agreement, for these divergences are a very important part of the picture. They do come out in the concluding comments by members of the committee, and these should be taken not as a sign of failure but as a sign of the immense difficulty and complexity of the problem and as pointers toward further work.
I would raise one or two questions about the general assumptions underlying the study that might lead to misleading conclusions if they were not brought up. One is the assumption implicit in many of the models that the GNP of the United States will grow by at least 2 percent over the next few decades, though there is some recognition that this rate of growth will eventually slow down. It seems to me at least possible that rates of economic growth will be much less than this. It is very dangerous to extrapolate the rates of growth for a very complex aggregate like the GNP.
Two factors suggest that the GNP of the United States may have a much slower rate of growth in the future than it has had in the past few decades. One is that as technological development proceeds, those industries that are subject to productivity increase usually decline in relative importance as the economy moves into industries and occupations where the increase in productivity is very difficult, like education, the arts, medicine, and government. A good deal of the increase in GNP in the last 30 years came out of the extraordinary increase in productivity in agriculture, which released some 30 million people from agriculture to produce other things. This process is now reaching its conclusion, even in agriculture, and agriculture is now such a small proportion of the labor force that even a substantial increase in productivity would not release very many people. Productivity in manufacturing was not rising very rapidly even in the last 30 years, and has now virtually slowed to a stop. Increase in knowledge and improvements in technology will, of course, continue to take place and will in part offset the increasing costs of energy and materials. But whether this offset will be sufficient to prevent virtually stationary or even declining GNP per capita is a real question. If we add to this the change in the political climate in the last 10 or 15 years, which has imposed great uncertainties on private enterprise and has created an increasing unwillingness to take risks, particularly those imposed by somewhat uncertain regulations, the chances of actual decline in GNP per capita seem much greater than the economics profession is willing to recognize.
It may be, of course, that this gloomy prophecy will be falsified as many similar prophecies have been in the past by unexpected advances in knowledge and technology. There are indeed a few possible technical developments that would transform the whole energy and economic picture of the world in a relatively short time. The development of a cheap and portable battery for storing large quantities of electricity might