The Economics of Multitasking

Multitasking is an inevitable part of everyday life. Perhaps it is most evident in the context of child care, where adults frequently look after children while also doing other things such as cooking or cleaning. Yet, far more broadly, most people in modern societies are engaged in simultaneous activities for a large fraction of their waking hours—the layperson’s definition of multitasking.

Before the 1960s, economic theory poorly addressed questions about the allocation of time.¹ However, in the 1960s and 1970s some economists worked on new ways to model time use. Becker (1965), in his well-known contribution “A Theory of the Allocation of Time,” argued that consumers maximize their utility by choosing commodities that are produced with market goods and time by a consumer facing both budget and time constraints. DeSerpa (1971) and Evans (1972) further attempted to improve time allocation models² by including extra constraints in their models; both DeSerpa’s and Evans’s models can be shown to be particular cases of Becker’s model by just redefining commodities in Becker’s model and keeping the new suggested constraints.

Time-use data show that people frequently multitask when performing household production activities, especially child care, and that this is especially true for women (see, e.g., Floro and Miles 2003; Kalenkoski and Foster 2008; Offer and Schneider 2011; and Zaiceva and Zimmerman 2011). It has been conjectured that this is because women have greater multitasking ability than men. Evidence from neuroscience (Weise et al. 2006) suggests the possibility of innate gender differences in cognitive functioning that may affect task performance. Preferences, social roles, and cultural constraints may also play a role in explaining differences (Booth 2009; Croson and Gneezy 2009; de Mel, McKenzie, and Woodruff 2009; and Gneezy, Leonard, and List 2009).

The transition from having no children to starting a family brings substantial changes in time use (Craig and Bittman 2008; Kotila, Schoppe-Sullivan, and Kamp Dush 2013). The increase in household workload is large and, although it varies somewhat across different social and policy contexts, it is present cross-nationally (Dribe and Stanfors 2009; Craig and Mullan 2010; Gauthier and DeGusti 2012). Parenthood also changes the composition of time use, adding requirements to perform not only child care but additional domestic work as well (Craig and Bittman 2008). Households meet this additional unpaid time demand in a variety of ways, including redirecting time to child care that nonparents can allocate to other things, such as leisure, sleep, and market work (Bianchi 2005). Other forms of adaptation include time compression—by doing more within the same period of time (Robinson and Godbey 1997).

The main benefit of multitasking is that it allows people to do more than one activity at the same time. However, people usually cannot give their full attention to two activities when they multitask (Just et al. 2001). Thus, multitasking may come at a cost. One part of this multitasking cost is that people may enjoy an activity less if they cannot give it their full attention. Talking to a friend while doing homework may not be as enjoyable as talking to a friend without anything else on the mind. Another important part of the multitasking cost is related to task completion time. When people divide their attention between two tasks, they tend to take longer time to complete each of the tasks, especially if a task is cognitively challenging (Rubinstein, Meyer, and Evans 2001). For example, two experimental studies find that students take a longer time to read a passage of text if they are sending and receiving instant messages while reading (Fox, Rosen, and Crawford 2009; Bowman et al. 2010).

Homework is an important component of the academic production function (Betts 1997; Aksoy and Link 2000; Eren and Henderson 2008, 2011) but often students are studying while doing another activity. Over the last decade, children have had increasing access to many types of media outlets that may distract their attention. The number of TVs and personal computers per household has grown; the growth of broadband access has increased computing potential on home computers, mobile phones, and other devices; and the options for devices on which to listen to music as well as the access to different types of music has expanded. Previous research has found that proximity to devices, such as placement of a computer near a TV, is a strong predictor of media multitasking (Foehr 2006). Using UK time-use data, Kenyon (2008) has shown that the Internet has changed how we use our time—both in the activities we choose and in the extent of our multitasking.

Individuals’ decisions about eating and food choice are complex. Eating is not just for nutrition and sustenance as there are individual preferences involved, as well as environmental, convenience, cultural, and other factors influencing decisions on what to eat, when to eat, and where to eat. Americans’ eating patterns have changed dramatically over the last few decades and, in particular, food away from home has grown as a meal or snack option. In addition, food has become ubiquitous—many food options are available in a variety of retail outlets; eating in a variety of venues such as at workplaces and in vehicles has become acceptable; and eating takes place at all times of day (Hamrick et al. 2011). In fact, a variety of food companies are now offering their products in packaging designed to fit in a vehicle’s cup holder (Perimeter Brand Packaging 2013). Chicken wings, mini cookies and crackers, and candy are sold in cups and pouches that fit in cup holders, facilitating on-the-go eating while driving. The number of fast-food outlets in the United States has grown tremendously,’ facilitating both in-restaurant and carryout options for consuming prepared food.

Time-use surveys are a primary source of data for studying parental investment in children—an important topic that has been researched extensively. Numerous studies have examined how much time parents spend in child care activities, by parental employment and marital status, as well as long-term trends in parental time spent in child care. In addition to these questions, time-use surveys are useful for measuring the aggregate amount of time devoted to child care activities. Much of this time, especially time spent looking after children, satisfies the third-person criteria for household production and should be included in nonmarket satellite accounts to the National Income and Product Accounts (NIPAs).¹ Thus, it is important to have accurate measures of the time spent in child care.

This chapter studies how recent changes in the organization of work, namely, the move toward multitasking, have affected the returns to work experience. In particular, I link two empirical observations about the returns to experience. First, Katz and Murphy (1992 ) showed that in the United States, while the returns to college have risen dramatically since the late 1970s, the returns to experience—the difference between the wages of older workers and those of younger workers at a point in time—for college graduates seemed to have been flat or even fallen. For high school graduates, the returns to experience increased from about 1976 to 1987. Autor, Katz, and Kearney (2008) update these results until 2005 and find that between 1987 and 2005, the returns to experience for college graduates did not change much, while the returns to experience for high school graduates rose through 1995 and then fell over the next 10 years. The second empirical observation that I study in this chapter is that for those entering the labor market in the late 1960s and early 1970s, wage growth over the first 10 years in the labor market was lower for college graduates than for high school graduates, while for those who entered in the late 1980s, wage growth increased over the first 10 years for college graduates so that it was almost as high as that of high school graduates (Aaronson 2001).

One major goal of this volume was to show that multitasldng can be incorporated into a standard economic household production model in which individuals choose to engage in multitasking in order to enhance their productivity and, hence, their utility or well-being. Chapters 2 and 3 presented complementary theoretical models that provide this economic motivation for multitasking and examined the implications of these models for time-allocation decisions. However, while these models take the household production literature a step forward by incorporating multitasking into existing theoretical models, they are limited in that they focus on household production only.