formal association of personhood with the human body.

Walras based his economic theory on the paradigm of scarcity, often viewed as a quantitative constraint on resources available, which implies behavioral limitations. Walras contended that the material world is essentially scarce. How is this consistent with "self-determination?" In trying to comprehend the above a difficult question arises, how do we know what we can, or can not, do? What events are realizable? Further, how do we distinguish between the limits of nature and lack of will. We may look to the past and say, if we didn't do behavior X we couldn't do it. However, we then must argue for why we can't behave as X now. Similarly, for looking to past behavior to determine what we wanted to do. Indeed, we do not even need to invoke scarcity or any limitations whatsoever in our explanations of behavior, the phrase "she willed X", without any reference to constraint, is sufficient. These issues find practical expression in courses taught by self-empowerment organizations such as LandMark Education So, why do we believe in scarcity? Because we have had experiences of willing X and having X not being realized, because it is self-evident.



For an entity to be a person (i.e. have control over things) means, in an econometric sense, that the probability of an event which encompasses the state of the person's things, is conditioned on the will of the person. Symbolically, we represent this statement as follows:

X is the event

Xw is the person's will

N(X) = P(X|Xw)

We shall refer to this distribution as the nomicity of the person. This representation assumes that the event and person are known without confusion.

What underlies the interpretation of this equation is that the universe is composed of awareness, the use of physical objects in discussion is merely a tool for succinctness of discussion. The only relevant control of things for persons is one that alters states of awareness. Moreover, when one ascribes personhood to an entity, based on some empirical evidence, it means that the entity's will has some effect on one's consciousness. The idea is that each individual has their own data set.

We can also discuss a persons power over events, as a function of the person's nomicity. The definition used here is


M = (N(X) - P(X)) / (1 - P(X))

Assuming that a person's nomicity is not pathological (i.e. if they will a realization they do not decrease the probability of it occurring), then M takes on a value between 0 and 1. This represents the fraction of probability of X not "naturally" occurring converted into the probability it does occur. A person is powerful if M is greater than zero.

In more formal language, the universe is composed of observations that are observer specific. Describing an observation as objective means that all observers realize measures of the associated event with almost perfect correlation. The only relevant nomicities for economic study are those which are powerful.

What drives the realization of an event are persons' wills. However, a person's will may not be (and often is not) singular. This dispersal of will also creates uncertainty in the outcome of events. This dispersal can be understood as a person's uncertainty with respect to the optimality of alternative realizations of an event. Thus, when someone goes to the market to buy goods, one reason that economists are unable to perfectly predict behavior is that the consumer himself/herself does not have a singularity of will. This distribution of will shall be referred to as the persons rationality, and is represented as

R(X) = P(Xw)

Hence, the probability of an event is the integral of nomicity times the rationality (or the sum of nomicities weighted by the rationality),

P(X) = ò N(X)*R(X)

Letting J represent the universal awareness (i.e. all perspectives/data sets) of an event, we have

P(J ) = ò N(J )*R(J )


P(X1,X2,...,Xn) = ò R(X1,X2,...,Xn) * N(X1,X2,...,Xn)

The interpretation of the previous equation is, the joint probability of all subjective realizations of an event is equal to the joint nomicity (what a society is likely to do given its individuals' wills) multiplied by the joint rationality (what a society's wills are likely to be).

The person in this theory is thus understood as an "invisible hand", a metaphysical being, identical to a probability distribution (or density), whose will and power are constructed from the decomposition of that person's influence on realized events. Empirical evidence is analyzable directly in terms of wills and feasibilities.


In traditional neoclassical theory whatever a person wills is implemented, where the will is determined by prices and preferences. This is seen in normative statements suggesting people behave in the way that is best for them (i.e. their market behavior is optimal). This is essentially the assumption of

P(X) = R(X),

where the distribution of R is understood to be over different persons. This implies that nomicity is the degenerate distribution at the point of will,


N(X) = 1 if X = Xw, 0 otherwise

Thus, when we generalize N(X) to be other than non-stochastic we are essentially implying that what happens is influenced by one's will, but is not necessarily one's will.

It is important to recognize that neoclassical economic theory makes this assumption. By doing so it allows us also to hypothesize a person with perfect nomicity. Indeed, it begs the question "how can there be more than one person if nomicity is non-stochastic?"

The Likelihood of God

The essence of belief in God derives from the insight that any other causative being(s) will be associated with a probability of less than one of "making things happen," and hence a lower likelihood for the events we do observe. It is as if the existence of doubt violates Occam's Razor. Anything unexplained can be understood as requiring an infinite number of hypothetical entities to explain it (this is analogous to an economist saying that an impossible outcome is understandable as an alternative with infinite cost).

With this in mind it becomes clear that a hypothesis that is built around a single (and exhaustive) causative agent would have great explanatory appeal. Indeed, monotheism, when it was first conceived, could be seen as both the beginning and the pinnacle of modern western thought. Every other model of the universe thereafter must demonstrate its value through non-explanatory criteria (e.g. predictability, practical use). Once humans invented "God" there was nothing left to do but deconstruction.

The hypothesis of God's existence is succinctly stated as the case when there is a single decision-maker whose nomicity and rationality are both degenerate (i.e. all mass at the observed behavior). Hence, R(X), N(X) and P(X) are all 1. The conclusion is that the likelihood of God given any dataset is one. Moreover, this is the maximum relative to any other competing theory/model.



If the existence of God is the most compelling explanation for any set of events, why have rational thinkers gone to such efforts to develop models of the world without God? I suggest that it is a continuation of our efforts to perceive ourselves as having some control of our own lives. For example, it is a fundamental principle of economics that expanding a decision-makers feasible alternative set (so as to include all previously available alternatives) is always associated with an increase in utility/welfare of that decision-maker. This is clearly false (consider that one can end up with enough psychic losses from the more demanding search for optimality to overwhelm whatever marginal gain in material wealth is acquired).

We can go further. The uncertainty of the modern era can be seen as a direct outgrowth of pursuing nomicity. This is a logical implication of an individual having non-zero nomicity. Such a person would have control over some variable that has yet to be determined. It is as if we created uncertainty so that we may experience a sense of power. Without a space for possibility there is no control. As any marketer can tell you, create the image of a different (and better) world in the mind of the consumer and it is as if they are compelled to demonstrate their autonomy. This is an insight so fundamental it is used as part of the Genesis story describing how mankind's plight has come to be.




This last insight is critical in understanding the methodological crisis developing in neoclassical economics. If economists are attempting to understand the economic world they can not exclude themselves. Hence, by traditional criteria they should be able to predict which methods will be used in the future. Further, they should be able to predict which methodological criteria will be used in the process. Clearly, such ability to predict is inconsistent with the assumption that economists are currently choosing the optimal methods and criteria. Well, not quite. The only way out is to understand methodology as merely another behavior to model without normative bias. But, this would imply that we can not argue for the superiority of any method of analysis, and hence could not distinguish between good/useful and bad/useless research. Either way the positivist neoclassical approach to the study of economics is self-contradictory since it can not simultaneously rationalize its own existence and be able to predict how economists will think (and, hence, analyze and measure) in the future. The obvious solution would be to admit that God even determines what methods we use to understand the world. But then what would we do?

The essence of the difficulties is the free will/determinism question. To arrive at a model with "perfect" explanatory power requires the assumption of a deterministic world (i.e. one run by god, or something god-like). However, the economic view is built around a perception of having some control of the world. We thus reject the hypothesis of God's existence. It is important to recognize that we do this not because of any data, statistical or scientific analysis. We do it because it is self-evident. Moreover, the use of a deterministic model along with a neoclassical methodology (where the methods used are understood as chosen and optimal) is logically inconsistent. One can not simultaneously assume that one's choices are predetermined, that one has come to conclude this through rational-willful deliberation, and that one is being logical. From a broader perspective, the hypothesis of a deterministic world is inconsistent with the belief that we are at choice with respect to the hypothesis we make. There is no meaningful, or useful, scientific inquiry in a deterministic world. However, there is no purely scientific/objective inquiry in a world at choice. Neoclassical methodology remains incomplete or inconsistent, or possibly both.



The hypothesis that God exists is supported by the very existence of data. It is as sound an explanation as any. However, its usefulness is questionable, since it assumes the powerlessness of the analyst. Thus, from an econometric standpoint we do not use this hypothesis in our research. We can either be certain of our place in the universe, or we can have the power to change it, but not both.

This paper does suggest that the objectivity of economic research becomes compromised when the neoclassical researcher does not include their own utility function/preferences in their work. This seems to imply that the first person that needs to be analyzed by the neoclassical researcher is the researcher himself, and further implies a breakdown of traditional neoclassical methodology. The researcher must be able to rationalize the constraints they face, or else defend himself from the criticism he is already assuming what he wants to conclude.

And yet, if we do not assume our constraints are determined beyond ourselves we can never make a rational choice. Do we choose to believe in God, or is that belief pre-determined? Can God-services be made a contestable market?





Ainsle, G.: 1986, `Beyond Microeconomics: Conflict Among Interests in a

Multiple Self as a Determinant of Value', The Multiple Self,

ed. J. Elster,Cambridge University Press, Cambridge

Axelrod, David A.: 1991, The Econometrics of Persons: Constructing The Invisible Hand

Becker, G.: 1976, The Economic Approach to Human Behavior,

University of Chicago Press, Chicago

Box, G. and Tiao, G.: 1973, Bayesian Inference in Statistical Analysis,

Addison-Wesley Pub. Co., Philippines

Davies, P. and Gribbin, J.: 1992, The Matter Myth , Simon & Schuster, New York

Dixit, A. K.: 1984, Optimization in Economic Theory, Oxford University Press, London

Elster, J.: 1979, Ulysses and The Sirens, Cambridge University Press, Cambridge

Frank, R.: 1987, `If Homo Economicus Could Choose His Own Utility Function,

Would He Want One with a Conscience?', American Economic Review, 77(4), 593 - 604

Freud, S.: 1938, A General Introduction to Psychoanalysis,

Garden City Pub. Co., New York

Friedman, M.: 1953, Essays in Positive Economics,

University of Chicago Press, Chicago

Gigliotti, G. A.: 1988, `The Conflict between Naive and Sophisticated Choice as a

Form of the `Liberal Paradox'', Theory and Decision, 24(1), 35 - 42

Jaynes, J.: 1990, The Origin of Consciousness in the

Breakdown of the Bicameral Mind, Houghton-Mifflin Co., Boston

Mood, A., Graybill, F., & Boes, D.: 1974 Introduction to The Theory of Statistics McGraw-Hill, New York

Nelson, Robert H.: 1991, Reaching for Heaven on Earth , Rowman & Littlefield, Maryland

Schelling, T.: 1984, Choice and Consequence, Harvard University Press, Cambridge

Simon, H.: 1976, `From Substantive to Procedural Rationality', in

Methods and Appraisals in Economics, ed. Latsis, S. J. Cambridge University Press, Cambridge

Simon, H.: 1978, `Rationality as Process and Product of Thought',

American Economic Review, Proceedings 68, 1 - 16

Simon, H.: 1987, `Rationality in Psychology and Economics', in

Rational Choice, ed. Hogarth, R. and Reder, M., University of Chicago Press, Chicago

Tversky, A. and Kahneman, D.: 1981, `The Framing of Decisions and the

Psychology of Choice', Science, 211, 453 – 458

Tversky, A. and Kahneman, D.: 1987, `Rational Choice and the Framing of Decisions', in

Rational Choice, ed. Hogarth, R. and Reder, M., University of Chicago Press, Chicago

Walras, L.: 1977, Elements of Pure Economics,

Translated by W. Jaffe, A. M. Kelley, Fairfield