We try to build a macroeconomic index, that includes some non-market variables, to be compared to the traditional GDP. Over the last twenty years answers to the welfare accounting problem have been different. Economists have used dynamic optimisation to rigorously derive an index that can be used to evaluate small projects and their contribution to well being. National Accountants are trying to extend the System of National Accounts (SNA) in the form of satellite accounts by increasing the system boundary. There is also a number of other studies which cannot be included in either of the previous categories and that we may call indices of welfare. They are not rigorously founded and start from common sense ideas of what should and what should not be considered as determinants of well being. This type of indices, however, has received wide attention thanks to their immediate comparability with GDP and to their characteristic of emphasising the long run trend of "welfare" as compared to GDP. Our work falls in the third category. We start from the Index of Sustainable Economic Welfare (ISEW) built up by Cobb and Daly in 1989 and we reproduce it for Italy, though revisiting the methodology associated to the construction of some of the variables. The results show that up to the seventies the Italian GDP only slightly overestimates the growth of economic welfare, but since then it seems to have been misleading, at least as compared to our index. While GDP has continued to rise, economic welfare has been stagnating. The importance of the environmental variables included seems to be decreasing over time going from the 38% in 1980 to the 31% in 1990. The cost of pollution (air, water and noise) has a greater weight in the sixties (about 13%) and falls to 8% in 1990. The ratio between GDP and environmental degradation has remained constant over the decades under consideration, thus contradicting the idea that the demand for environmental quality increases with income.