In neoclassical economics, the material world is reduced only to prices and quantities. Empty boxes like "good X" and "good Y" are exchanged on the market at certain prices in certain quantities, being the equilibrium quantities and prices at which demand is equal to supply.
If you take a real-world promotional leaflet of a retailer chain, you will be given a significantly larger number of elements on which to make your choice (to visit the shop or not, possibly in search for one or more goods you want to buy before the promotional period ends).
In this paper, we shall explore mismatches between ECON101 textbook theory of economics and empirical evidence from the real world that you can collect by yourself, so to replicate (confirming, widening or rejecting) our analyses, as a scientific approach would authorise and ask for.
In particular, we shall claim that "quality" is a missing element in the description of goods, that goods are inserted in broader categories (in a sort of tree structure), that quantities can have several measurement units at the same time while being prevalently discrete (and not continuous), that prices - asymmetrically distributed with odd endings - can be expressed not only in absolute but also in percentage terms, that time is crucial to understand the setting in which actual purchasing acts occur. Not only the triangle price-quantity-quality is considered by the consumer but one has to recognize that there is no one-to-one correspondence between any couple of this trilateral relation.
We shall raise issues that are all too obvious in marketing science but remain at the margin in microeconomics. In so doing, we shall offer to students and researchers a further justification for paying attention to a number of alternative economic models and approaches to consumer that we and our colleagues have being developing for years.
Special buys can be advertised in written materials delivered to all houses in a certain geographical range from the point of sale. These home-drop advertising materials report a careful selection of products in promotion which should convince the potential customer to go to the shop until the offer is valid.
Click on this image to download the full leaflet
We use these written materials as a static, replicable, limited source for information flows from retailer to consumers, containing the minimum amount of information needed to attract customers. By contrast, within the shop, the consumer will actively move and act in the venues, zooming in and out from broad categories to specific products, will visually evaluate alternatives under her/his eyes, will look at what others are buying and saying. An analysis of the shop environment is much more complex. The leaflet is an easier material for analysis, while retaining the key elements that attract to POS.
You can find many more promotion leaflets at the end of the paper, while we remain sure that a more extensive comparison across leaflets will enrich the analysis we are going to present, so we invite you to collect leaflets in your area and reflect upon them. Ours is more a method than a conclusion.
But let's start. In the image above, representing the front page of a real leaflet, you shall see the brand of an Italian retail hard-discount chain (MD), the claim of the week ("Offerte strepitose" = "Amazing offers"), three products with their features.
On the left you see a pack of milk and 8 pieces of information: 1. the image of the product package, 2. the name of the product, 3. the qualifier(s) of the product, 4. the normal price, 5. the discounted price, 6. the percentage of price reduction, 7. the quantity as volume, 8. a text underlining the exceptionality of the offer (in other images you'll find also 9. the brand and further elements):
The image of the package in turn represent a whole set of images, including the brand ("Malga paradiso"), flowers, a container, and texts. In other words, in order to identify and qualify the product X of neoclassical approach, in real life it's necessary to highlight several elements, both textual and visual, some of which replicated and given twice, of which we distiguish:
1.1. "Product name" (here: Latte = Milk);
1.2. "Product qualifiers" (here: "parzialmente scremato" = partially screamed = low fat), which are a sort of sub-categories of the product, which distinguish it from other variants;
1.2. "Quality hints" (here: U.H.T., which is a process to obtain safety and longer shelf life), aimed at shortly push up the consumer's assessment of quality (as they were bullet points);
2. visual elements, which can refer e.g. to the "exhibited product" (when transparent glass or plastics reveal the exact colour and texture of the good) or to the "exalted product" (when a photo has been manipulated so as to show a particularly tempting product), to its "origin" (when the environment in which the product was produced is suggested), to its "use" and its "ultimate effects" (e.g. on happiness of the people).
In neoclassical theory, consumer is given preferences and a utility function. In real world, his/her assessment is endogenously changing, by on going support of hints and suggestions.
Price information is split in three elements: the normal price, the discounted price and the percentage of discount. A fourth element, the absolute difference between normal and discounted price, is not given, although the viewer can calculate it (which can be easier or more difficult, depending on the actual numbers).
The customer is given several pieces of information, from which it is expected that he/she will realize that has the opportunity of buying a good at a particularly interesting conditions (while probably retaining that also the normal price is good, in comparison what other retailers charge). At the same time, it has been created a tension between the different price elements and induce at least some activity from the consumer (look, read, calculate), hopefully concurring to the further step of a mental inner order to go to the shop before the deadline for the offer is elapsed.
It is not just the price P that let the consumer take decisions. It's the comparison with other reference prices (present in the leaflet and in the memory) that pushes action. As with prospect theory, action is the result of a comparison with a reference point and the shop itself provides "food for thought". Giving stimuli and suggesting a frame for interpreting them: both are relevant for marketing activities and sale promotion.
The unit of measure are the currency (the Euro sign) for the absolute prices and the percentage sign for the discount. All too often in models you keep out both, but the real consumer need such support to allocate the pure number into a useful information.
In particular, it's interesting to note the extensive use of percentages (whose actual formula of computation is not always immediately performed by real consumer), which are used here as well as for expressing macroeconomic dynamics. Percentages are used because of two main reasons: they make comparisons easy and they establish a loose relation with logarithmic sensibility to stimulus (some psychological research has shown that the minimal threshold for perception reaction rises with the increase of the total value at hand).
To provoke consumer reaction of positive suprise and of taking the inner decision of visiting the shop, it's useful to show high percentages of discount in a large enough number of items. An hyper-rational consumer would sum up the absolute value saved and compare with the cost to travel to the shop (for time spent and gasoline used). A bounded rational consumer would simply have routines about where to shop and might take note to go to that shop before the promotion ends (with different consumers having different rules of behaviour: some might totally ignore leaflets; others might study them carefully and stick to them when shopping; in the middle most would give a look, include the shop in their shopping routes and hop in shopping both discounted and non-discounted items).
As for the price numbers, they are integer Euro plus two decimals (a head with a tail). The theoretical minimum price is thus 0.01 Euro. All price numbers will be multiples of this minimum non-divisible unit, which is also the smallest value of a currency coin. It's impossible to have prices like 1.2342342342342 Euros in a grocery shop.
Outside grocery, it has been noticed that many prices are multi-dimensional, consisting of numerous price dimensions, as they relate to e.g. monthly payments for a certain number of months. As the customer has already difficulties in coping with simple arithmetics with one-dimentional prices, this paper shows the difficulties which multiplies when the price is multi-dimensional, arriving at the conclusion that "results from a laboratory study indicate that under conditions typical of the market place, consumers do not evaluate multi-dimensional prices rationally. Instead they utilize a simplified model, resulting in inaccuracies in their price perceptions".
Quantity can be expressed not only in weight (e.g. kilo) or volume (e.g. litre) but as the number of items inside the same pack.
In this example, in a pack you find 10 items ("Croissant"), which cumulatively weight 360 grams. Please note that the price is for the pack, not for a full kilo or for each croissant, as you cannot purchase the individual item, you have to buy the pack. This information is useful for many reasons.
Each item will likely represent a "consumption dose" during the actual consumption over time: the consumer will eat one croissant. In ten days of actual consumption she/he will have exhausted the pack, which meanwhile would have entered into her/his cumulative bundle. This gives a hint for consumer about when to come back to the shop and repurchase the product to avoid inventory breaks.
Moreover, this information is useful to judge the relative convenience with other competing substitute goods. Conversely, by bundling several items in one package, the supplier can modify the terms of comparison, increase the sold quantity per purchase occasion, and generate prices per item that can have a large, even infinite, number of decimals. However, these "prices per item" are non-transactionary, the consumer cannot purchase the items separatedly, they are just notional values, results of a mental computation.
Meanwhile, the retailer can choose to supply several package sizes of exactly the same product, as we widely analysed in this paper. But it cannot provide a continuum of sizes.
Coming back to the leaflet: the transactionary price is referred to the number of packs you buy, not the weight of them or the number of items included in each pack. The quantity purchased is the number of packs (1, 2, 3,...). This cannot be a number with decimals (1.234 packs).
All this is at odds with the smooths movements of the equilibrium point in the classical demand and supply diagram from courses in microeconomics:
Nothing here prevents the equilibrium to be reached at a quantity of 1.234; it can even occur that the quantity will be irrational (such as the square root of 2). Quantities can assume any value, they are totally flexible. Simmetrically, prices here can have any number of decimals. In equilibrium, the actual price and quantity purchased can have any number of decimals. Microeconomics prices are said to be flexible as quantities, with no restrictions about the ending number (e.g. an ending in 3 like with 1.233 is perfectly possible and equally likely as any other ending).
On the contrary, in real life, packaged goods are necessarily sold in "natural numbers" (discrete quantities). And prices never end in 3.
What about meat and other fresh goods that are sold on weight? The price is here given by weight.
In this example, the meat is sold at 7.99 euros per kilo. If in the shop you'll find it packed, they have already weight it and each pack will have a slightly different price.
If you can ask (or take directly by yourself - as it happens with fruits and vegetables in open spaces) the butcher (or you, respectively) will use a balance. The balance has a certain precision (e.g. 1 gram) under which the quantity cannot be measured. The price will be calculated by multiplication. Although the quantities are now many more than before, they continue to be discrete numbers (natural numbers of the precision-level of the balance). Conversely, given the price of the good, the minimal precision of the balance might lead to an actual price lower than the minimum currency threshold (1 cent in Euro countries). A price per kilo of less than 10 euro makes the cost per gram lower than 1 cent, to the effect that it is ignored in the final bill the consumer has to pay.
In sum, also goods purchased by weight have quantities and prices that are multiples of non-further-divisibile units (e.g. 1 g and 1 cent). The smooth, continuous, twice differentiable curves of demand and supply in neoclassical textbooks ignore this real-world texture on which real consumers make their choices.
Over time, the total consumption of a grocery good will be equal to the "consumption dose" times the number of consumption acts (consumption seized occasions). We define the consumption dose as the typical quantity used in every occasion, with containers (e.g. a cup of milk) and packaging (e.g. a croissant unit) offering a discretization actually used by consumers. For instance yogurt consumption usually occurs with single doses in a container, even when the product package links two containers:
The consumer usually finishes the full dose. When consumer can more finely choose the quantites, a consumption dose is the result of consumer practice having no (increasing or decreasing) trend (as with doses of shampoo used to wash hair, which are widely different each time but are without a trend: the sum of a sufficiently large number of doses (e.g. 10) remains constant over time).
The quantity in the cumulative bundle is reduced each act by one dose (on average), sending useful alerts for consumer routines to prevent inventory break and to assess urgency of repurchasing: the signal to go to buy might be sent when the quantity in the cumulative bundle approaches the level of inventory break before the second next visit to POS is forecast (so it's necessary to buy in the next visit to POS)
In choosing the quantity, one could look at the forecasted time to come back to POS and posit that the consumer will purchase a quantity that safely allows for the consumption acts x consumption doses that will be made over that period.
One could take into account how much is still in the cumulative bundle at home (so that the sum of it and the purchased quantity safely) as well as the number of components of the household: many people do not live alone as single but purchase for their family and other people living in the same house, which de-cumulate the inventory together.
In a model, POS could be visited according to a schedule plus variation of the schedule introduced for emergency reasons (e.g. forthcoming exhaustion of inventory in the cumulative bundle of special "necessity" items).
Duplication in purchases could happen in order to remove
the possibility of emergencies or to take advantage of a special promotion
or to take advantage of the POS competitiveness in price and quality -
to avoid coming back.
In other terms, purchased quantity is not a simple function of price: it depends on the categorization of the good by the consumer, on the routines establishing the number of consumption occasions and which occasions are actually seized, on the consumption dose, and on the time of repurchase - all this cumulated for all the people living together at the household.
For instance milk purchases depends on milk being categorized as "standard necessity" with routines dictating that a cup of milk is eaten at breakfast each day (the cup being the "consumption dose"), thus over time the quantity of milk consumed (in terms of bottle of milk) will be equal to the number of days times the dose divided by the number of doses in each milk bottle. The quantity purchased will be allocated across different shops, possibly because of their relative price convenience (but also because of purchase trip decision), with large price discounts motivating more-than-usual quantity purchased. Price is important in categorization, in the habit formation, in the allocation across shops and specific shopping occasions but there is no one-to-one price-quantity relation. There is no curve of demand except as a possible, but not necessary, "emerging property" over time of consumer behaviour, each act being inserted in routines.
In neoclassical theory, demanded quantity is a function of price only, often a linear one or one deriving from indifference curves. In the real world, price is just one component of a much more structured set of routines and information.
In non-grocery items, like consumer electronics or other durables, quantity is often a size, with one or more dimensions.
In this example of image, the product (Fornello a gas) is qualified with "2 fires" and with "Color Black". The dimensions are in centimetres (width and length).
This information is useful for the consumer to imagine how the good fits the cumulative bundle of containers at home and represent a feature of the good having a potential impact on performance (e.g. easeness of manipulation).
Needless to say, the consumer cannot change the size of
the product, can only choose not to buy it and buy another one instead
(which is different from the first because of dimensions, but sometimes
also because of other features, leading to the need of balancing the changes).
The quantity purchased by the consumer is usually one (or zero if not buying!). You usually have one copy of a book, not many (unless you are purchasing them for others). In certain cases, successive acts over the years lead to have more item than one in the cumulative bundle (possibly in relationship to the number of persons in the households). In other instances, a house might require a certain number of items, so that a first collection of that number is purchased when the house is used for the first time, then - depending on breaks and replacement policies - over time new items are purchased.
However the purchase in the same shopping occasion of more than one (perfectly identical) item of durable goods is unlikely. The attention of the consumer is on "which one to buy", more than "how many to buy". Think at your visits in a bookshop.
If you browse entire full leaflet examples, freely downloadable here, you'll notice that many items are collected under broad categories (e.g. Fresh goods, Canned goods, Consumer electronics, etc.). In some offer seasons, innovative broad categories are temporarily introduced (e.g. Christmas presents, Easter eggs, etc.). The whole device of "category" is to provide an "introduction" to produts, to help localize them, offering a logical succession which does not disorient the customer and allow for some quick visual browsing, as well as to offer a "family" of relatively similar goods for comparison purposes.
Inside the shop, categories and the product tree are particularly important to cut the time to localize the good you were looking for. Too long a search would discourage the consumer - if it happens too often, the consumer would remember the awful shopping experience and would avoid to come back all together.
Needless to say, a key piece of information contained in the leaflet is the time of validity of the offer in promotion. In a real-world sample of ours, we found that most are valid for two weeks. However, marketing campaigns try and propose many other terms of validity.
Furthermore, the address of the shop, its available parking facilities, payment methods, opening hours and other services are included in the leaflet.
Let's now explore the informational processes of consumers and retailers in terms of inputs and outputs. In standard textbooks, supply and demand are expressed as equations in which price and quantity are reversibly and symmetrically linked, to the effect that consumers and producers both respond to price and quantity "signals".
In our real-world supermarket, the consumer looks at the price (and the other elements we mentioned earlier) and decide whether to buy or not, how many units to buy of which one of the many goods in offer. The consumer "processes" pieces of information on prices and provides an output as a quantity (zero, one, two,...). He/she does not negotiate the price, nor proposes a price at which she/he promise to buy a certain quantity. The informational flow from consumer to producer (retailer) is a quantity response to a price (and the other elements) signal.
Conversely, the retailer, whose main task is to fix the price, will take into consideration a number of elements (its overall positioning in the competitive retailing landscape, its costs, the advantages in attracting new customer, etc.), will estimate how much it can sell at certain level of prices (an expected demand in function of the price) then the outcome of its information process is a price. From (expected) quantity to prices is, in all too simplified analysis, the direction taken by retailer. Furthermore, each day it receives a quantity signal (how many units are sold, how many remain on the shelves), to the effect that, looking at "unusual" or "unwanted" quantity outcomes (e.g. too little sales of a certain item) may decide to change the price (e.g. by reducing it) and to undertake a number of other promotional efforts (e.g. to put the product on an adv leaflet).
In sum, real-world consumers and retailers are asymetrically sending and receiving price and quantity signals to each other, while also using further types of signals:
* time signals - the retailer highlight the duration and the end of the promotional offer, the consumer elaborates on whether she/he has time to go to the POS; when there will expectedly be next promotional seasonM on how much time elapsed since last purchase of the item; the current state of his/her cumulative bundle, his/her experience of the good, etc.
* quality signals - consumers can protest about failures in quality of the good, the retailers gives quality hints about the products in order to differentiate it from others;
* popularity signals - consumer look in the streets, in others' houses, in television, etc. how many and who has bought the good, so to follow (or snobistically not follow) a fashion; retailers screen other competing POS to highlight if new products are in their supply and may decide to include them (or an imitation) so that their own customers will not be disappointed while searching for them;
* information-rich tools - e.g. retailers use advertising formats with some (or even) a lot of informations (including parking availability, payment methods, etc. - photo).
Meanwhile, both are immersed in a flow of information about how the general macroeconomy is going, how the specific location is developing, etc. and in broad social trends and influences.
Finally, it is clear that different groups of consumers and different retailing formats would be characterised by a varying degree of attention to a number of signal sources and contents. For instance, Turkish bazaars might allow for a wide negotiation of prices between the customer and the seller.
This means that a general theory of consumption should allow a wide variety of Point of Sales where to shop, spatially distributed and with a range of channels of communication with the perspective customer (as well as possibly providing post-sale assistance) and interaction settings.
In normative economics, we judge what is good for society - or a collection of individuals - and suggest to policymakers what can be effective in changing behaviours. For instance you might judge that people should smoke less, eat less junk food or pollute less. As normative neoclassical economist, you would suggest either to raise prices (e.g. through a tax on the final good prices, VAT, etc.) or to establish quantity quotas (max production, max sales, etc.). Your expectation is that higher prices will lower consumption (but you can forecast by how much only if you know the "real" demand curve) and that quotas will raise prices so that only the most dependent consumer will continue to buy (or the richest of them!).
Because of the symmetry between price and quantity in signals, you probably will suggest that it's irrelevant whether the policymaker will use a price or a quantity policy.
In other cases, you may add some realistical detail about specific administrative costs or on the side of informational imperfections, so as to prefer prices or quantity instruments.
A large part of the debate in climate change economics between carbon taxes and cap-and-trade can be seen as a discussion on whether to employ price or quantity restrictions, possibly involving quite elaborated evaluations like this.
Our - very basic and sketchy - previous analysis would, however, suggest that you can (and should) suggest a much wider number of policies targeted at quality, time, product substitution, retail chain landscape, etc. to really move the system. This is, in much deeper and wider development, what we proposed in a long book on "Innovative economic policies for climate change mitigation".
Real people use a much wider number of signals and elements for choosing than just prices and quantities. Real-world retailers know it and subtly force the communication with consumers along lines that put their supply in an attractive light, so to have consumer shopping in their point of sales.
Neoclassical economics disregard these aspects as irrelevant details. Evolutionary agent-based models of consumer behaviour can instead include these empirical observations into simulations allowing for aggregate properties to emerge, including multi-stage simulation games in the distribution landscape during the different phases of the business cycles.
Now it's up to you
If you receive at home some ad leaflets or you find them at the shop, then please carry out an analysis of what is shown, which elements are present (prices, quantities, and what?), how prices and discounts are given, how are quantities expressed, and how our analysis is confirmed or not. What else can be added? What about odd pricing (i.e. the prevalence of price ending with odd numbers like 9)? What about the absence of the "3" as ending? If you find something interesting, let us know.
|More on consumer behaviour...|