1
Modeling: The Explanatory Model and the Individual Predictive Model

This first chapter begins by making preliminary assumptions (section 1.1) of the modeling, first, of an explanatory model (section 1.2) and second, of an individual predictive model that derives from it (section 1.3).

1.1. Preliminary assumptions

In this section, three assumptions are made: the first relating to the nature of uncertainty, the second to the nature of consequences and the last to the nature of investor rationality.

1.1.1. Nature of uncertainty

The central question of our problem concerns the determinants of the choice of projects by individual investors. However, a choice between several alternatives is a decision; therefore, this research is indeed centered on a "decision". The consequences of this decision are not deterministic; they are uncertain and depend on hazards. The question that arises from this observation is that of the nature of the uncertainty at stake, a question analogous to that posed by the theory of individual decision that, according to Kast (2002), aims to "propose a framework for studying rational behavior in the face of uncertainty by distinguishing different types of uncertainties".

Uncertainty taken in the common sense that concerns us, "the unforeseeable nature of the result of an action"1, is specified by scientific literature, in particular, by the oft-repeated distinction, made by Knight (1921), between risk and uncertainty. For the economist, risk characterizes decisions for which the consequences are probabilizable, either by a priori calculation or on the basis of statistics. Conversely, uncertainty characterizes decisions that are not probabilizable, due to the uniqueness of the situation.

According to Le Heron (2012), uncertainty is not probabilizable in the social sciences, because human decisions do not come under the law of large numbers. Moreover, the decisions taken change the state of the world so that their repetition is not possible.

In the social sciences, in the field of entrepreneurial finance, in the venture capital sphere, the question of selecting projects financed by the venture capitalist (screening) is not far from our problem. Differences are due, on the one hand, to the sophistication of the venture capitalist - a true financing professional - and, on the other hand, to the stage of development, after the start-up, of the company awaiting funds.

However, Dubocage (2006) noted the extent of the uncertainty that venture capitalists face. The latter can base neither their evaluation nor their selection on objective probabilities. In fact, the business plan offers a limited quantified vision of the future to understand it and analog methods run up against the singularity of the evaluated companies, which are generally carriers of disruptive innovation. The author also decided to use the uncertainty framework proposed by Knight in order to theoretically clarify the mode of judgment of venture capitalists.

This approach corresponds to that of Keynes (1936) according to whom decisions relating to distant future horizons cannot, in most cases, be based on mathematical calculations, due to the lack of data, which can be scarce or even nonexistent. Decision-makers then make the choice they can: "[...] choosing between the alternatives as best we are able, calculating where we can, but often falling back for our motive on whim or sentiment or chance".

For these reasons, Keynes developed the concept of state of trust, that is to say, a subjective evaluation of the probability of occurrence of a future event that constitutes, according to him, a more realistic basis of appreciation than that of mathematical calculation to describe the method of determining the actions of entrepreneurs.

We can compare Keynes's concept of state of trust with the successive works of Ramsey, De Finetti and Savage, all three of whom were critics of frequentist probability and who developed the concept of subjective probability. These represent a judgment on the trust placed in the realization of an uncertain future event (Kast 2002).

We choose to place ourselves in the uncertain; there are indeed no objective probabilities associated with the consequences of investment choices. However, subjective probabilities could be associated with them. But what consequences are we talking about? The next point clarifies our view on this subject.

1.1.2. Consequentialism

It is important to clarify that in our approach, decision-making investors are not exclusively or even necessarily consequentialist, that is to say that while the consequences can be taken into account in their choice, it can also rest on the intrinsic value granted to the actions (investments, in our case) themselves, independent of their consequences, which amounts to saying that the rationality of investors is located between two areas of rationality: instrumental rationality and axiological rationality. As for the consequences to be considered, we consider that they are multifold because they depend on the objectives of the individual. Box 1.1 sheds light on the nature of these consequences.

Box 1.1. Consequences that matter

First, in accordance with the recommendation of Kast (2002), it is essential to form hypotheses about the objectives that the decision-maker wishes to achieve:

Economic theory is built on the description of the behavior of agents (consumers, producers); management models must make hypotheses about the representation of the objectives to be achieved.

In the description of the issue in the first chapter of Volume 1, we indicated that the firstorder choice was determined by investors' objectives, which are themselves associated with the investors' motivations, whether financial and/or not financial and also that we make hypotheses of financial and/or non-financial objectives quite naturally.

Second, it seems natural to us to assess the consequences of decisions in the light of these objectives. As for uncertainty, it relates to future events that determine the consequences of decisions.

For the sake of clarity, let us illustrate this point. Let us consider an equity crowdfunding investor who is firstly motivated by an altruistic impulse and secondarily by an enrichment motivation. Above all, they want to help a cause that is important to them and ideally hopes to get a financial benefit without it being their primary investment motive.

We have:

Motivation 1 = altruism Motivation 2 = enrichment To these two motivations, we associate two objectives, respectively:

Objective 1 = defend a cause via the funded project Objective 2 = improve the profitability of their portfolio of financial assets The pair of objectives (Objectives 1 and 2) has already determined the choice of investment in equity crowdfunding rather than in any other asset class (first-order choice).

After choosing one project rather than another on a platform (second-order choice), the investor will assess the consequences of their choice with regard to the objectives (Objectives 1 and 2). In other words, the evaluation of the choice relates to the consequences that matter to the investor; it consists of answering the following questions:

• - will the financed company allow for progress for the supported cause?
• - will the profitability obtained be satisfactory?

Thus, there are a multitude of possible combinations of objectives, and therefore of consequences, for each investor, the latter being moved by specific motivations.

1.1.3. Rationality of the equity crowdfunding investor: a triptych of rationality

According to Buttard and Gadreau (2008), in the standard paradigm:

Instrumental rationality characterizes agents guided in their choices by the selfish search for personal utility. It makes reason a tool for selecting actions that generate utility, the individual a determined, normalized, and selfish being, and the organization (firm, market, or a hybrid) an entity that can be reduced to the agents that make it up.

Thus, the agent endowed with an instrumental rationality maximizes its utility. While this behavioral postulate is simplified to an extreme, it has the advantage of making it possible to model decisions. Our approach deviates from this paradigm, but we keep the idea of an agent in search of an optimum. However, as we will define below, the (local) optimum of our individual predictive model does not relate to selfish utility and is not necessarily determined by reason.

Furthermore, we can contrast economic science and its model of the homo economicus, giving a single teleological explanation of behavior, with other social sciences, such as psychology or sociology, which are interested in the multiple causes of behavior (Viviani 1994). Our approach is intermediate, as with economists; the individual has motivations oriented toward objectives or goals, but these are not the only factors explaining their investment choices. The preferences of the individual, as well as the unique situational context of the choice, also determine the formation of antecedent psychological variables of the choice. In other words, the teleological explanation is part of a more general explanation integrating other causes.

Besides this, the theory of rational choice that defines an instrumental rationality oriented toward a goal of maximizing a selfish...