Consumers exercise choice on a daily basis, and the theory of preferences is a fundamental part in the mechanism of choice. Consumers will always aim for a maximised level of utility, and to do so, choice and preferences are needed. Not all utility is gained from consumption, but it is the main method of increasing, or at least trying to increase, our utility. Economists, psychologists, and sociologists, will all have different views on choice and the theory of preferences within behavioural economics. The two go hand-in-hand. Choice looks at what goods consumers decide to consume, and preferences is the ordering of these goods based on relative utility. The theory of preferences looks at what goods or resources are favoured by individuals, and why – e.g. ‘what would be a better choice for me, in order for me to be most satisfied?' Preferences seems to avoid certain considerations, such as price and availability, but with a given range of choice, it is thought that consumers will be able to identify a specific order of preference within. Choice and preferences are personal, and we would think that the more choice we have, the greater our utility is likely to be. This is because of the increased likelihood of being able to receive our most-preferred goods. However, psychologist Barry Schwartz (TED, 2006) explains that paradoxically, with so many items to choose from, it becomes harder to even make a choice in the first place. He goes on to say that even if we get past this, and finally make a choice, we find it easy to imagine an alternative choice that would've been better. This provokes us to regret the decision we made in the first place, leaving us with a negative effect of consumption, even if the original choice was good at the time.
Whenever there is a set of goods or resources available for consumption, consumers will make a choice. The best choice is said to come from having a set of preferences which are ‘rational'. This highlights the theory of rational decision making, where we look at preferences in more detail. Rational decision making is the idea that a consumer makes a choice based on some sort of logic – there are reasons behind choices. This includes preferences. But what makes a rational decision rational? We would expect consumers to have a basic idea on what they want to consume already, but this may be affected by various influences; from basic social trends, to cultures, and even a change in preferences over time. Martin Peterson (Peterson, 2017) talks about the ‘normative' side of decision making theory. He implies that decision making resists these various influences, and therefore, opens up a space for all consumers to be rational. However, there are four main assumptions about preferences within rational decision-making theory that lead the preferences of individuals into being ‘rational'. These are referred to as the four axioms of rational choice.
The first axiom of rational choice is the property of non-satiation, or monotonicity, of preferences. This is the idea that no matter how much a consumer consumes, they will still benefit from an increase in utility by consuming even more, regardless of what good or resource is consumed. It is easier to explain with the idea that out of two similar goods, X and Y, even if good X is preferred to good Y, a higher quantity of good Y would be preferred to a lower quantity of good X. This means that a consumer is totally focused on consuming more or less, not choosing between specific goods or particular amounts thereof. In simpler terms, non-satiation is the ‘more is always better' scenario. “The intuition behind this argument is that when an agent is allowed to discard unwanted objects for free, adding objects to the agent's bundle never makes the agent worse off” (Yang, 2012). It implies that there is an unlimited availability of everything, and that when anything is removed from their choices, consumers will never be worse off by consuming any quantity of any other good available to them. Utility will always increase with the quantity consumed.
The second assumption looks at preferences as being ‘complete' and ‘rankable'. This is the idea that consumers are able to make comparisons across everything that they consume, so that they know what certain combinations of goods that they would prefer. The set of goods that a consumer chooses to consume, or ‘consumption bundle', is chosen out of a wide range of different, possible, combinations. With this assumption, “a consumer can determine whether she prefers the first bundle to the second bundle, the second to the first, or is indifferent between the two (i.e., views them equally)” (Goolsbee et al., 2016). We can use it to break down consumer behaviour and compare all sorts of combinations of goods. It helps us to learn many things in lots of different areas of focus; from consumers, to trends, markets, prices, and everything in between. The ‘completeness' element suggests that consumers will always know of every possible combination of goods that is available to them, and the ‘rankable' element tells us that consumers' preferences will follow a strict order – in the sense that consumers will always be able to decide what they would prefer, and that they will always be able to decide from all possible combinations of goods or resources that they could ever potentially consume.
Third, and probably the most important, is the assumption that consumers' preferences are transitive. Transitivity is the idea of there being a fixed relation in the preference of goods, and that preferences will hold throughout different options of choice. For example; if a consumer prefers good X to good Y, and also prefers good Y to good Z, then we can say that the consumer prefers good X to good Z due to the ‘transitive' nature of preferences. “The transitivity of preference assumption is meant to rule out irrational preference cycles” (Serrano and Feldman, 2011). When given a choice between good X and good Z, even though no direct preference was stated, the original preferences should automatically hold. Most choice theories suggest transitivity (Kahneman & Tversky, 1979; Luce, 2005; von Neumann & Morgenstern, 2007), also cited in ‘Transitivity of Preferences' (Regenwetter et al., 2011).
Finally, there is the assumption of a ‘diminishing marginal rate of substitution'. The marginal rate of substitution (MRS) is described as the quantity of one particular good that a consumer is willing to give up in exchange for a certain quantity of another particular good, whilst maintaining the same utility level. It looks at combinations which all provide the same level of utility, or satisfaction. The MRS is the rate of which one good can be substituted for another, in order to preserve this level of satisfaction. It is seen as the rate of quantity changes between points (Garratt and Qin, 1994). The MRS changes when each of the two goods are consumed in different quantities. This can be modelled on an indifference map. With each quantity consumed on a separate axis, it enables us to plot the possible combinations of two, consumable, goods on a graph. When there is a change in the quantities consumed, the points we plot for our different combinations move down a simple, negative, convex-shaped, indifference curve. An ‘indifference curve' is a line demonstrating combinations with equal utility – where the consumer is indifferent between each point on the curve. When different combinations are chosen, the MRS changes between each arrangement. This helps us to break down consumers' behaviour by seeing a model of what different combinations could possibly be chosen by consumers. The three diagrams below illustrate what a basic indifference curve looks like, what happens when combinations change, and how the MRS is derived.
Mathematically, and from point-to-point on an indifference curve, the MRS depicts the slope of the curve at any specific point – the gradient of the curve. With two goods, ‘x' and 'y', the marginal rate of substitution is calculated by using a general formula: . The marginal rate of substitution assumption does not look at the preference of each combination in relation to another, it looks at which combinations that the consumer would prefer by the same amount. Nevertheless, we are now able to see where the ‘diminishing' marginal rate of substitution sets in. Seen at either end of the curve, it is where a consumer starts to give up less and less of one good for a specific increase in the other. The diagram below clarifies the law of a ‘diminishing' marginal rate of substitution.
Introducing a ‘budget constraint' line helps us to utilise applications of this model when analysing consumer behaviour. The ‘budget line' is a straight line of negative gradient that shows us the possible combinations of goods that the consumer can afford. For example, let's say there is a consumer who wants to buy a combination of ice-creams, at £2, and chocolate bars, at £1. An indifference curve would be drawn connecting the points from all of their utility-maximising combinations. If the consumer only has £10 to spend on these goods, we can draw our budget line going through all of the possible combinations that would cost exactly £10 – five ice-creams, four ice-creams two chocolate bars, three ice-creams four chocolate bars, and so on. All combinations of the goods that are on or below the budget line are affordable. Some will be desired, some not, but we can use this to see that consumers are then most likely to choose from the combinations that are below this budget line, and also on the indifference curve. A similar situation can be seen on the indifference maps drawn below. The range of choice is seen to get greater when the budget line moves out from the origin, therefore, chances of utility maximisation are higher when consumers have more money, or when prices fall.
If we start to use these axioms of rational decision making on indifference maps, we can see how they aid the make-up of our model. The ‘completeness and rank-ability' assumption is seen on the second of the three diagrams above. It lets us see that of combination one (), and combination two (), the utility from is greater than that of so, subsequently, combination two is going to be chosen by the consumer. It just so happens that is also on the consumers' indifference curve. Non-satiation enables us to see that the further out from the origin we are, the higher our utility is likely to be. This is demonstrated on the diagrams below.
The property of transitivity is only really seen with more than two goods, but by only using two goods, we can already see how the non-satiation, completeness and rank-ability, and MRS assumptions enable us to analyse consumer preferences – modelling and optimising. This technique allows us to obtain information on consumer behaviour by using indifference maps. The maps can be used to look at the combinations of specific goods and services that are chosen by consumers. Action can then be taken on these selections. Producers can see what types of goods, and what combinations of these goods, that are more likely to suit consumers – e.g. normal, inferior, complementary, substitute, giffen, or normal goods. Different products may suit different types of consumers, and various pricing or marketing strategies may be applied from this.
However, there are many external arguments against this model; various individuals' stand-points will question rational behaviour differently, and in different ways, but each of the aforementioned assumptions can highlight implications for the model. In practice, there will be situations in which consumers will feel more satisfied, or dissatisfied, after consuming more of the same good or resource – an implication for the non-satiation assumption. Completeness and rank-ability is quite vague in the sense that it only looks at consumption bundles as a whole. It fails to look at the make-up of these bundles, where the analysis of preferences across what is inside these bundles could be much more beneficial to current, or the construction of new, economic theories. The marginal rate of substitution assumption is also quite vague. It only looks at the quantity of one good that will be given up for a certain quantity of another. If we were to acquire a better understanding of the model, there may be new economic theories developed that focus on the utility side of the alternative combinations of goods, not solely on the rate of substitution for the same utility.
There is one major problem, which engulfs most economic theories, that also criticises the model. This is the argument that when we model circumstances such as this; reality, or our perception of reality, needs to be changed, or simplified, in order for our model to work. Donald Hoffman, Professor of Cognitive Science at the University of California, explains that over time “our perceptions have been shaped not to show us reality as it is” (TED, 2015). It obvious to see that we will not be able to take everything into account when developing the model. Some variables, or features of reality, will need to be left out in order for it to work. It is then a question of what to include and what to exclude. If certain mechanisms of the model are included or excluded for the wrong reasons, then there is a chance that the model will either be too simple, leading us to the wrong conclusions, or too complicated, giving us a model that fails to work. Work by Amos Tversky and Richard Thaler on ‘preference reversals' highlights problems with the majority of these four main assumptions. They discuss the axioms in a similar context, and that when applied, behavior is likely to vary across situations that economists consider the same. They explain that “alternative auction mechanisms which are equivalent in theory”, or similar situations, may produce different outcomes if the situations themselves have an influence on consumer behavior (Tversky and Thaler, 1990). This suggests that even though all four axioms of rational decision making are theoretically realistic, they are unhinged when applied in a real-life context. Solid empirical evidence is still required for a more complete understanding of rational decision-making theory.
By taking a further look at the assumption of transitivity, we can break down the theory and see how it opens the model to criticism. ‘The irrational, the unreasonable, and the wrong' (Margalit and Bar-Hillel, 1981) suggests that there is further concentration on transivity as it is arguably the most fundamental axiom of rational choice. Transitivity criticises the model by highlighting a set of underlying problems, and this identifies certain situations as being ‘irrational'. What good to us is rational decision making, that is now seen as irrational? It seems logically impossible; but when given a set of similar goods (X, Y and Z), even though they are different, in some circumstances it may be hard to distinguish what would bring a greater utility out of the three – i.e. irrationality. This removes the transitive property from our preferences.
Let's use an example. Transitivity would imply that out of the goods mentioned above; if a consumer prefers X to Y, and Y to Z, then X is preferred to Z. When faced with a choice of all three, then surely, the consumer will follow their order of preference. When the options are just X and Z, then X should rationally be chosen. But what if only Y and Z are available? Even though X is not available anymore, the order of preference should hold, and Y should be chosen. However, without the option of consuming a given quantity of good X, the consumer may now decide to change their preferences. It may be that Z brings a greater utility than Y in this specific situation, so Z is chosen. This removes the ‘rational' aspect of the decision, and logically, does not make sense – shouldn't X still be the most sought after of the three goods? And shouldn't Y always be chosen over Z? Our options of choice may not be the same all of the time, and for any situation similar to this, a consumer's decision then becomes more complicated. Preferences change, but while this explanation seems irrational, the decision may not have been quite so ‘irrational' at the time.
Various experiments over the years have failed to contribute enough to the idea of intransitivity. This is because many experiments are set up with the goal of getting a desired conclusion. They simply test enough people before results match their hypothesis. But, recent experimental evidence from Derek L. Hansen and Dr. Mark Showalter (Hansen and Showalter, 2014) showed that a significant proportion of their surveyed participants were “not consistent in their decision making”. Two thirds of the participants were intransitive at least once, and, as situations became more complex, the level of intransitivity increased. However, they concluded that the results do not directly relate to the real world. “It is possible that people would be more transitive in a real life setting where the stakes are higher”. Consumers are not as intransitive as their survey originally implied. There was strong evidence that most of the participants were irrational, but “the level of intransitivity was not great enough to warrant the abandonment of current economic theory”.
This then leads us to believe that there are not necessarily any fundamental preference relations which are intransitive. We can see that there is a loss in accuracy of the theory by assuming transitivity for all consumers, but rational decision-making theory is simple, and this makes up for that loss. The assumption should be more focused on the idea that consumers' preferences are not consistently more non-transitive rather than transitive, or more transitive rather than non-transitive. The inaccuracy of economists, when developing such theories, may reduce the validity of their application in the modelling of certain situations. The rigidity of rational decision-making theory may need to be addressed by economists.
...(download the rest of the essay above)