RESEARCH ARTICLE | SUSTAINABILITY SCIENCE OPEN ACCESS The global human day William Fajzela, Eric D. Galbraitha,b,1 , Christopher Barrington-­Leighc,d , Jacques Charmese, Elena Friea, Ian Hattona, Priscilla Le Mézof , Ron Milog , Kelton Minorh , Xinbei Wani,j, Veronica Xiaa, and Shirley Xua Edited by B. Turner, Arizona State University, Tempe, AZ; received November 21, 2022; accepted April 18, 2023 The daily activities of ≈8 billion people occupy exactly 24 h per day, placing a strict phys- ical limit on what changes can be achieved in the world. These activities form the basis Significance of human behavior, and because of the global integration of societies and economies, many of these activities interact across national borders. Yet, there is no comprehensive Understanding how the global Downloaded from https://www.pnas.org by UNIVERSITAT AUTONOMA DE BARCELONA UAB IBB BIBLIOTECA on July 10, 2023 from IP address 158.109.174.204. overview of how the finite resource of time is allocated at the global scale. Here, we human system functions is estimate how all humans spend their time using a generalized, physical outcome–based crucial if we are to sustainably categorization that facilitates the integration of data from hundreds of diverse data- navigate planetary boundaries, sets. Our compilation shows that most waking hours are spent on activities intended adapt to rapid technological to achieve direct outcomes for human minds and bodies (9.4 h/d), while 3.4 h/d are change such as artificial spent modifying our inhabited environments and the world beyond. The remaining intelligence, and achieve global 2.1 h/d are devoted to organizing social processes and transportation. We distinguish development goals. But, the vast activities that vary strongly with GDP per capita, including the time allocated to food scope and diversity of human provision and infrastructure, vs. those that do not vary consistently, such as meals and transportation time. Globally, the time spent directly extracting materials and energy endeavors presents a major from the Earth system is small, on the order of 5 min per average human day, while challenge for holistic assessment. the time directly dealing with waste is on the order of 1 min per day, suggesting a large Here, we address this problem by potential scope to modify the allocation of time to these activities. Our results provide providing a global estimate of a baseline quantification of the temporal composition of global human life that can be time use by all humans, expanded and applied to multiple fields of research. integrating economic and noneconomic data within a time use | sustainability | global | sociology | economics consistent framework. Our At present, we lack a coherent global understanding of human activities. This is not to findings provide a bird’s eye say that the study of human activities has been overlooked. On the contrary, activities perspective on what our species comprise the core of our species’ behavior, and for decades they have been documented does, including how economic by diverse fields including economics (1–3), sociology (4, 5), history (6, 7), and anthro- activities fit into the backdrop of pology (8–10). However, economists have focused primarily on paid work activities, life, and reveal activities for relegating other activities to leisure or unpaid work, while sociologists, historians, and which there is significant anthropologists have often focused their attention on the activities that take place outside potential for change. the formal economy. Because of deep methodological differences, these studies are very rarely combined, and they have not been previously integrated at the global scale. A coherent interdisciplinary understanding of activities is important at present because, although the motivations for people to act are couched within the contexts of their own lives, activities are coordinated through economic and societal links to generate a globally integrated human system (11). The food we consume, the clothes we wear, and the material objects we use are largely produced by others in distant parts of the world. Similarly, threats to planetary boundaries, like climate change and biodiversity loss, are the collective outcomes of human activities across the planet (12, 13). Although the consequences of any human undertaking vary greatly with the available technology and other contextual features, the time spent on tasks is a key factor in determining outcomes, whether pro- Author contributions: W.F. and E.D.G. designed research; W.F., E.D.G., J.C., E.F., P.L.M., K.M., V.X., X.W., and S.X. ducing food, constructing buildings, or tackling environmental problems (14, 15). Because performed research; W.F., E.D.G., C.B.-­L., I.H., and R.M. global outcomes emerge from the sum of individual actions, it is crucial to understand analyzed data; and W.F., and E.D.G. wrote the paper. how global activities influence local changes and vice versa. The authors declare no competing interest. Compounding the disciplinary divisions is a geographic fragmentation of activity data. This article is a PNAS Direct Submission. The collection and analysis of activity data have tended to be carried out at the national Although PNAS asks authors to adhere to United Nations naming conventions for maps (https://www.un.org/ scale, tailored to the specific needs and objectives of individual countries, and most analyses geospatial/mapsgeo), our policy is to publish maps as have focused on wealthy populations. Although there have been some efforts to compare provided by the authors. sociological time use data for adults across countries (16–20), they have not previously Copyright © 2023 the Author(s). Published by PNAS. This open access article is distributed under Creative been used to characterize the global human system in a broader sense. Economic activities Commons Attribution-­NonCommercial-­NoDerivatives are more often captured in global databases (21), yet where global economic analyses have License 4.0 (CC BY-­NC-­ND). been carried out, they have typically focused on individual sectors of the economy, or 1 To whom correspondence may be addressed. Email: eric.galbraith@mcgill.ca. relied heavily on monetary valuations to combine activities among countries. Because This article contains supporting information online at wages and capital valuations can vary dramatically between countries, a monetary per- https://www.pnas.org/lookup/suppl/doi:10.1073/pnas.​ spective does not provide a clear picture of the ends to which humanity’s global supply of 2219564120/-­/DCSupplemental. labor is directed. Published June 12, 2023. PNAS 2023 Vol. 120 No. 25 e2219564120 https://doi.org/10.1073/pnas.2219564120 1 of 9 Table 1. Categories used to harmonize data, according to the Motivating-­Outcome-­Oriented General Activity Lex- icon (MOOGAL) Group Category Description External outcomes Food provision Providing food to humans, including agriculture and fishing, the processing of food items, cooking, serving, and cleanup Nonfood provision Providing raw materials and energy to the technosphere, including mining, lumber, fossil fuels, and renewable energy Technosphere modification The construction and maintenance of buildings, infrastruc- ture, and movable artifacts Maintenance of surroundings Cleaning surfaces and arranging the spaces that humans inhabit, taking care of accompanying plants and animals, disposing of wastes Direct human outcomes Somatic maintenance Caring for the cleanliness, appearance, and health of human Downloaded from https://www.pnas.org by UNIVERSITAT AUTONOMA DE BARCELONA UAB IBB BIBLIOTECA on July 10, 2023 from IP address 158.109.174.204. bodies, including medical care and childcare Deliberate neural restructuring Education, both formal and informal, research in the aca- demic and private sector, and religious activities Experience oriented Engaging in activities to provide desired experiences, includ- ing through use of media, interactive hobbies and sports, socializing, and meals Organizational outcomes Organization Activities that do not directly support any of the above outcomes, but instead serve to change the locations of entities, or allocate the time and access rights of humans, including through commerce, finance, real estate, law, and governance In contrast to monetary metrics, which are not strictly physical, global human day. We combined available data collected by all humans exist for exactly 24 h per day and spend each minute national statistics agencies, international organizations, and doing something (22, 23). This 24-­h time budget constraint is appli- researchers from over 140 countries, wherever available during cable at any scale, as well as over the history of human development. the period 2000 to 2019 to avoid the economic and social A complete and holistic quantification of how global humanity allo- disruption of the COVID-­19 pandemic (SI Appendix, Tables S1 cates its ~190 billion hours per day could therefore provide a firm and S2). We interpolated within geographical regions to coun- grounding from which to assess how human behavior is changing tries with incomplete or missing data in order to account for over time, as well as the scope and plausibility of strategies to simul- undersampled populations. We assessed the full human lifespan taneously achieve multiple goals, such as the 17 internationally by weighting population-­specific time use estimates using age-­ agreed upon Sustainable Development Goals (SDGs) (24, 25). For structured demographic data (Methods). example, time provides a simple basis for assessing the overall feasi- Our approach is enabled by a generalized categorization of bility of reallocating labor to constructing nonfossil fuel energy sys- activities (29), the Motivating-­Outcome-­Oriented Generalized tems (advancing SDGs 7 and 9) or dealing with plastic pollution Activity Lexicon (MOOGAL), which allows for the integration (SDGs 12, 14, and 15) while maintaining meaningful employment of data originally collected for diverse sociological, economic, and in a globalized economy (SDGs 8 and 9). It also represents an impor- anthropological purposes. The lexicon is comprised of eight cat- tant, human-­centered perspective on development and the evolution egories (Table 1), which are subdivided into 24 subcategories of human experience in the face of social and technological shifts, (Table 2). The subcategories are described in physical, rather than including the accelerated transformation of labor markets through colloquial, terms to limit ambiguity in their application across urbanization, automation, and artificial intelligence (26–28). cultures. Since the MOOGAL lexicon is designed to combine Making use of the 24-­h constraint requires a holistic assessment economic and noneconomic data, it differentiates based on the of activities. The disciplinary division between paid and unpaid activ- motivating outcome that causes the activity to be undertaken, ities can be resolved by combining observations from economic and rather than whether or not the activity is undertaken for pay. For sociological time use data with a harmonized set of activity categories. example, both paid daycare work and unpaid care of young chil- Given that global economic networks exchange vast quantities of dren by parents are classified under physical childcare, while food materials and goods across national borders, understanding how the preparation includes both cooking at home and working at a res- time budget constraint relates to physical outcomes also requires taurant. Similarly, the time invested by humans as both bus pas- assessing activities for the complete population of the world. sengers and bus drivers would be included together within human Combining national data requires taking into account discrepancies transportation, since changing the locations of humans is the in the subsets of the populations surveyed (e.g., labor force and legal intended outcome for both activities. We produced concordance adults), as well as addressing numerous variations in reporting con- matrices for crossmapping all time use survey and economic activ- ventions and activity categorizations. These methodological hurdles ities to the MOOGAL subcategories, resulting in 3,956 MOOGAL have impeded the development of a unified global perspective on subcategory definitions. Although it is not common to report activities. uncertainty estimates for time use data, we endeavored to provide a partial estimate of the uncertainty range for each global average A Holistic Estimate of Global Time Use value. This uncertainty range aims to represent uncertainty in the initial data, the association to MOOGAL subcategories, and from Here, we assemble a complete estimate of what humans are interpolation to countries with missing data, and should be seen doing, averaged over time and across the entire population, to as approximate (see SI Appendix for a detailed description of the provide an aggregated high-­level view that we refer to as the full method). 2 of 9 https://doi.org/10.1073/pnas.2219564120 pnas.org Table 2. Subcategories of the MOOGAL Category Subcategory Definition Examples Food provision Food growth & collec- All activities related to the Crop and animal production. Fishing, hunting, tion growth of edible organic and trapping. Ploughing, clearing of land, matter, its collection, and sowing, planting, transplanting. Kitchen initial storage gardening. Collecting, storing, and stocking of products. Fish farming. Gathering wild products Food processing Transformation of food to pre- Food manufacturing. Milling, husking, pound- vent spoilage and detoxifica- ing. Beverage manufacturing. Food process- tion, or to create storable ing and preservation. Jarring and canning beverage and food products Food preparation Transformation within days or Cooking. Preparing meals for the home. hours of eating. Includes Washing dishes. Catering. Food and bever- Downloaded from https://www.pnas.org by UNIVERSITAT AUTONOMA DE BARCELONA UAB IBB BIBLIOTECA on July 10, 2023 from IP address 158.109.174.204. cleanup of preparatory age preparation and serving. Parboiling. surfaces, serving, and Bread baking. Serving meals/snacks. Clearing washing of dishes table Nonfood provision Materials The extraction of substances Mining and quarrying. Digging out clay, gravel, from the Earth system to be and sand. Mining of metal ores. Forestry and used for artifacts, buildings, logging. Nonferrous metal mining. Stone and infrastructure cutting Energy Extraction and provision of Oil and gas extraction. Electric power genera- energy tion, transmission, and distribution. Gas production and supply industry. Petroleum processing. Mining of uranium and thorium ores. Gathering firewood and other natural products used as fuel Technosphere Buildings Construction and mainte- Home maintenance. DIY home improvement. modification nance of residential, Construction work. Building and extension of commercial, and industrial dwelling. Construction activities for own buildings home Infrastructure Construction and mainte- Civil engineering. Telecommunications. nance of structures to Construction of roads, railways, and bridges facilitate the transport of people, materials, and information Artifacts Creation and maintenance of Manufacturing of base goods. Manufacturing movable objects of textiles. Manufacturing of pharmaceuti- cals. Manufacturing of computer, electronic, and optical products. Making handicrafts, pottery, printing, and other crafts. Assem- bling machines, equipment, and other products. Vehicle maintenance and repairs. Production of goods for own household use Maintenance of Inhabited environment Maintaining the cleanliness Laundry. Indoor cleaning. Washing clothes surroundings and order of inhabited and shoes. Ironing. Cleaning dwelling. Care spaces and materials, of house plants. Ground maintenance. Pet including home, workspace care. Care of textiles interiors, and grounds Waste management Dealing with waste and Waste management and remediation services. unintended by-­products Sewerage. Recycling. Sewage and refuse outside of inhabited build- disposal, sanitation, and similar activities. ings and their immediate Waste disposal. Removing trash environment Somatic Hygiene & grooming Maintaining the cleanliness Washing yourself, getting dressed. Bathing. maintenance and appearance of the body Personal care. Grooming. Private activities. Personal hygiene Physical childcare Physical and practical care of Physical care of children: washing, dressing, young people, including and feeding. Supervising children needing cleaning, feeding, and care. Minding children. Physical care of minding young children to preschool children ensure safety Health care Medical care and physical Medical care for family members. Health care support to persons in need to oneself. Medical examination or treat- ment. Physical care of sick or disabled adult. Receiving medical/personal care from profes- sionals. Mental health PNAS 2023 Vol. 120 No. 25 e2219564120 https://doi.org/10.1073/pnas.2219564120 3 of 9 Table 2. (Continued) Category Subcategory Definition Examples Sleep & bedrest Time spent in bed and/or Sleeping, naps, sick in bed. Incidental sleeps sleeping and naps. Bedridden due to disease Deliberate neural Schooling & research Deliberate education and Education. Attending class. Homework and restructuring research activities research. Studying and learning. Remote education learning activities. School, tech- nikon, college, university attendance Religious practice Religious practice and Ritual ceremonies. Praying. Religious activities. ceremonial, social, or cultural Private prayer, meditation, and other events spiritual activities Organization Material transportation Transport undertaken to Road freight transport. Shipping. Loading, move artifacts, raw materi- unloading, handling, and other transporta- Downloaded from https://www.pnas.org by UNIVERSITAT AUTONOMA DE BARCELONA UAB IBB BIBLIOTECA on July 10, 2023 from IP address 158.109.174.204. als, and food tion services. Postal service, couriers, and messengers. Warehousing industry. Trans- porting in vehicles. Fetching of water Human transportation Transport of persons for the Travel to/from work. Travel for social and purpose of changing their cultural activities. Travel to or from school. location Commuting, job, and study-­related travel. Public transport. Transport of passenger by motorized and nonmotorized transports. Journeying Allocation Activities that are not directly Wholesale and commission trade. Retail. motivated by a specific Banking. Financial and insurance industry. outcome for humans or the Public administration and defense. Local external world, but instead government services. Extraterritorial contribute to determining organizations and bodies. Accessing govern- the allocation of time and ment services. Real estate industry. Legal access rights to humans and accounting activities. Petty trading, street and door-­to-­door vending, hawking. Grocery shopping. Purchasing goods. Shop online stores. Paying household bills. Household management. Job search Experience oriented Meals Activities centred on eating Eating and drinking. Eating meals/snacks. and drinking, including Pubs and restaurants. Coffee, refreshments. associated socializing Meals associated with work. Visit to restau- rant, café, bar Active recreation Recreation that involves an Active sports. Ball games. Walks. Wushu and elevated metabolic activity, Qigong. Hiking. Walks in forest and on land. whether purely for the Walking the dog. Water sports experience or including a fitness motivation Passive, interactive, and Activities undertaken for the Watching TV. Listening to radio, personal social purpose of producing a media device, or other audio. Reading. Using desired experience, including computer to read and watch/listen to passive observation of media programs. Doing nothing, rest, and relaxa- or surroundings, interactive tion. Arts and hobbies. Computer games. engagement with devices or Visual, literal, and performing arts. Museum/ other people, and socializing exhibition. Spectator to sports, exhibitions, concerts. Socializing. Attending or hosting social events. Telephone calls. Discussing, gossiping. Family and socializing. Visiting relatives and friends The Global Human Day. Our resulting estimate of the global human comprising the largest group, are motivated by the immediate day is shown in Fig. 1, reported as the number of hours per day consequences they have on humans. These activities include engaged in each activity, averaged across all humans, where the area taking care of the appearance, cleanliness and health of human of each colored cell is proportional to the amount of time. Sleep and bodies, the deliberate restructuring of human neural pathways, bedrest, the largest category (9.1 ± 0.4 h), is shown as the adjacent and the generation of desired experiences. The most crescent. This sleep estimate is significantly larger than the global time-­consuming subcategory is passive, interactive, and social average of 7.5 h of sleep per day recorded among adults by wearable activity which includes reading, watching screens, playing games, devices (30), a difference we attribute to the inclusion of children in going for walks, socializing, and sitting doing nothing, and occu- our estimate and to the time spent in bed but not sleeping (Methods). pies an average of 4.6 ± 0.3 h or ≈31% of the average waking For the ≈15 h per day of human life not devoted to sleeping day. The second large group includes activities motivated by and bedrest, the activity subcategories can be summarized external outcomes (3.4 h), i.e., intended to produce physical according to three large groups. Direct human outcomes (9.4 h), changes in the world outside humans themselves. These changes 4 of 9 https://doi.org/10.1073/pnas.2219564120 pnas.org Downloaded from https://www.pnas.org by UNIVERSITAT AUTONOMA DE BARCELONA UAB IBB BIBLIOTECA on July 10, 2023 from IP address 158.109.174.204. Fig. 1. The global human day, including both work and nonwork activities. The time devoted to each activity, averaged across the entire human population of ≈8 billion people, is indicated by the area of each colored shape in the Voronoi diagram. Direct human outcome activities aim to modify the bodies, neural structures, and experiences of humans. Activities with external outcomes are intended to modify the immediate surroundings of humans, including construction and maintenance of the technosphere, and the provision of food, energy, and materials from the Earth system. Activities with organizational outcomes include moving humans and cargo, as well as activities that allocate labor and access rights such as trade, finance, law, and governance. The time spent in each subcategory is listed below the diagram, in hours per day, with approximate confidence intervals that reflect contributions from the original data sources, interlexicon associations, and interpolation. include extracting materials and energy from the natural envi- a unique perspective on how economic activities fit into the overall ronment, producing food, creating and maintaining movable distribution of human activities. Economic activities are defined here objects and immovable constructions, and maintaining the as those within the scope of what the International Labor Organization cleanliness and tidiness of the spaces humans inhabit. Activities considers “employment,” including work for pay or profit as well in the third large group are motivated by organizational out- as the production of nonmarket goods within households. These comes (2.1 h), including activities that modify the locations of economic activities account for ≈2.6 h (158 min), roughly 11% of humans and materials, and an array of activities that are not the global human day, or one-­sixth of waking hours over the average directly motivated by particular physical outcomes, but instead lifetime. Although this may appear small, it is equivalent to a 41-­h serve to allocate the time use and access rights of humans (29). work week among the global labor force (which is approximately 66% Allocation is achieved by mechanisms that vary between cultures of the working-age population, those aged 15 to 64 y). and economic systems, including legal and political systems, When the global economic activity is viewed on its own (Fig. 2), finance, policing, and shopping. we see that almost one-­third involves the growth and collection of food (44 ± 3 min), mainly in the form of agriculture. Roughly one Time Devoted to the Global Economy. Because our analysis includes quarter of economic activity is dedicated to allocation (37 ± 2 min), both economic and noneconomic data at the global scale, it enables which includes retail, wholesale, real estate, insurance, finance, law, PNAS 2023 Vol. 120 No. 25 e2219564120 https://doi.org/10.1073/pnas.2219564120 5 of 9 Daily minutes spent in economic activity, by outcome Averaged over all humans e neural restructuring Technosphere T infrastructure 7 min creation schooling & 35 min research passive, Experience-oriented 2 min interactive & artifacts religious practice active social Somatic rec Maintenance 5 min buildings of Downloaded from https://www.pnas.org by UNIVERSITAT AUTONOMA DE BARCELONA UAB IBB BIBLIOTECA on July 10, 2023 from IP address 158.109.174.204. inhabited Surroundings health 5 min care environment material waste physical transportation childcare hygiene & grooming energy human food transportation Non-food preparation materials Provision 5 min allocation food growth & collection Organization Food 47 min Provision food processing 52 min Food Food growth Technosphere 44 min & collection 22 min Artifacts Non-food 4 min Materials Maintenance of Inhabited provision creation provision surroundings 4 min environment 52 min 35 min 5 min 5 min 5 min Food preparation 10 min Buildings 0.4 Waste 2 min Energy min management Food 3 min processing 3 min Infrastructure Somatic Organization maintenance 4 min Health Care 47 min 37 min 5 min Deliberate Schooling Experience- Passive, 7 min & research neural 1 min Hygiene & oriented 2 min interactive & Material restructuring grooming 2 min social 7 min transportation 7 min 0.2 Religious 0.2 practice 0.4 Physical Active min min recreation Human min childcare 2 min transportation Fig. 2. The global economic day. Voronoi tree is calculated as in Fig. 1, for the average time spent in paid employment and unpaid or other own-­use/household production of goods, averaged across the global population. Average times per subcategory are shown at the bottom of the figure, in minutes per day. The sum of all economic activities is ≈2.6 h per day, equivalent to a 41-­h work week among ≈66% of the working-­age population. and governance. The production of artifacts, which includes the man- ($US PPP) as a proxy. Our data reveal particularly notable trends in ufacture of vehicles, machinery, electronics, domestic appliances, and four activities vs. GDP per capita, shown in Fig. 3. The time spent all other movable goods as well as their intermediate components, growing and collecting food is large in low-­income countries (>1.0 h) accounts for roughly one-­seventh of the total economic activity (22 but becomes very small in high-­income countries (<5 min). This ± 2 min). The remaining economic time is mostly partitioned among striking trend can be largely attributed to labor-­saving technologies the construction and maintenance of buildings, freight and other (27) that allow the same amount of food to be produced with an material transportation, food preparation, and schooling and research. order of magnitude less time. The decrease of approximately 1.2 h in food growth and collection over the income range is roughly Variation with Material Wealth. Because our data compilation counterbalanced, perhaps unsurprisingly, by an increase in the time includes formal, informal, subsistence, domestic and care work, spent engaged in experience-­oriented activities (passive, interactive, as well as nonwork activities, normalized to total populations, it and social interactions plus meals and active recreation, a ~1.5 h allows a comprehensive comparison of how activities vary between increase). There are also significant increases of the time spent in countries. To provide an overview, we show how activities vary allocation activities (~0.4 h) and on infrastructure construction and in relation to material wealth, for which we use GDP per capita maintenance (~0.1 h) across the range of GDP per capita. 6 of 9 https://doi.org/10.1073/pnas.2219564120 pnas.org A B C D E F G H Downloaded from https://www.pnas.org by UNIVERSITAT AUTONOMA DE BARCELONA UAB IBB BIBLIOTECA on July 10, 2023 from IP address 158.109.174.204. Fig. 3. Activities vs. GDP per capita at the country scale. Each circle represents the population-­average time per day for one country. Panels A–D show subcategories with highly significant linear trends, while panels E–H do not show significant trends. Panels A, B, and D include only countries with economic data, while other panels include only countries with time use survey data. In contrast, other activities are remarkably invariant vs. GDP terms: When summed over the global population, ≈780 million per capita. Food preparation (0.9 ± 0.1 h), human transportation person-­hours are dedicated to material extraction and energy (0.9 ± 0.2 h), hygiene and grooming (1.1 ± 0.2 h), and meals (1.6 provision daily, equivalent to ≈285 billion person-­hours per ± 0.2) show no detectable trends with GDP per capita (Fig. 3 year. On their own, these large sums might give the impression E–H). This invariance does not imply that the portions of time that material extraction and energy provision comprise a major devoted to these activities are universal across humans, as they component of human activity. But instead, when viewed as a certainly vary among individuals. Rather, our results do not show relative fraction of the whole, they are found to be remarkably a consistent variation with GDP at the population level, suggesting small compared to activities such as hygiene and grooming, that material wealth does not play a large role in determining the which consume ≈3.2 trillion person-­hours per year, roughly allocation of time to these activities. Together, these wealth-­invariant 12-­fold more. activities comprise 4.5 ± 0.4 h or 30% of the waking day. The concept of Energy Return On Investment (EROI) offers an The relatively constant time spent in human transportation is interesting perspective on the human time invested in energy provi- particularly notable, given that travel is often thought of as a cost sion (37). An EROI ratio compares one form of usable energy with that might be alleviated with technology (31), analogous to the large the energy investment required to provide it. Our results allow a effect of labor-­saving technology on agriculture (27). Instead, our unique assessment of the average return on human metabolic energy data imply that material wealth has a negligible effect on travel time investment. Although the usable energy could be quantified at mul- at the population level (Fig. 3F). This supports long-­standing spec- tiple points, for the sake of argument, we choose the final global ulation regarding an inherent travel time budget (32–34), which was energy supply, approximately 13 TW in 2019 (38), to compare with originally based on more geographically limited data, and is consist- the human metabolic energy expenditure used in the activity of ent with the hypothesis that the built environment and transporta- providing that energy supply. If we assume a typical adult human tion technology tend to coevolve to maintain a psychologically mass of 70 kg, with a basal rate of 1.1 W kg−1, working at a typical bounded average daily travel time (35, 36). This general observation rate of 2.5-­fold the basal rate (39), our results imply that the average has implications for the potential to reduce transport-­related energy global rate of human metabolic energy invested in all forms of energy expenditure: If travel time is relatively invariant at the population provision (2 min per day) is ~2 GW. This is equivalent to an EROI scale, per capita energy consumption can only be reduced by decreas- ratio for the human to final energy supply of more than 5,000-­fold. ing the energy consumption per travel time. It follows that the energy This EROI ratio is a factor of 5 to 10 lower than values estimated cost per travel time is the key variable of interest, and that—on their previously for Italy and the United States by ref. 40, which we attrib- own—reductions of the average energy cost per distance traveled are ute to a combination of international exchange and the inclusion of unlikely to reduce overall energy consumption in daily personal low-­efficiency energy forms such as firewood collection in our global transport. estimate. We emphasize that this is a crude estimate, but it shows the extremely high average EROI, relative to metabolic energy, Time for Sustainable Action. Our analysis shows that the achieved by the global human system. activities through which humans directly modify the state of Because the global supply of energy and materials is currently our planet account for a relatively small portion of the global provided with a small fraction of the total time (amounting to ≈ human day. For example, the extraction of all raw materials 3% of the global economic time), the time allocated to these such as wood, minerals, and rock requires only 4 ± 1 min, while activities could be altered to a relatively large degree without nec- energy provision, including the extraction and refining of all essarily having a large impact on the time allocated to other activ- fossil fuels, is achieved with 2 ± 1 min. This is not to say that ities. This assessment suggests that climate change solutions, such these activities are minor in aggregate physical or economic as shifting labor away from fossil fuel industries and into the PNAS 2023 Vol. 120 No. 25 e2219564120 https://doi.org/10.1073/pnas.2219564120 7 of 9 construction of global renewable power infrastructure, are highly Time Use Surveys. To provide a baseline for average daily time use, we obtained feasible in terms of the global time budget, in that there exists a nationally representative time use surveys from 58 countries comprising approx- clear physical scope for humans to reallocate time among the imately 60% of the world population. These surveys are conducted by national relevant activities without significantly disrupting the overall dis- statistical agencies with the goal of providing a broad understanding of how the tribution of time at the population level. The potential of population allocates their time to a set of activities. Most of these surveys report labor-­saving technologies to reduce the time required for food any formal economic participation as a single activity, e.g., “work for employment” growth and collection, highlighted above, suggests that a synergy (with some exceptions). Survey data are collected via self-­completed time diaries, might exist between the mechanization and electrification of agri- telephone or in-­person interviews, online questionnaires, or a combination of these culture in low-­income countries (41) and the construction of solar methods. The measure we use is the population-­weighted average daily time spent on each activity among all respondents, which is the product of the participant time energy infrastructure. This does not speak to the policy and eco- (the average time spent on each activity only among those who engaged in the nomic measures required to engineer such transitions, which are activity) and the participation rate (the percentage of all respondents engaging in likely to be highly challenging. But, it does indicate that a future the activity). When available, the aggregate survey data were downloaded from the with quite different infrastructure and energy flows can be phys- respective national statistical agency database, and translated as necessary. If the ically achieved without requiring a great disruption in the overall Downloaded from https://www.pnas.org by UNIVERSITAT AUTONOMA DE BARCELONA UAB IBB BIBLIOTECA on July 10, 2023 from IP address 158.109.174.204. aggregated datafile could not be located, the relevant table from the survey report composition of human activities. was manually transcribed to a computer-­readable format. The full list of countries, Meanwhile, the amount of time devoted to dealing with waste, including the source location, is given in SI Appendix, Table S1. outside of our dwellings and their immediate surroundings, appears The quality of each survey is assessed according to several key characteristics, to be very small (≈1 min). We caution that our assessment of waste notably survey duration, data collection method, and lexicon length. Three quality management time may have been unable to capture time invest- levels (A, B, and C) are associated with a 5, 10, and 20% baseline uncertainty on ment by some government agencies, consultants, and the informal the time values reported in the survey, respectively. For details regarding the economy, so the ~1 min per day may be an underestimate. quality assessment, see SI Appendix, section 6. Nonetheless, this small time investment stands in stark contrast to Economic Activity. The main source for economic data was ILOSTAT, the online the time spent cleaning and arranging within our dwellings (≈40 repository of labor statistics managed by the International Labour Organization min). It seems plausible that many waste problems, including the (ILO) Department of Statistics. Mean annual employment and mean weekly work- accumulation of ocean plastics and water contamination by toxins, ing time data, recorded under the International Standard Industrial Classification could potentially be greatly alleviated through a relatively small (ISIC) of economic activities, were used to calculate mean daily working time of reallocation of the total human time budget. Motivating such shifts the entire population. We also obtained the comparable economic activity data of time allocation requires dedicated policy and economic strategies, for Canada, China, Japan, and Russia, which did not archive data with the ILO. In but our analysis suggests that the time is available for the global total, economic data were available for 139 countries, representing 86% of the human system to address 21st-­century sustainability goals. world population. Outlook. The holistic approach to time use presented here can Youth Population. While the time use surveys are nationally representative, many serve as a foundation for future work. Additional dimensions do not include youth below a minimum age, which varies by country between 14 of global time use, beyond the physical outcomes focused and 18 y. The absence of children from most time use surveys can be generally upon here, can be resolved through the global application of attributed to the view of time use as a metric of human capital, a context in which complementary standardized lexicons (29). These additional children are not considered useful (44). In order to correct for the consequent bias, dimensions might include social interactions, physical context, while also providing a complete description of human time, we assembled a comple- or technology use. The geographic distribution of activities mentary dataset on the time use of youth aged 0 to 17 y and used this to construct an age-­structured model of total youth time. Student enrollment data from the World can be directly linked to material, energetic, and monetary Bank and UNICEF were paired with educational instruction time from the OECD to flows, as well as to subjective experiences, to enhance process calculate average daily time spent in schooling for each age. Youth employment data understanding. Changes over time in the global human day, were taken from the 2016 ILO Global Estimates of Child Labour report, which covers informed both by historical records and ongoing monitoring, children between ages 5 and 17 y in 105 countries and provides child employment can potentially provide further insight on long-­term behavioral rates in agriculture, industry, and services. Youth working hours were obtained from mechanisms and the roles that changes in time allocation play the World Bank. The activities occupying the remainder of average daily time were in societal transitions. Globally standardized time use patterns estimated using data averaged from time use studies conducted in 10 countries, can be applied as human-­focused alternatives to GDP (42), as well as a global youth sleep time study (see SI Appendix, section 4 for details). and calculations of potential changes in time use may help Sensitivity tests without the youth model are provided in SI Appendix, section 9. to chart pathways toward SDGs, applicable for planetary-­scale governance (43). Time, it has been said, is the coin of life—and Sleep. Sleep, as recorded by wearable devices among >18-­y-­old adults, consumes 7.5 h in a globally connected society, it is essential to have a thorough per day, while the self-­reported estimates averaged over all ages result in an average global understanding of how that coin is spent. of 9.1 ± 0.4 h of sleep plus resting in bed, or being in bed but not sleeping. The 1.6 h discrepancy is partly attributable to the well-­characterized overestimation of total sleep time in time diaries (45, 46). Prior comparisons of sleep time have found this difference Methods to exceed 1 h (45) due to the inclusion of sleep onset latency, offset, nighttime waking, and other activities in the self-­reported sleep time. In addition, youth, who are included Our estimate of the global human day is constructed as part of the Human in our global estimate but not the wearables data, generally sleep longer hours. Chronome Project (https://humanchronome.org/), based on three primary com- ponents, namely nationally representative time use surveys, national statistics Activity Categories. Given the diversity of our data sources, it was necessary to of employment and working time according to economic activity, and a mul- crossmap across a large number of activity categorization systems, known as lexi- ticomponent time use model for youth aged 0 to 17 y. We also include sleep cons. Activities were reclassified according to the motivating outcomes that cause the data from wearable devices for comparison with time use survey data. The data activities to be undertaken, using the MOOGAL as described in ref. 29. The MOOGAL sources are briefly described below, followed by the strategy used to interre- subcategories are intended to apply to any human population and epoch and are late activities through a common lexicon, and a brief discussion of uncertainty. generally well aligned with commonly used sociological and economic lexicons A more thorough description of the method is provided in SI Appendix, including (HETUS, ICATUS, ISIC) for the majority of activities. Activities that are coordinated supplementary figures and tables. between multiple people, such as those carried out in economic activities organized 8 of 9 https://doi.org/10.1073/pnas.2219564120 pnas.org by firms, are categorized according to the motivating final output, consistent with regions based on the ILO subregions and separately interpolate both time use standard economic practice. We apply a priority scheme such that Priority 1 subcate- and economic data to the missing countries in each region. For each missing data gories are identified in preference to Priority 2 where both co-­occur, and Priority 3 has type, countries are filled using the population-­weighted average of the sampled the lowest priority. Thus, Priority 3 experience–oriented activities (e.g., reading) are countries in the same region, for each subcategory. only coded as such if they are not identified as contributing to a Priority 1 or Priority 2 outcome (e.g., schooling and research). Because activities are frequently reported Data, Materials, and Software Availability. All data used in this work, in terms that include more than one MOOGAL outcome, our coding system allows as well as scripts to compile the results, are deposited in zenodo (10.5281/ fractional partitioning of an observed activity between MOOGAL subcategories. For a zenodo.7941615) (47). given lexicon of length n, each activity is associated to the 24 MOOGAL subcategories ACKNOWLEDGMENTS. We thank Victoria Reyes-­García, Jeroen Van Den Bergh, in an n × 24 matrix by assigning a relative weight between 0 and 1 to each subcat- David Carozza, Dror Etzion, Majdi Hunter-­Batal, Maxwell Kaye, Maria Pastor, Lior egory. These weightings indicate the portion of time from the original activity that is Greenspoon, Eric Kolaczyk and Brian Robinson for helpful comments and discussion. associated with the given subcategory, as estimated by human coders. A weighting of 1 indicates that the activity is uniquely associated with the single subcategory, while a weighting of 0 indicates that the activity is entirely excluded from that sub- Downloaded from https://www.pnas.org by UNIVERSITAT AUTONOMA DE BARCELONA UAB IBB BIBLIOTECA on July 10, 2023 from IP address 158.109.174.204. category. An activity that cannot be entirely associated with a single subcategory is Author affiliations: aDepartment of Earth and Planetary Sciences, McGill University, Montréal, split fractionally among subcategories, that together sum to 1. All activity definitions QC H3A 0E8, Canada; bInstitut de Ciència i Tecnologia Ambientals, Autonomous University of Barcelona, 08193 Cerdanyola del Vallès, Barcelona, Spain; cInstitute for Health and Social were estimated independently by at least three coders, discrepancies were reviewed, Policy, Faculty of Medicine and Health Sciences, McGill University, Montréal, QC H3A 1G1, and mean values were used wherever unequal but defensible estimates co-­occurred. Canada; dBieler School of Environment, McGill University, Montréal, QC H3A 2A7, Canada; e Centre for Population and Development, Institute of Research for Development, University Interpolation to Countries with Missing Data. Our dataset includes direct of Paris, 75006 Paris, France; fLaboratoire de Météorologie Dynamique, Institut Pierre Simon Laplace, Ecole Normale Supérieure Ulm, 75006 Paris, France; gDepartment of Plant observations for 145 countries. Of these, both time use and economic data were and Environmental Sciences, Weizmann Institute of Science, Rehovot 76100, Israel; hData available for 52 countries, time use data alone were available for 6 countries, and Science Institute, Columbia University, New York, NY 10027; iDepartment of Epidemiology, Biostatistics, and Occupational Health, McGill University, Montreal, QC H3A 1Y7, Canada; economic data alone were available for the remaining 87 countries. To assess and jInstitute of Public Health, Epidemiology, and Development, College of Health Sciences, the entire global human population, we group countries into 17 geographic Université de Bordeaux, Bordeaux 33076, France 1. G. S. Becker, A theory of the allocation of time. Econ. J. 75, 493–517 (1965). 24. L. M. Fonseca, J. P. Domingues, A. M. Dima, Mapping the sustainable development goals 2. A. B. Krueger, D. Kahneman, D. Schkade, N. Schwarz, A. A. Stone, National time accounting: The relationships. Sustainability 12, 3359 (2020). currency of life in Measuring the Subjective Well-­Being of Nations: National Accounts of Time Use 25. United Nations, “Transforming Our world: The 2030 agenda for sustainable development and Well-­Being (University of Chicago Press, 2009), pp. 9–86. (Department of Economic and Social Affairs, 2015). 3. M. Aguiar, E. Hurst, The macroeconomics of time allocation in Handbook of Macroeconomics 26. S. Vallas, J. B. Schor, What do platforms do? Understanding the gig economy Annu. Rev. Sociol. 46, (Elsevier, 2016), pp. 203–253. 273–294 (2020). 4. B. Cornwell, J. Gershuny, O. Sullivan, The social structure of time: Emerging trends and new 27. A. D. Foster, M. R. Rosenzweig, Economic development and the decline of agricultural employment. directions. Annu. Rev. Sociol. 45, 301–320 (2019). Handb. Dev. Econ. 4, 3051–3083 (2007). 5. A. Szalai, The Use of Time: Daily Activities of Urban and Suburban Populations in Twelve Countries 28. D. Acemoglu, P. Restrepo, Automation and new tasks: How technology displaces and reinstates (Mouton, 1972). labor. J. Econ. Perspect. 33, 3–30 (2019). 6. E. P. Thompson, Time, Work-­Discipline and Industrial Capitalism. Past Present, 56–97 (1967). 29. E. Galbraith et al., Interdisciplinary applications of human time use with generalized lexicons. PLoS 7. H.-­J. Voth, Time and work in eighteenth-­century London. J. Econ. Hist. 58, 29–58 (1998). One 17, e0270583 (2022). 8. D. R. Gross, Time allocation: A tool for the study of cultural behavior. Annu. Rev. Anthropol. 13, 30. S. S. Jonasdottir, K. Minor, S. Lehmann, Gender differences in nighttime sleep patterns and 519–558 (1984). variability across the adult lifespan: A global-­scale wearables study. Sleep 44, zsaa169 (2021). 9. A. Johnson, C. Behrens, Time allocation research and aspects of method in cross-­cultural 31. S. Luh, R. Kannan, T. J. Schmidt, T. Kober, Behavior matters: A systematic review of representing comparison. J. Quant. Anthropol. 1, 234–45 (1989). consumer mobility choices in energy models. Energy Res. Soc. Sci. 90, 102596 (2022). 10. R. Bhui, M. Chudek, J. Henrich, Work time and market integration in the original affluent society. 32. D. Metz, Time constraints and travel behaviour. Transp. Plan. Technol. 44, 16–29 (2021). Proc. Natl. Acad. Sci. 116, 22100–22105 (2019). 33. Y. Zehavi, “The “UMOT” project” (US Department of Transportation, 1979). 11. J. Liu et al., Framing sustainability in a telecoupled world. Ecol. Soc. 18, 1–19 (2013). 34. A. Schafer, D. G. Victor, The future mobility of the world population. Transp. Res. Part Policy Pract. 34, 12. S. Motesharrei et al., Modeling sustainability: Population, inequality, consumption, and 171–205 (2000). bidirectional coupling of the Earth and Human Systems. Natl. Sci. Rev. 3, 470–494 (2016). 35. G. Hupkes, The law of constant travel time and trip-­rates. Futures 14, 38–46 (1982). 13. W. Steffen et al., Planetary boundaries: Guiding human development on a changing planet. Science 36. C. Marchetti, Anthropological invariants in travel behavior. Technol. Forecast. Soc. Change 47, 75–88 347, 1259855 (2015). (1994). 14. C. W. Cobb, P. H. Douglas, A theory of production. Am. Econ. Rev. 18, 139–165 (1928). 37. C. A. Hall, J. G. Lambert, S. B. Balogh, EROI of different fuels and the implications for society. Energy 15. E. D. Galbraith, Earth system economics: A biophysical approach to the human component of the Policy 64, 141–152 (2014). Earth system. Earth Syst. Dyn. 12, 671–687 (2021). 38. International Energy Agency, “Key World Energy Statistics 2021” (2021). 16. J. Charmes, Time use across the world: Findings of a world compilation of time use surveys (UNDP 39. J. Deyaert, T. Harms, D. Weenas, J. Gershuny, I. Glorieux, Attaching metabolic expenditures to Human Development Report Office Background Paper N. Y., 2015). standard occupational classification systems: Perspectives from time-­use research. BMC Public 17. K. Fisher, J. Gershuny, A. Gauthier, Multinational time use study: User’s guide and documentation Health 17, 1–10 (2017). (Centre for Time Use Research University of Oxford, 2012). 40. M. Giampietro, K. Mayumi, The Biofuel Delusion: The Fallacy of Large Scale Agro-­Biofuels Production 18. J. Charmes, Variety and change of patterns in the gender balance between unpaid care-­work, paid (Routledge, 2009). work and free time across the world and over time: A measure of wellbeing? Wellbeing Space Soc. 41. T. Daum, R. Birner, Agricultural mechanization in Africa: Myths, realities and an emerging research 3, 100081 (2022). agenda. Glob. Food Secur. 26, 100393 (2020). 19. J. I. Gimenez-­Nadal, A. Sevilla, Trends in time allocation: A cross-­country analysis. Eur. Econ. Rev. 56, 42. R. Costanza, M. Hart, J. Talberth, S. Posner, Beyond GDP: The need for new measures of progress” 1338–1359 (2012). (The Pardee Papers, 2009). 20. G. Vagni, B. Cornwell, Patterns of everyday activities across social contexts. Proc. Natl. Acad. Sci. 115, 43. F. Biermann et al., Navigating the Anthropocene: Improving earth system governance. Science 335, 6183–6188 (2018). 1306–1307 (2012). 21. A. Aguiar, M. Chepeliev, E. Corong, D. van der Mensbrugghe, The global trade analysis project (GTAP) 44. R. W. Larson, S. Verma, How children and adolescents spend time across the world: Work, play, and data base: Version 11. J. Glob. Econ. Anal. 7, 1–27 (2023). developmental opportunities. Psychol. Bull. 125, 701–736 (1999). 22. J. Gershuny, O. Sullivan, What We Really Do All Day: Insights from the Centre for Time Use Research 45. M. Basner, A. M. Spaeth, D. F. Dinges, Sociodemographic characteristics and waking activities and (Penguin UK, 2019). their role in the timing and duration of sleep. Sleep 37, 1889–1906 (2014). 23. M. Manfroni, R. Velasco-­Fernández, L. Pérez-­Sánchez, S. G. Bukkens, M. Giampietro, The profile of 46. M. Basner et al., American time use survey: Sleep time and its relationship to waking activities. time allocation in the metabolic pattern of society: An internal biophysical limit to economic growth. Sleep 30, 1085–1095 (2007). Ecol. Econ. 190, 107183 (2021). 47. W. Fajzel et al., The global human Day. Zenodo. 10.5281/zenodo.7941615. Deposited 16 May 2023. PNAS 2023 Vol. 120 No. 25 e2219564120 https://doi.org/10.1073/pnas.2219564120 9 of 9