Moreover, economic analyses of sustainability examine consumption or sustainability at the highest level of aggregation—the level of average consumption today and in the future. This level of aggregation masks a number of important ways of disaggregating the complex ensemble of economic and environmental activities. It omits details such as the sectoral or asset breakdown (for example, the separate trajectories of reproducible capital, stocks of subsoil minerals and timber, the quality of air and water, the health of salt marshes, and the value of stocks of genetic material). It assumes that within a sector or asset class, substitutes (including technology) will replace high-priced goods and services. It also overlooks the distribution of consumption among different groups within a country or among countries. It does not distinguish among different future generations and focuses only on the present generation versus the generalized future.
In addition, most treatments of sustainability do not deal with issues of uncertainty or risk. Must a path be sustainable with absolute confidence, or on average, or 90 percent of the time? Would we prefer a certainty of nondeclining consumption over an alternative that involves a robust growth in living standards plus a tiny chance of a small decline? How would we feel about a promising technology that offers a 99.9 percent probability of sustainability and a 0.1 percent chance of extinction? A short journey down the road of stochasticity raises numerous unanswered questions about the concept of sustainability.
In limiting the present analysis to this highly generalized and aggregated version of sustainability, it is recognized that many worthwhile goals will be overlooked. An alternative view of sustainability, for example, might hold that "maintaining capital intact" should apply at a more disaggregated level than the entire asset base of an economy. This narrower perspective might hope to protect certain assets or flows or subsystems—such as a suite of species or a group of important ecosystems, or even "natural capital" more generally—so that future generations could enjoy them at today's levels. Such a perspective depends on ''sector-specific" and "use-specific" definitions of sustainability. Defining sustainability in this narrower sense is often useful as a guide to policy making or as a practical shorthand way of expressing certain desirable conservation goals, but it generally is too narrow and subjective to serve as a principle for constructing measures of national income. Moreover, if taken literally, the injunction to keep "natural capital intact" is probably infeasible because human activities inevitably cause the levels of some natural assets somewhere to decline.
Sector-specific or use-specific definitions of sustainability raise other practical problems. One issue is selection of the list of assets to be maintained. Which specific set of resources is to be maintained? Who selects