Much of classical political economy was concerned with the question of how to achieve the “good life” for the greatest possible number. Utopian writers like Karl Marx decried the “idiocy” of rural life and the soulless drudgery of the industrial era . His vision of the good life was one in which an individual could “hunt in the morning, fish in the afternoon, rear cattle in the evening, criticize after dinner” (Marx and Engels 1977, p.169). The idea that a rich life is the ultimate concern of economics is implicit also in the capabilities approach of\nSen. “Development as Freedom” (Sen 1999) can be seen as one long argument to the effect that extending human capabilities is both a prerequisite for and constitutive of development. Extending “freedom” is about expanding the capacities of individuals to act across a broad front of human endeavour. Increased consumption may be necessary for this, but it is not sufficient. By contrast an orthodox microeconomist would view well-being in the first instance through the lens of utility functions. In these utility is derived from consumption, i.e. utility functions typically take consumption bundles as their arguments. Increasing consumption is the path to increased utility and hence well-being. Many utility functions also implicitly assume that consumers value variety in consumption. For instance the Cobb-Douglas utility function has a value of zero unless the individual consumes something of every type of good on offer. With a CES utility function u (x1, . . . , xn) = x? 1 + . . . + x? n it is possible to get a positive utility without consuming all products only if ? > 0. Even in that case, if products are infinitely divisible then at the optimum the agent will consume something of all goods available. It is only in the presence of indivisibilities and nonlinearities in the budget constraint that there will be systematic variation in the number of items consumed by different agents. Indeed showing empirically that consumers do value variety has proved somewhat tricky (Gronau and Hamermesh 2001). A connection between the variety of activities undertaken and the variety of commodities consumed has\nbeen made in a recent paper by Gronau and Hamermesh (2001). Their starting point is the home production literature and a Beckerian time allocation model (Becker 1965). They make the point that it may be easier to analyse the “demand for variety” in activities rather than commodities because of the inherent nonlinearities in time budget constraints. In particular they argue that time costs in switching from one activity to the next will in a straightforward way induce systematic patterns of time allocation. They proceed to show that there is a significant education gradient, with more educated people packing more non-work activities into the average day. They suggest that this is because more skilled people are more efficient at switching between activities. We will discuss their paper and their results in more detail below. Our paper makes three contributions. Firstly we show that the Gronau and Hamermesh results hold up in the context of a developing country, namely South Africa. Educated South Africans engage in more non-work activities than their less educated compatriots. This is despite the fact that they also spend much more time each day working. In short, they manage to cram more non-work activities into a shorter time-span. Secondly we argue that access to certain infrastructures (in particular electricity) has the effect of increasing the variety of activities and, arguably, consumption. A simple explanation for this is that some “capacities” are foundational, in the sense that they enable certain kinds of activities and consumption. This may provide yet another reason why people with higher education seem to lead fuller lives. This argument resonates well with Sen’s approach, since it suggests that development consists in expanding these capacities. Arguably they are also necessary for the eradication of certain “unfreedoms”. Finally we suggest that the results presented by Gronau and Hamermesh and the South African figures can be put into the same continuum. It appears as though variety in a cross-country sense increases with GDP. This suggests that Marx may be both right and wrong — right in the presupposition that modern industrial societies would have richer possibilities than the agrarian past, but wrong in the assumption that much of the time would be given over to work-related drudgery.