Roy Davidson says that the term "wealth" remains undefined. The implication is that not much, if any, progress has been made in this realm. I have recently become interested in this subject and I wonder whether he is looking at the issue from the best vantage point. The question to be asked is why a definition of wealth is important. The usual reason has something to do with comparing one economy or society with another economy or society (or with the same one at a different time). But this is not the only question that a definition of wealth may help one answer. A concept of wealth can also help one evaluate arguments for or against government policies -- in particular, policies in favor of or against market intervention. It seems to me that the main consequence of the marginal utility revolution was that it provided a utilitarian basis for evaluating arguments favoring various government policies. It did this by defining wealth in terms of consumer utility. Economists wanted a way to judge the claims of special interests. The spokespersons for those interests claimed that this or that government policy is good (or bad) because it is in (or not in) the "public interest." The models built by the marginal utility theorists provided frameworks to evaluate such claims by defining the public interest as the interests of consumers in a market economy. This later became the foundation for modern cost-benefit analysis, stemming largely from the work of A.C. Pigou. There are, of course, very big problems in trying to implement cost-benefit analysis. There have also been major abuses of cost-benefit analysis. However, the concept of wealth developed by the marginal utility theorists is nevertheless useful, I would maintain. Consider an example of a market intervention -- the U.S. anti-dumping law which has recently been ruled to be against the WTO trading rules. No reasonable intellectual today -- and surely no good economist -- would argue that the anti-dumping law increases the wealth of the U.S. because in increases the benefits to U.S. steel companies and U.S. steel workers. They would point out that the increase in rents to people in the steel industry are offset by the decrease in rents to people on other industries. In the meantime consumers are harmed. Anyone who has learned cost-benefit analysis would recognize that such an argument is incomplete. True, costs and benefits, like wealth, cannot be measured. But that is not the point. The point is that arguments referring to how the "public interest" is served or how the peoples' wealth will be raised by some policy cannot be regarded as sound unless they employ the cost benefit framework of the type based on the work of the marginal utility theorists. We cannot define wealth precisely but we can tell what is necessary to make a sound utilitarian argument that wealth is higher or lower due to some government policy relating to market intervention. From this point of view, I would argue that there has been progress in the development of the concept of wealth in economics, even though the concept has not been, and probably never will be, defined well or clearly. Pat Gunning