Discretionary time casts new light upon time as a candidate for the “appropriate currency of egalitarian justice”. Time – the authors claim – has “some very special properties” that make its candidature particularly palatable: it is inherently egalitarian, it is inherently scarce and it is a necessary input to any human activity. Whoever is interested in egalitarian justice, then, should also be interested in making time the equalisandum.

Goodin et al. are not the first advocates of time, the tradition going back – as the authors themselves admit – at least to Marxian economics. The originality of Discretionary time must then lay elsewhere. Traditional supporters propose to equalize the proportion of the day as a discretionary resource over various classes of people distinguished, for example, by some pre-determined social characteristic. That, Goodin et al. claim, amounts to equalizing spare time – i.e. time spent in leisurely activities. Translated in the lexicon of egalitarian language, traditional views aim at granting “equality of outcomes” in the distribution of time among people. Discretionary time, on the contrary, defends a sort of “equality of opportunity” time-egalitarianism where preoccupation is with giving “the just distribution of control over the resource of ‘time'“ (p. 4). The relevant issue is to make sure that any individual has “fewer constraints and more choices in how she can choose to spend her time” (p. 4). Temporal autonomy (or “autonomous control” over one's own time), measured in terms of discretionary time, should be the equalisandum. Taking into account certain social standards that determine how much time a person has to spend in six OECD countries – Australia, the USA, France, Germany, Sweden and Finland – to satisfy his basic needs, Goodin et al. are able to determine the algorithm that equalizes discretionary time (i.e. that grants the same autonomous control over discretionary time) and the policies that each country's welfare system should implement to achieve such a result. The authors argue for their thesis on the basis of a carefully blended mixture of analytical considerations and statistical research; a blend which should be a model for whoever wishes to take social policy analysis seriously.

The book articulates its argument over roughly 450 pages, the last third of which lists a remarkable sequence of tables produced to illustrate the main conclusions and policy prescriptions. Its substantial length and technicalities go unnoticed, partly for the easily accessible language, partly for the neat and rational organization of the argument's frame, divided in six parts, including a three-chapter introduction and a conclusion. The introduction lays down the main aims and characteristics of the analysis (first chapter), the main concepts (chapter two) and the main empirical findings (chapter three). As in Macbeth, where we wait for the murderer to realize that his hands are stained with blood, Discretionary time introduces its results since the very inception and yet keeps its attractiveness as we observe the four subsequent parts unveiling how temporal autonomy may be translated into the metric of discretionary time.

Special attention is devoted in the second part of the book – Time pressure – to peoples’ exacting commitments over their time. In a time outcome-egalitarian philosophy, time pressure measures people's consumption of their time. The more intense their commitments are, the stronger the time pressure and the lesser their spare time. Quite contrarily, in a resource-egalitarian philosophy, time pressure is related to how people need to allocate their time. Once requirements imposed by necessary activities are factored out, time pressure may well become illusory since it results from people's choices beyond what is necessary, choices that can hardly be the subject matter of resource-egalitarianism. The peculiar role that the second part serves for the overall structure of the book becomes apparent. By focusing on time pressure, it illustrates how Goodin et al.'s measurement of discretionary time departs from time outcome-egalitarian philosophies, how such a departure affects policy prescriptions and how it lays down a methodological exercise quintessential of the book's proposal and replicated in the subsequent pages. The three parts which follow – Welfare regimes matter, Gender regimes matter and Household regimes matter – constitute the public policy backbone of the entire book. There the authors apply their measure of temporal autonomy to three different “social regimes” leading to 18 (recall they are considering six countries) case-studies that provide a wide and attractive empirical articulation for their theoretical framework.

Discretionary time is a measure of autonomously controlled time. Any discussion must take as its starting point how it is actually calculated and the nature of the statistical information used by the algorithms. Consider a week as the unit of measure (or 168 hours): discretionary time is the number of hours left once the bare necessities imposed by 1) earning poverty line income, 2) satisfying – by unpaid household labour – the “core domestic work” and 3) ensuring minimal personal care are accounted for. So, for example, let us take a dual-earners, childless, American family. On average each member spends 11.89 hours to earn a poverty line income, 7.13 hours in unpaid household labour and 55.16 hours in personal care. According to the authors’ algorithm then, his or her discretionary time is given by 168 hours minus 74.18 – i.e. the time spent on necessary activities – or 93.82 hours.

First, note that the measure combines time and money information. The data derive from two surveys, the Multinational Time Use Study – that delivers the temporal information – and the Luxembourg Income Study – that provides the monetary information. The statistical procedure nests the temporal variables derived at the individual level from the first dataset into the income information provided by the LIS dataset. The combined dataset uses monetary units to measure and implement time-egalitarianism. This is interesting because the authors’ support for time's candidature for the post of the appropriate currency of egalitarian justice is then a qualified one. Though interesting, yet it is unsurprising since public policies cannot mint any extra time for their beneficiaries: the length of the day is fixed and state intervention powerless. Public policies may touch, on the contrary, people's pockets by introducing, for example, a minimum salary or child-care subsidies that alter, eventually, the amount of time required to satisfy bare necessities.

Once the public purse's role is (re)established, one would expect redistribution to increase discretionary time. Quite surprisingly, Goodin et al. show that “the net effect of government taxes, transfers and child-care subsidies is to reduce temporal autonomy slightly, averaging across the population as a whole” (p. 149). The oddity of this result is noteworthy and, I fear, problematic for the time egalitarian project; it is unfortunate that the reasons are not explored in full. Utilitarian outcome egalitarian philosophies would recommend transfers from the rich to the poor since the latter's greater marginal utility of money ensures an increase in utility for society as a whole. Time egalitarianism cannot rely upon the value of preference satisfaction at the margin but reasons on the basis of a similar compensation logic. If in a two-persons – one rich and one poor – hypothetical society, a tax-funded subsidy in favour of the poor is introduced which reduces the rich's discretionary time for the full amount of the tax, it might not benefit – in temporal terms – as much the poor if he is sufficiently far away from covering necessary activities, leading to the book's forecasted overall decrease in temporal autonomy. Of course, things are more complicated in real life (taxes are not related one-to-one to discretionary time, in n-persons societies income is unevenly distributed, taxation is progressive, etc.), but the basic logic would not change. A perverse logic in an egalitarian perspective since it would establish – as the authors perceptively note – a welfare state for the benefit of the politically cared for only.

One favourable consequence of such an odd result is the book's advocacy for a more subtle array of public policies than blunt redistribution. Such a conclusion is reinforced if one looks at the sophisticated articulation of the effects of state intervention with respect to gender regimes (chapter 12) and at the consequences of household regimes (chapter 16) where the authors show that the formal and informal rules that govern family's birth, life and dissolution matter. So, institutions matter.

The difficulties with an egalitarian project based upon discretionary time are far more general than the issue of redistribution would suggest: they involve time's performance as the currency of egalitarian justice. As any resource-egalitarian philosophy would recommend, policy makers should “strive to equalize people's ‘necessary time’ in the various necessary tasks of daily life, regardless of their social circumstances [. . .]. How people then actually choose to exercise their discretion and spend that time should be left up to them” (p. 112). In other words, Discretionary time wishes to determine quantitatively and equalize the time required by necessary activities, computed using objective information, and leave people free to allocate their time to activities for which they shall then be held responsible. It follows that “social policy should be interventionist in egalitarian directions with respect to opportunity (discretionary time); but social policy should be impervious to differences in the actual amount of ‘spare time’ people have left over” (p. 112). These two intimately connected conclusions are defended by the authors as germane to resource egalitarianism since necessary activities are construed as independent of people's responsibility; and as consistent with liberalism's value neutrality with respect to differences in people's use of discretionary time.

The point is that egalitarianism has its raison d'être in the moral worth of compensating people for the circumstances of their situations for which they do not bear responsibility. As an informational basis, time is too coarse to implement a genuinely egalitarian policy. Consider the case of gender regimes. We want to equalize discretionary time. Redistributive policies lead – according to Goodin et al.'s statistics – to a reduction of discretionary time for childless women and to an increase of women's discretionary time, if they have children. So, if I want to equalize the time resource, I should tax childless women only if they have no responsibility for their circumstances. If they do have responsibility, then taxing them is unfair under the general perspective offered by time egalitarianism since their circumstances do not deserve compensation. How odd a result for an egalitarian philosophy.

The oddity is certainly attributable to the peculiar consequences of redistribution upon discretionary time. But I wish to suggest that it also involves the inadequacy of time as a foundation of egalitarianism. The relationship between the equalisandum and the policy variable used to implement the egalitarian prescription would require far more thoughts than space permits in this brief note. But, it looks like that while hard line resource egalitarianism equalizes resources by redistributing resources, time resource egalitarianism equalizes discretionary time by redistributing resources since time is impermeable to public policy. I wonder whether such a discrepancy between aims and tools is at the root of the difficulty under examination.

The final issue I want to tackle concerns Goodin et al.'s interpretation of autonomy as capacity and, in general, the sense in which discretionary time may be considered a measure of freedom, as the authors suggest even in the book's title. In presenting discretionary time's philosophical credentials a distinction is made between two dimensions in the interpretation of autonomy. The first one, autonomy as capacity to form one's own principles, is a sort of necessary (“lexically prior” – as the authors say) condition for being autonomous. If taken by itself, such a condition would make people autonomous “only in a minimal way”. The second one, the capacity to act upon one's own principles, is the dimension of autonomy that Discretionary time wishes to measure and equalize: “[h]ow autonomous any given person is, once over the threshold of having adequate capacity to form principles of his or her own, depends just on how much capacity he or she has to act on principles of his or her choosing” (p. 30). The main goal of the book is then control over time a decision maker possesses.

I must confess some uneasiness. Goodin et al.'s interpretation of autonomy as control can hardly be captured by information about control over time. In the most favourable case – which, however, is not the one supported by the book – control over time is a precondition of temporal autonomy, alike the first dimension of autonomy. Suppose I equalize the necessary time over the three domains – satisfying bodily, financial and household necessities – the authors deem as relevant for a given group of individuals. People then enjoy freedom in the sense that they have the possibility of choosing how to employ the spare time that is now available. At best, if one happily subscribes to the positive interpretation of freedom, people then have opportunity for choice. But to claim that they also enjoy autonomy – i.e. capacity to act on principles of their choosing – seems to me a logical stretch that is not supported by the information provided by Goodin et al.'s measure of freedom.

Control requires more than mere availability of resources – whatever they would be: time, money or opportunities. It requires a liaison between the available resources and the use to which a decision maker puts them. It calls for a psychological process of decision making that allows the individual who has formed her own principles to act on their basis. This can only be measured by observations that reflect the underlying psychological process that leads to a given choice; observations that cannot be provided by exclusive reliance upon discretionary time.