Navigation – Plan du site
Time in Economics, part II

Consumption Choices and Time Use: History, Theory and Potential Empirical Evidence

Choix de consommation et emploi du temps : histoire, théorie et évidences empiriques potentielles
Sergio Nisticò
p. 219-238


La traduction anglaise de l’ouvrage de H.H. Gossen, The Laws of Human Relations and the Rules of Human Action Derived Therefrom (1983), éditée par Georgescu Roegen, a initié une nouvelle approche de la théorie de la consommation qui met l’accent sur le rôle fondamental du temps dans le comportement économique individuel.

Cet article souligne les différences analytiques et méthodologiques entre le traitement standard du temps comme « manque à gagner » à la Becker et le rôle du temps dans la nouvelle approche, dont la pertinence peut également être évaluée à la lumière des données à recueillir lors d’enquêtes sur l’emploi du temps et les dépenses de consommation.

Haut de page

Texte intégral

1In the very last lines of his seminal paper A Theory of the Allocation of Time, Gary Becker suggested that “time budgets, which have not been seriously investigated in most countries, including the United States and Great Britain, should be integrated with the money budgets” (Becker, 1965, 517). The main aim of this paper consists in following Becker’s suggestion by investigating how a sort of ‘economic time-use diary’ could be designed and administered in order to get some important, and not yet explored, clues about how people spend their time and their financial resources, the latter being allocated either on a pay-as-you-go basis (income flows) or from a reserve fund (wealth stock). In turn, and as the great majority of human beings in the world are very well aware of, a part of our time must be devoted to earn precisely those financial resources needed to buy the goods and services we need to enjoy our time. The secondary, though not less important, aim of the paper is to show that following Becker’s suggestion could be easily done within a theoretical context, substantially different from Becker’s, that does not rely on any ‘unobservable’ utility function.

  • 1 For an interesting review of an early debate on social interdependencies in consumption in the pre- (...)

2The circumstance that what we do and what we buy are strictly related to each other has been emphasized by those schools of thought, e.g. the institutionalists, the post-Classicals and the post Keynesians that, rejecting methodological individualism, emphasize the social dimension of consumption.1 In fact, even the economic decisions of the isolated man par excellence, Robinson Crusoe before his meeting with Friday, cannot be understood, theoretically, without a clue about what motivates his behavior on the island.

3As we know, a pen and a notebook survived the shipwreck and Mr Crusoe’s preferred intellectual activity was writing a diary. After exploring his new territory, he starts to take decisions about his life on the island. Assuming that Mr Crusoe does not act randomly but as a result of conscious decisions guided by some sort of cost-benefit analysis, nobody would deny that reading his notes could give some important clues about what type of regularities drive his economic choices. Those notes might reveal how Mr Crusoe distributed his time on the island among various activities such as:

  • gather ‘first order’ inputs (pieces of woods, stones, leaves etc.);

  • learn how to, and actually produce second order inputs, such as a rudimental knife or a ladder;

  • learn how to, and actually gather or produce third order inputs (row fish, coconuts, elaborated pieces of woods or ropes assembled together in the form of shelter);

  • enjoy the shelter and the meals;

  • personal care;

  • rest;

4and, if Mr Crusoe is precise enough, the diary will also contain the information about the amount of time he devoted to write the diary itself or to create some nice art craft to serve as object of contemplation. Finally, the diary would also contain Mr Crusoe’s own judgment about how pleasant or unpleasant the various activities he felt through time, possibly discovering that most of the time devoted to instrumental activities, e.g. catching fish, or producing the art crafts were not less pleasant than the time devoted to eat the meals or to find repair from the thunderstorm under the shelter. Reading the diary might reveal not only what type of decision making drives Mr Crusoe’s choices but also the differences between those activities aimed to ‘produce’ pleasant time in the short run and those aimed to increase the satisfaction productive capacity, in the sense of Marshall.

5In a hypothetical variant of Defoe’s novel, Mr Crusoe can rely only on coconuts to be gathered from the trees without the assistance of any inputs. However, the natives living around the archipelago, rather than being dangerous cannibals, fully respect Friday’s property rights on his coconuts and send out his representative merchant, let’s call him Friday, to trade inputs of any order as well as final goods for Friday’s coconuts that are used as money in the archipelago. Luckily, a pen and a notebook survived the shipwreck also in this version of the story where Mr Crusoe’s diary will report much more time devoted to coconuts gathering, which will replace the time that in the previous version was dedicated to the production of the various inputs. Apart from this (non negligible) difference, the diary, with all details about the various activities that Mr Crusoe dedicates to the production of shelter, meals and art crafts, as well as to the use (consumption) of those goods or to planning and learning how to expand the set of potentially pleasant activities will still be illuminating in understanding how our isolated man takes his economic choices. It is in particular Mr Crusoe’s own judgement about the pleasantness of the various time uses, including his source-of-income activity (coconuts gathering) that can shed light on his choice mechanisms. We could discover, for instance, that rather than expanding coconuts gathering to have more ‘money’ to buy second order inputs, Mr Crusoe prefers buying only first-order inputs and to devote his time to producing those second-order inputs in a do-it-yourself activity, not necessarily to ‘save’ coconuts but possibly to enjoy the time devoted to that challenging activity. We could also discover that Samuelson’s revealed preferences approach to proving the existence of a well-behaved demand function (Samuelson, 1938) is of little help since, after repeating the same set of activities for a few weeks, Mr Crusoe’s demand pattern changes significantly in line with a path-dependent perception of ‘pleasures and pains’ associated with the various activities; moreover, we could discover that such a change in demand takes place regardless of any change in the exchange value of his endowment (time and accumulated coconuts stock) relative to the goods supplied daily by Friday. In fact, Samuelson, whose aim was that of “dropping off the last vestiges of the utility analysis”, wouldn’t have wasted his time in reading what the diary tells about Robinson’s time use on the island. In Samuelson’s approach, the information about the number of coconuts paid to Friday in exchange of the various items traded daily is enough to infer and hence “assume in the beginning as known, i.e., empirically determinable under ideal conditions, the amounts of n economic goods which will be purchased per unit time by an individual faced with the prices of these goods and with a given total expenditure” (Samuelson, 1938, 62). In Becker’s view, instead, time use matters, since Mr Crusoe chooses about time-consuming activities (commodities in Becker’s peculiar jargon); and the amount of time needed to perform them is an essential component of their cost to be measured in terms of foregone coconuts, as it is in the neoclassical treatment of labour supply in the consumption-leisure space, where it is taken for granted that the opportunity cost of, or the best alternative to, time spent in all the various activities listed in Mr Crusoe’s diary is coconuts gathering, i.e. income earning.

  • 2 Chapters V and VII of Samuelson’s Foundations (1947) contain a brilliant summary of the different u (...)

6The stand taken in this paper is wholly different from Becker’s and Samuelson’s and implies going back to the origin of the subjective theory of value, as it was sketched by Smith’s with his notion of labour commanded (Smith, 1976 [1776], 34) and articulated by Gossen in his laws of pleasure (Gossen, 1983 [1854], 6). We are sure that Mr Crusoe’s decisions about what to trade with Friday are inextricably linked to what he plans to do on the island and that, outside the atemporal, thought-experiment definition of the alternative price-quantities pairs that constitute a demand function, the number of coconuts that he must be give in exchange might play a much less important role than mainstream demand theory suggests. In fact, with the advent of the marginalist revolution, the atemporal version of the neoclassical theory of consumption became established and the ensuing debate focused on the usefulness of the notion of utility to build the foundations of a rigorous demand and supply theory of value to be opposed to the Classical solution, shared by Gossen, that under competitive conditions, costs of production alone determine long-run relative prices.2 If we take the standpoint that explaining demand (choices) and explaining prices should not necessarily be done contextually within the same theoretical model, then the former explanation could be freed from all those axioms needed to yield the requisites of existence, uniqueness and stability of a vector of relative prices. In fact, an empirically robust theory of choice could be elaborated starting from observation of how individuals allocate their time and financial resources while time itself flows.

7Unfortunately, nowadays a few people around the world keep a detailed diary of what they do and what they buy. On the other hand, time-use surveys conducted by means of diaries have become ordinary job within National Statistical Offices and other institutions engaged in social research. The aim of those types of surveys has been that of taking a snapshot and capturing the evolution of social relations within the households. We share Becker’s mentioned argument that, once the potential of time-use surveys is extended outside their typical sociological domain, analysis and interpretation of the collected data could provide important insights into what motivates individual economic choices. On the other hand, we are sure that Becker’s suggestion to look at time-use data could also shed light on the potential explanatory power of different theories of consumption, possibly alternative to both Samuelson’s and Becker’s approaches, as an antidote to the “possible drying up of the sources of discussion with regard to the new issues that continually arise from the observation of the real world, but also of the very ‘fundamentals’ of the discipline” (Nisticò and Tosato, 2002, 2).

8The rest of this paper tries to show that reading Robinson’s, or anyone else’s, diary and being equipped with some fundamental economic notions such as opportunity cost or rate of return could give, alone, some clues about economic decision making without reference to any type of utility function, or given-preferences assumption. In a context in which individual rationality does not serve the purpose of deriving a well-behaved demand function, even the extent to which individual choices are driven by socio-institutional constraints or by maximization could be tested, an alternative maximand of individual decision makers under uncertainty being possibly detectable from the same diaries, namely the ratio between pleasant and unpleasant time, à la Bentham, in a given time horizon. In this perspective, Section 1 identifies the main existing research projects and data collection about time use. Section 2 reviews some important reference points in the history of economic thought as well as the recent theoretical approaches to time use and identifies a possible ‘measure’ of the economic efficiency of different activities. Section 3 discusses how the existing surveys could actually be extended in order to serve the purpose of testing alternative theories of consumption. Section 4 concludes.

1. Time Use Data Collection: History and the State of the Art

9Time-diary surveys are attracting the interest of researchers and policy makers from all around the world since decades. The potential of time-use surveys was pioneered by The Institute for Social Research at the University of Michigan, and in particular by his former director Thomas Juster. The book Time, Goods and Well-Being that Juster co-edited with Frank Stafford (Juster and Stafford, 1985) is still unparalleled in showing how time-use surveys can foster our understanding of individual behaviour. Two other reference points for researchers in this field are Time for Life by Robinson and Godbey (1997) and The Economics of Time Use by Hamermesh and Pfann (2004).

10According to a research conducted by Kimberly Fisher (2015) at the Centre for Time Use Research of the University of Oxford, surveys on time use are being conducted in more than 90 countries around the world and about 110 countries have conducted at least one survey in the last 50 years. The U.S. Census Bureau, sponsored by the Bureau of Labor Statistics (BLS) annually surveys how US residents spend their time. Given the mission of the BLS, the American Time Use Survey (ATUS) aims, essentially, at monitoring the changing patterns of labour. The overall goal of all those surveys is essentially to foster social accounting (Juster and Land, 1981) and, in particular, to monitor social change in its various dimensions, while their use in economics have been limited to grasp the role and impact of non-market activities on individual wellbeing. In this latter perspective, sponsored by the National Institutes on Ageing, a well‑being module has been added to ATUS. This module collects information about how the respondent felt, emotionally, while performing the selected activities with emotional states covering the six following alternatives: happy, tired, stressed, sad, pain, meaningful. As far as Europe is concerned, the National Statistical Offices of several member countries regularly conduct time-diary surveys. Since the second half of the ‘90s, EUROSTAT started a coordination process that resulted in the Guidelines on Harmonised European Time Use Surveys (HETUS). The latest version of HETUS has been published in 2013. Of particular interest for the purpose of the present paper is the part of these guidelines relating to an additional module on experienced wellbeing, similar to the one recently added to ATUS. Inspired by the 2011 French Time-use survey, the guidelines suggest to include in each line of the diary the additional question “Was this moment pleasant or unpleasant?”. In contrast with Becker’s suggestion mentioned at the outset of this paper, the guidelines do not include any hint about collecting data on income and on the expenditure necessary to perform the various activities. However, these data are essential to endow the time-use surveys with the potential of being suitable for an ‘economic accounting’ purpose.

11The next section highlights the origins of the approach here suggested and discusses how collecting data about the pleasantness or unpleasantness of the various activities, including work, and about the fraction of the weekly income devoted to buy the inputs of the various activities can be used for an economic investigation of individual behavior.

2. A Suggested ‘Timeful’ Alternative to the Axiomatic Approach: Back to Bentham and Gossen

12As hinted at above, Becker’s seminal approach to consider households as producers incurring also time costs, valued as foregone earnings, among the inputs of the household production function neglects the circumstance that time should be considered not just a cost but also an output of the individuals’ production function according to whether its flow is felt as unpleasant or pleasant during a specific activity.

13This latter circumstance was clear to Hermann Heinrich Gossen, whose Laws of Human Relations (Gossen, 1983 [1854]) are centered on the idea that enjoying pleasant time is the goal of human action and that the intensity of enjoyments is a function of both the length and the frequency of time spent on activities and not of the units of commodities consumed (in a moment of time). In fact, contrary to the neoclassical version of the subjective theory of value as it emerged with the marginalist revolution, Gossen did not set its analysis in the price-quantity space and considered time units as the fundamental choice variable, since “enjoyment must be so arranged that the total life pleasure should become a maximum … and if nevertheless, the actions of people seem so extraordinarily varied, this is solely the result of their different views as to the magnitudes of different pleasures” (ibid, 3).

  • 3 The implications of the role of time in consumption for individual wellbeing had already been thoro (...)

14The need to rescue Gossen from the minor role of being a mere forerunner of marginal utility theory has been pointed out by Georgescu-Roegen in his brilliant introduction (Georgescu-Roegen, 1983) to the long awaited English translation of Gossen’s book and by Ian Steedman in his Consumption Takes Time (Steedman, 2001). In contrast with Becker’s, Steedman’s contribution emphasizes that most of the predictions of standard consumption theory do not hold when time constraint in consumption choices is taken into account. Moreover, Steedman shares Georgescu’s reading of Gossen according to which the continuity thesis between Gossen and the marginalist revolution is unwarranted.3 We should also take into account the circumstance that, not only Gossen had a path dependent conception of preferences about time use, but also that Gossen, contrary to Jevons, Menger and Walras, was not interested in developing a theory of value alternative to the classical one based on cost of production alone, which he fully endorsed in chapter 7 of The Laws (Nisticò, 2016).

15Gossen made no reference to the authors who inspired his Laws. However, it is hard to disagree with Schumpeter (1987 [1954], 1022) that he was a Benthamite. Gossen’s declared aim to explain what drives human behavior, as summarized by the first sentence of his book, “man wants to enjoy life and makes it his goal to increase pleasure enjoyed throughout life to the highest possible level” (Gossen, 1983 [1854], 1), clearly echoes the first sentence of Bentham’s An Introduction to the Principles of Morals and Legislation:

  • 4 This paper does not deal with paternalism and the ‘aggregation problem’ associated with Bentham’s u (...)

Nature has placed mankind under the governance of two sovereign masters, pain and pleasure. It is for them alone to point out what we ought to do, as well as to determine what we shall do (Bentham, 1823, 1).4

16Bentham’s intention to develop a theory of choice based on a cost-benefit analysis, as reflected by pleasures and pains, is even more evident in his A table of Springs of Action. In fact, The Table is a sort of ante litteram manifesto of the economics—and—psychology research program aiming to understand what drives human activities per se, i.e. regardless of what will become the fundamental concern of the early marginalists, namely Jevons and Warlas, to rely on the notion of marginal utility as a foundation of a demand—and—supply theory of prices to be opposed to Ricardo’s cost—based theory of value.

17One of the most interesting features of Bentham’s Table is its being framed in a timeful context, where both past and future pleasures and pains play a fundamental role in explaining present choices:

57. Derivative pleasures are derived: viz. by memory or imagination.

58. Derived from imagination if the conception entertained of them be accompanied with belief of their future existence, they are ditto of expectation; and if pains, of apprehension: if not, ditto of imagination merely.

59. From recollection of pleasures may be derived pleasures or pains of imagination: and so from recollection of pains.

60. Pleasure is a spring of action no other than as action is regarded as a means of obtaining it — and so of pains, putting avoidance instead of obtainment.

61. Inert pleasures are: 1. Ditto of mere recollection; 2. Of mere imagination; 3. Of expectation, where not regarded as capable of being rendered by action more certain, or speedy, or greater.

62. Objection. By recollection of pain from injury is produced desire of revenge and thereby, action.

Answer. In so far as desire, so is expectation of eventual pleasure: in which, and not in the pain, the action has its immediate source (Bentham, 1815-17 [1983], 10).

18And Gossen, like Bentham, emphasized the path-dependent evolution of both the intensity and the duration of pleasures driving human actions:

Now, on the one hand, the life of a human being covers a considerable time span, and there are large number of pleasures in life that man can obtain immediately; yet those pleasures have the consequence of imposing later, disproportionate deprivations. On the other hand, the most elevated, the purest pleasures become comprehensible, become real pleasures, only after man has educated himself for their appreciation (Gossen, 1983 [1854], 3).

19It is possibly because of his (unwarranted) association with neoclassicism that in their “Back to Bentham” article in memory of Amos Tversky, Kahneman, Wakker and Sarin (1997) neglected Gossen altogether when rightly pointing out that late neoclassicism deviated, regrettably, from the Benthamite agenda to rely on pleasant and painful time—consuming experiences, both ‘recollected’ and/or ‘expected’, by memory or imagination, to explain the ‘springs’ of economic actions.

20Kahneman et al. point out Stigler (1950) when mentioning the change in the notion of utility endorsed by late neoclassicism; and Stigler did not forget Gossen, blamed precisely for his attempt to cast pleasures and pains in a theoretical context in which time plays a fundamental role:

Gossen’s discussion of the laws of satisfaction is concerned only with individual acts of consumption, such as the eating of slices of bread. Correspondingly, in his early diagrams marginal utility is a function of time (duration of the act of consumption), and only after a considerable elaboration of this approach does he take quantity of a (perishable) commodity as proportional to duration of consumption (Stigler, 1950, 314).

21After eradicating time from the Bentham-Gossen original approach, the next step in the direction of the modern axiomatic derivation of demand functions from the mere assumption of rationality (optimization), was to get rid of all psychological investigations characterizing experienced utility, as Kahneman et al. designate Bentham’s notion of hedonic experience:

experienced utility is essentially ignored in modern economic discourse. The rejection of experienced utility is justified by two standard arguments: (i) subjective hedonic experience cannot be observed or measured; (ii) choices provide all necessary information about the utility of outcomes because rational agents who wish to do so will optimize their hedonic experience (Kahneman, Wakker, and Sarin, 1997, 375) .

22Rejecting the argument that hedonic experiences cannot be measured is the essence of the authors’ proposal to go back to Bentham. Kahneman et al. do so by starting from the assertion that “the outcomes that people value are normally extended over time” and focus on both remembered and predicted utility of ‘temporally extended outcomes’, arguing that this notion of utility can be observed and measured by instant utility that “can be derived from immediate reports of current subjective experience or from physiological indices” (ibid., 376, emphasis added). As it is well known, the authors’ conclusion is that the ‘peak’ and the ‘end’, rather than the duration, of hedonic experiences play a fundamental role in determining how individuals remember, and hence predict, ‘utility’.

23Showing how time-use surveys could provide the systematic recollection of individual experiences, thus allowing to start over with the Bentham-Gossen agenda as suggested by Kahneman et al., is the goal of section 3.

24On the other hand, if ‘pleasures and pains’ have to be considered components of time-consuming activities, the solution for the ‘measurement problem’ should be extended to goods and services, whose consumption is a fundamental input of our activities.

25In fact, it is by going back to Smith’s notion of ‘labour commanded’ that an economic accounting of individual activities can be given in terms of time units. Smith was crystal clear about keeping the theory of value, intended as an explanation of relative prices, separate from a subjective theory of how individuals value what they already have, or have bought from other people. As we know, Smith adhered to a cost of production theory of relative prices, relegating the labour theory of value ‘to that early and rude state of society’, and endorsing the theory of the component parts ‘in every improved society’ (Smith, 1976 [1776], 53-57). On the other hand, in chapter V of book I of The Wealth of Nations, immediately preceding chapter VI dedicated to the rules ‘for exchanging [different objects] for one another’ (ibid.), Smith distinguished between the real and nominal price of commodities starting from the Bentham-like assumption that individuals are ‘rich or poor’ according to whether their wealth can allow them to avoid the unpleasant time necessary to “enjoy the necessaries, conveniences, and amusements of human life” (ibid., 34). It is in this context that a subjective theory of value, based on ‘labour time’ survives in The Wealth of Nations together with the cost-of-production theory of relative prices:

  • 5 Ricardo and Marx considered the theory of component parts at odds with Smith’s idea that labour com (...)

What every thing is really worth to the man who has acquired it, and who wants to dispose of it or exchange it for something else, is the toil and trouble which it can save to himself, and which it can impose to other people. What is bought with money or with goods is purchased by labour, as much as what we acquire by the toil of our own body (ibid.).5

26In line with Smith’s measure of value in terms labour commanded, Becker’s idea to compute the cost of consumption time in terms of foregone earnings can be fruitfully ‘reversed’ by computing the cost of consumption goods in terms of the unpleasant time needed to earn the income necessary to pay for them. By so doing, it is possible to identify the efficiency of each form of time use (activity) in terms of a ‘rate of return’ (Nisticò, 2014; 2015) computed as a ratio of the pleasant over the unpleasant time units spent on it, according to the following equation:


  • ej and pj denote, respectively, the unpleasant and pleasant shares of the total time Tj devoted to activity j (e.g. washing dishes and enjoying dinner at home) and eL represents the unpleasant share of working time;

  • the gijs denote the services of all n market goods used up during the activity j, mi their market price, w the individual’s income rate.

28Note that, since the ratio mi/w represents the ‘labour commanded’, i.e. the amount of working time (supposedly unpleasant only for the fraction eL) necessary to buy one unit of gij, the whole denominator of (1) is the flow of unpleasant time necessary to perform activity j.

29As we will see below, the fruitfulness of this approach lies in the circumstance that all variables of equation (1) are to be found in Robinson Crusoe’s or anyone else’s diary. On the other hand, one of the main problems to be addressed in a time-based approach to consumer behavior is the role of what comes ‘before’ and ‘after’ consumption decision; therefore, a careful periodization of the analysis is needed. Some of our activities are certainly recurrent:

The household does things more or less easily at different times in large part because it exists an environment that is different at different times. … Stores open and close; trains leave; children get home from school and friends from work; the sun comes up, bringing cheap light and the inputs for a suntan. These environmental changes, given endowments, preferences, prices, and production functions, determine the household’s utility-maximizing schedule of activities. The particular elements of the environment that are incorporated into this analysis are those entirely repetitive, rhythmic, and hence fully anticipated changes that occur in equilibrium day after day or week after week or year after year. It is, of course, a rhythmic equilibrium because it reflects adjustment to a rhythmic environment. And because that environment is exogenous, the household is linked irrevocably to the real time of the world outside it (Winston, 1982, 190).

30However, a significant part of Mr Crusoe’s as well as of our activities are precisely aimed to break the repetitive rhythmus of our daily life. Mr Crusoe could take the troublesome decision to truncate an activity started with the aim of increasing the quality of his life on the island, with a significant impact on his preference ordering in the weeks to follow. Consumers fail, not so much differently from entrepreneurs. On the other hand, projects for new activities and new buying options may take shape, to be treated, analytically, as investment taken under uncertainty. Therefore, any theorist accustomed to look for optimum solutions, might be disappointed in observing that data collected through diaries do not show that individuals’ allocation of time fulfils the optimum, maximum pleasant time condition characterized by the equality of the marginal rates of return (∂rj/∂Tj) on all activities performed within a given time period (Nisticò, 2014). This is why, Hicks’ (1946) temporary-equilibrium approach, appropriately extended to household decision-makers, can serve the twofold purpose of providing a suitable theoretical method and to cope with the ordinary viewpoint that our daily life flows through a time frame conventionally divided into ‘units’ (say minutes) and ‘periods’ (say weeks). Within Hicks’s temporary-period method, individuals allocate time units among the various activities, possibly trying to maximize the share of pleasant time within a week; moreover, upon comparison of expected and actual satisfaction outcomes, they formulate a Gossen-type plan that extends to the weeks to follow. Hence, Hick’s sequence of temporary equilibria extended to consumption choices should be characterized both by the recognition that each new weekly plan is rooted on the information provided by the past, but also on ‘pleasures and pains’ projected into the future, in a dynamic sequence that hardly converges towards any intertemporal equilibrium path:

Even when we have mastered the “working” of the temporary equilibrium system, we are even yet not in a position to … examine the ulterior consequences of changes in the data. These are the ultimate things we want to know about, though we may have to face the disappointing conclusion that there is not much which can be said about them in general. Still, nothing can be done about these further problems until after we have investigated the working of the economy during a particular week (Hicks, 1946, 246).

31There is also room for Marshall within Hicks’ method. Some of our decisions can be labelled as ‘short run’ since they are taken assuming as given the capacity to produce enjoyment, as well as a series of institutional constraints, such as the impossibility to change our job, or place of residence, within a week’s time. However, we normally act also with reference to a long run, precisely with the aim of adapting our ‘enjoyment-productive capacity’ and/or to remove the obstacles preventing the implementation of important changes in our future weekly plans.

  • 6 It is interesting to note that just a few years before, Becker tried to show that the “main implica (...)

32To what extent is there room for maximizing within a timeful decision making process where what to buy depends on what to do? To what extent can we choose all the details of our weekly plan rather than being ‘forced’ to accept most of it, as a combination between the heritage of our past choices and social constraints? There are no easy answers to the above questions and some hints from how individuals choose in their everyday life are definitely welcome, since any robust theory should be grounded on observed facts. Becker suggested that the ‘extended’ time-use surveys might corroborate his 1965 conception of time use as a cost (foregone earnings) within a ‘full income’ constrained utility-maximisation.6 Actually, in order to test the possible optimality of the intra-week allocation of time units or any other decision rule, we need only assume that individuals are able to distinguish between pleasant and unpleasant time units. Then, by integrating time-use surveys with a module collecting data about the individual’s source of earnings and expenditure on the market goods used up to perform each activity, we can compute the rates of return of the various activities in order to check the robustness of the strong-rationality assumption of the traditional approach to consumption. Moreover, if surveys could be conducted ‘longitudinally’, we could test whether preferences about time use can actually be considered ‘given’ week after week and whether any preference ordering is really separable from constraints.

3. The Relevant Data to Test Alternative Time-Use Theories

  • 7 The integration of the diary with the Pencil and Paper interview (PAPI) has proven to be very succe (...)

33As mentioned above, in order to test alternative economic explanations of how individuals choose what to do and what to buy, the ‘economic time use survey’ should ask the respondents to specify, first of all, whether each time span (e.g. a quarter of a hour) spent on any activity (including work) has been pleasant or unpleasant, while the collected information about the individual’s income and the monetary cost of the ‘inputs’, i.e. of the goods and services used up during the same activity, will allow the identification of the ‘rate of return’, as defined in equation (1), for each specific form of time use. A series of important details of the proposed ‘economic time-use survey’ need to be defined though, for what concerns in particular its design. In this perspective, the experience on time use surveys conducted at the national and international levels mentioned in section 1 can, and inevitably should, represent the reference point. More specifically, the suggested ‘economic time use survey’ can draw from the now established 24-hours diary collection, which has proven to be more effective than interviewing the respondent with stylized questions about the time spent, retrospectively, in some predefined and pre-coded activities. As it will be clear below, the 24-hours diary will have to be collected more than once from each respondent, while the usual structure of the diary will have to be adapted and complemented to comply with the aim of integrating the strictly economic information in equation (1). This adaption can be done, upon completion of the diary, by an interviewer meeting and asking the respondent further questions, listed in an ad hoc questionnaire, focusing on the activities registered in the diary.7

34An important feature of equation (1) is the dependence of the rate of return of all activities on the share of working (or any other income-earning activity) time felt as unpleasant. Therefore, interpretation of the reports according to gender, age, professional status (e.g. employee, self-employed, student, retiree), income class and educational level might reveal that the choice mechanisms adopted by the respondents vary according to professional and income status. In particular, students, rent earners and retirees might exhibit models of choices different from those individuals whose earnings derive from paid, possibly unpleasant, work, in line with the behavioural findings on ‘mental accounts’ (Thaler, 1985; Shefrin and Thaler, 1988). Moreover, since observation and interpretation of the collected data should give important insights about (i) how income and time budgeting jointly drive individual consumption choices, (ii) what motivates individual economic choices (e.g. maximisation or satisficing) and (iii) what, if any, is the ‘maximand’ variable (e.g. pleasant time or an underlying utility associated to any bundle of activities) and, (iv) about the time period relevant for the expected outcomes of the choices, the following crucial issues will have to be addressed in conducting the suggested survey:

  1. A ‘week of time’ as the minimum observation time period, e.g. by randomly selecting a weekday and a weekend day, for the survey to be meaningful, and the connected trade-off between longer observation periods, quality of the answers and number of individuals willing to write a diary;

  2. The need to distinguish between final (e.g. having dinner at home) and instrumental (e.g. cooking and washing dishes) activities and the possibility to identify ‘roundabout’ activities comprising both instrumental and final time uses (e.g. watching a movie in a theatre including the time needed to reach the theatre and walking or driving back home);

  3. The possible overlapping of activities and their classification as ‘time intensive’ (e.g. reading a novel or watching TV) or ‘goods intensive’ (e.g. having a family dinner in an expensive restaurant);

  4. The impact of consumption goods and services on the pleasantness of the activities;

  5. How to split the cost of the services of durable goods (e.g. cars or housing) among the various activities reported in the survey (e.g. commuting or travelling during the weekend, watching TV or having breakfast at home);

  6. How to capture the time flows devoted to the maintenance of the durable goods;

  7. Identify those forms of time use that can be treated as ‘investment activities’, i.e. whose expected return might show up after the time period (the week) in which the administration of the questionnaire takes place;

  8. The relative weight of different sources of income (e.g. salaries, pensions, rents, scholarships) possibly associated with the age of the respondent;

  9. Decide whether collecting data about the pleasantness or unpleasantness of the activities as ‘binary’ data or on a scale;

  10. Design the after-diary interview in order to infer the type of individual rationality from the chosen bundle of goods and activities, with particular reference to the extent to which the respondents might ‘adjust’ their time to increase the overall pleasantness of time, considering that some specific forms of time use might depend on habit or social/institutional constraints rather than on deliberate choice.

35For the planned survey to be able to address these issues, the design of the ‘economic time-use survey’ should focus on the following elements:

    • 8 An interesting method combining diaries and ‘habits questionnaires’ to avoid the problems connected (...)

    The overall observation period should somehow cover an entire calendar year in order to catch the specificity of some ‘seasonal’ activities. In fact, though a one-day random observation would perfectly fit the need of investigating decision making over those activities individuals undertake on any given day, such as eating, personal care or sleeping, this is not the case when the focus is on individual overall behaviour and not on reporting about some specific forms of time use. However, it is unfeasible to ask the respondent to take a 24-hours diary for 365 consecutive days. Therefore, the choice must be taken about whether and for how many times re-administer the 24-hours diary to the same respondent within a year or rely on the after-diary questionnaire, filled under the guidance of the interviewer, to gather information about the activities that vary across the days of a week, across the weeks of a month and across the months of a year.8

  • The survey could, in principle, be conducted with ‘light’ paper diaries based on a number of predefined activities for which the respondents should report their daily occurrence. Collecting a light survey is faster and easier for the respondent, which increases the response rate. On the other hand, a full-scale diary asks individuals to specify, in their own words, the chosen activities for each time interval. Full-time diaries allow collecting more accurate information about sequence and possible overlapping of activities, though placing a possibly excessive burden on the respondent. Both forms of diary should, in any case, be integrated with the question on experienced well-being included in the French Time Use Survey since 2010: “was this moment pleasant or unpleasant”, with a scale ranging from -3 (very unpleasant) to +3 (very pleasant).

  • The more information is collected through the diary, the less will have to be collected through the auxiliary questionnaire administered by the interviewer. Such a trade-off will have to be carefully considered. In fact, assessing the distinction between final, intermediate or investment activities, and the possible overlapping of some of them, could be done either way. The diary could be integrated with a column asking the respondent, for each time interval, to indicate the right attribute of the activity; alternatively, this specification could be postponed to the moment of the interview. The first option is preferable if one considers that the auxiliary questionnaire will inevitably bear the burden of collecting the typically economic information about costs of the goods and services used up during the activities, the sources and the amount of income of the respondent, the indirect use (possibly neglected while filling the diary), of durable goods, the activities undertaken in other days than those in which the diary was taken.

4. Policy Implications and Conclusion

36The relatively recent economic approach to time-consuming consumption behavior ‑ based on the idea that since consumption takes time, individuals should be considered producers of enjoyment activities whose output requires both goods and time ‑ lacks any reliable collection of data for its potential to be tested. We have tried to show how the gap in such a collection of data can be filled by appropriately extending the existing time use surveys, whose relevance both for academic researchers and policy makers, has been limited to the following areas:

  • The incidence, and the possible value, of non-market activities relative to the official figures in national accounts;

  • The occurrence of unpaid work within households and the related issue of gender differences;

  • The relative incidence of home and on-the-job paid work and the development of new work schemes;

  • The possibility to collect information about quality of life and well-being of individuals and households;

  • The overall incidence, and the different forms, of leisure activities, such as those that have a direct impact on well-being and those that have a direct impact on health;

  • The impact of transportation facilities on the quality of life.

37So far, time-use surveys conducted by social researchers have been able to catch the evolution of our habits and ways of living about the topics included in the previous list, thus supporting policy making in targeting the possible interventions to the benefit of specific population groups. Moreover, the existing surveys can be useful for those interested in attributing a monetary value to a series of non-market activities relevant for our wellbeing. However, none of the preceding areas are relevant for understanding ‘how people choose’ about their money and their time concurrently. In fact, the potential of the new suggested surveys lies precisely in the possibility to ‘observe’ economic choices over activities and, at the same time, measure their outcome in terms of the amount of pleasant time that each form of time use is able to produce. Moreover, a reliable collection of data about how people spend their time together with their money could provide important insights into how to extend the scope for policy making. The areas of these new policies could include the definition of innovative welfare policies capable of increasing the opportunities for a more even distribution of the main activities (education work, leisure) across the life-cycle, or the introduction of incentives for the adoption of more efficient uses of durable goods. In particular, besides the obvious positive impact on future developments of consumption theory, the findings of the suggested economic time-use surveys could provide useful insights on:

  • The different impact on wellbeing of having a choice to substitute among different activities belonging to the same ‘set’ (having dinner at home or at restaurant) and having a choice to reallocate lifetime among the activities belonging to different sets (e.g. more education and less leisure or work);

  • The possible policies capable to increase the individuals’ degree of freedom as to the second type of choices mentioned above, given that the reallocation of our lifetime across different sets of activities might have a significant impact on wellbeing despite being quite unfeasible within each phase of the life-cycle;

  • The consequences of the ongoing increase in life expectancy on choices as to the allocation of individual lifetime in general and as to retirement plans in particular, with the possible need to enforce policies whose aim is to support a more even distribution of non-working years throughout the life-cycle;

  • The ‘idleness’ of most of the durable goods through time and the possible implication of increasing supply of ‘just in time’ consumption technologies (e.g. hiring car services as an alternative to owning a car);

  • The uncertainty surrounding intertemporal choices of individuals having to ‘invest’ their scarce time (in order to acquire the skills necessary to raise the average returns of their consumption activities) might require the introduction of some forms of policy support, similar to the widespread diffusion of economic incentives to those firms willing to increase their productive capacity.

38In the event that our suggested ‘economic time-use survey’ will prove to be effective in gathering information on the efficiency of individual choices about time use, it could represent the starting point of a longer term longitudinal analysis, whose impact on the understanding of economic human behaviour would be much stronger than that of cross-section analyses, whatever their spatial diffusion.

A previous version of this paper has been presented at the Creativity and Motivations Research Center of the University of Cassino and Southern Lazio and at the 2016 STOREP Conference in Catania, where I could benefit from constructive criticism and suggestions. I am particularly indebted to Marina Bianchi for the many interesting hints about the implications of the paper and some relevant literature supporting my argument. I thank the anonymous referees. The usual caveat applies.

Haut de page


Becker, Gary. 1962. Irrational Behavior and Economic Theory. Journal of Political Economy, 70(1): 1-13.

Becker, Gary. 1965. A Theory of the Allocation of Time. Economic Journal, 75(299): 493-517.

Bentham, Jeremy. 1815-17 [1983]. A Table of the Springs of Action (The Collected Works of Jeremy Bentham, ed. by A. Goldworth). Oxford: Clarendon Press.

Bentham, Jeremy. 1823. An Introduction to the Principles of Moral and Legislation. Oxford: Clarendon Press.

Bianchi, Marina and Eleonora Sanfilippo. 2015. Social Interdependencies in Consumption: An Early Economic Debate on Social Distinction, Emulation, and Fashion. History of Political Economy, 47(4): 547-575.

Fisher, Kimberly. 2015. Metadata of Time Use Studies. Last updated 31 December 2014. Oxford: Centre for Time Use Research, University of Oxford.

Georgescu-Roegen, Nicholas. 1983. Herman Heinrich Gossen: His Life and Work in Historical Perspective. Introductory Essay to Gossen (1983 [1854]). Cambridge, MA: The MIT Press.

Gershuny, Jonathan. 2012. Too Many Zeros: A Method for Estimating Long-term Time-use from Short Diaries. Annals of Economics and Statistics (105/106): 247-270.

Gossen, Herman H. 1983 [1854]. The Laws of Human Relations and the Rules of Human Action Derived Therefrom. Cambridge, MA: The MIT Press.

Hamermesh, Daniel S. and Gerard A. Pfann. 2004. The Economics of Time Use, Contributions to Economic Analysis, vol. 271. Bingley: Emerald Publishing.

Hicks, John R. 1946. Value and Capital, 2nd ed. Oxford: Oxford University Press.

Juster, Thomas F. and Kenneth. C. Land. 1981. Social Accounting Systems. New York: Academic Press.

Juster, Thomas, and Frank Stafford. 1985. Time, Goods and Well Being. Ann Harbor: University of Michigan Press.

Kahneman, Daniel, Peter P. Wakker, and Rakesh Sarin. 1997. Back to Bentham? Explorations of Experienced Utility. The Quarterly Journal of Economics, In Memory of Amos Tversky (1937-1996), 112(2): 375-405.

Linder, Staffan B. 1970. The Harried Leisure Class. New York and London: Columbia University Press.

Nisticò, Sergio. 1991a. On Marx’s Critical Reading of Adam Smith’s Theory of Prices. In G. Caravale (ed.), Marx and Modern Economic Analysis. Aldershot: Elgar, 149-167.

Nisticò, Sergio. 1991b. Prices and Distributive Antagonism in the Work of Smith and Kalecki. Journal of the History of Economic Thought, 13(1): 54-69.

Nisticò, Sergio. 2002. Different Perspectives on Distribution within Classical Political Economy. In Sergio Nisticò and Domenido Tosato (eds), Competing Economic Theories. London and New York: Routledge, 115-129.

Nisticò, Sergio. 2014. Production of (Pleasant) Time by Means of (Unpleasant) Time: Some Notes on Consumption Theory and Time Use. Metroeconomica, 65(2): 276-297.

Nisticò, Sergio. 2015. Enjoyment Takes Time: Some Implications for Choice Theory. Economics: The Open-Access, Open-Assessment E-Journal, 9(2015-8): 1-40.

Nisticò, Sergio. 2016. Some Notes on Gossen’s ‘Submerged and Forgotten’ Approach to Consumption and Time. CreaM Working Papers, 2016(1): 1-26.

Nisticò, Sergio and Domenico Tosato. 2002. Introduction. In Sergio Nisticò and Domenico Tosato (eds), Competing Economic Theories. London and New York: Routledge, 1-12.

Robinson, John P. and Geoffrey Godbey. 1997. Time for Life. The Surprising Ways Americans Use Their Time. University Park: The Pennsylvania State University Press.

Samuelson, Paul A. 1938. A Note on the Pure Theory of Consumer’s Behaviour. Economica, 5(17): 61-71.

Samuelson, Paul A. 1947. Foundations of Economic Analysis. Cambridge: Harvard University Press.

Schumpeter, Joseph A. 1987 [1954]. History of Economic Analysis. London: Routledge.

Shefrin, Hersch M. and Richard H. Thaler. 1988. The Behavioral Life-Cycle Hypothesis. Economic Inquiry, 26(4): 609-643.

Smith, Adam. 1976 [1776]. An Inquiry into the Nature and Causes of the Wealth of Nations. The Glasgow Edition of the Works and Correspondence of Adam Smith, vol. II, ed. by R. H. Campbell and Andrew S. Skinner. Oxford: Oxford University Press.

Steedman, Ian. 2001. Consumption Takes Time: Implications for Economic Theory. London: Routledge.

Stigler, George J. 1950. The Development of Utility Theory. I. Journal of Political Economy, 58(4): 307-327.

Thaler, Richard. 1985. Mental Accounting and Consumer Choice. Marketing Science, 4(3): 199-214.

Viner, Jacob. 1949. Bentham and J. S. Mill: The Utilitarian Background. The American Economic Review, 39(2): 360-382.

Winston, Gordon C. 1982. The Timing of Economic Activities. Cambridge: Cambridge University Press.

Haut de page


1 For an interesting review of an early debate on social interdependencies in consumption in the pre-axiomatic, Marshallian era of neoclassicism, see Bianchi and Sanfilippo (2015).

2 Chapters V and VII of Samuelson’s Foundations (1947) contain a brilliant summary of the different uses of the notion of utility in the early versions of neoclassical theory of demand. However, Samuelson merely follows Jevons and Walras in considering Gossen a mere precursor of the atemporal marginal utility notion defined in the quantity-utility space. For an account of Jevons’s and Walras’s unconvincing readings of Gossen, see section 2.

3 The implications of the role of time in consumption for individual wellbeing had already been thoroughly investigated also by Staffan Linder in The Harried Leisure Class (Linder, 1970).

4 This paper does not deal with paternalism and the ‘aggregation problem’ associated with Bentham’s utilitarianism, that is with the idea, in Viner’s words, that “it is the function of legislation to coerce or bribe individuals to make their behavior coincide with that required by the greatest-happiness principle, and of education and moral leaders to mould men's desires so that they spontaneously associate the happiness of others with their own happiness” (Viner, 1949, 365).

5 Ricardo and Marx considered the theory of component parts at odds with Smith’s idea that labour commanded explained the real price of commodities. For a critique of Marx’s and Ricardo’s reading of Smith’s theory of prices, see (Nisticò, 1991a; 1991b; 2002).

6 It is interesting to note that just a few years before, Becker tried to show that the “main implication [of the traditional theory of household choices]—that market demand curves are negatively inclined—can also be derived from a wide variety of irrational behavior” (Becker, 1962, 2).

7 The integration of the diary with the Pencil and Paper interview (PAPI) has proven to be very successful in raising the completion rate of the recent Italian survey on time use conducted by ISTAT to more than 70%. An important innovative feature could be the use, at least for a share of the units observed (those, whose hectic life is incompatible with the filling of a 24-hours diary) of a smartphone-based application, reminding the respondent, at regular time intervals, to specify the type of activity she or he is undertaking, its degree of pleasantness, its location and, possibly, other related details. So far, the only ‘technology-based’ surveys rely on ad hoc websites, rather than on ‘portable’ applications, whose use could significantly increase the response rate.

8 An interesting method combining diaries and ‘habits questionnaires’ to avoid the problems connected with high levels of non-response has been proposed by Gershuny (2012).

Haut de page

Pour citer cet article

Référence papier

Sergio Nisticò, « Consumption Choices and Time Use: History, Theory and Potential Empirical Evidence », Œconomia, 7-2 | 2017, 219-238.

Référence électronique

Sergio Nisticò, « Consumption Choices and Time Use: History, Theory and Potential Empirical Evidence », Œconomia [En ligne], 7-2 | 2017, mis en ligne le 01 juin 2017, consulté le 19 janvier 2019. URL : ; DOI : 10.4000/oeconomia.2708

Haut de page


Sergio Nisticò

University of Cassino and Southern Lazio,

Haut de page

Droits d’auteur

Licence Creative Commons
Les contenus d’Œconomia sont mis à disposition selon les termes de la Licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Haut de page
  • Logo Association Œconomia
  • Logo CNRS
  • OpenEdition Journals