Knowledge that Transforms

Jerome Stein's article in a recent issue of this Review is subject to misinterpretation which could lead to seriously misleading results. He makes several arguments against the Council's conclusion that Many, though not all, pollution problems are local in character, and therefore determination of the appropriate level of environmental quality in these cases is likely to be more accurate if it is done locally rather than by the Federal Government . (quoted by Stein, p. 532). One of his arguments is that local pricing (effluent charges) would result in inefficiency because the marginal product of pollution would then differ from region to region. Possible confusion results from the fact that two separate measures of pollution are not consistently distinguished. The first is a physical measure like pounds of biochemical oxygen demand or tons of sulfur oxide discharge. The other is the dollar value of damage. In Stein's Figure 1, the horizontal axis represents pollution damage and what is demonstrated is that a dollar's worth of damage must be charged for at the rate of a dollar if efficiency is to prevail. This is emphatically not the same as saying that every pound of discharge should be charged for at the same rate, no matter where in the nation it occurs. But the latter is a conclusion that could easily be inferred from Stein's discussion. instance, Stein says, For example, if the firms were ordered to reduce the rate of pollution by Ax, . (p. 533) where the correct reference is to pollution damage. A similar sliding over from pollution emission to pollution damage occurs again in the first full paragraph of page 535. The translation of pounds of discharge into dollars of damage depends, among other things, on meteorological and hydrological conditions and on the presence and preferences of receptors. Charging a dollar for a dollar's worth of damage requires quite different rates of charge per pound of discharge depending upon the locality. A uniform national charge might be defended on other grounds (see Kneese) but not as a requirement for efficiency strictly speaking. Charges which reflect local conditions could in principle be set locally or nationally. If the Council's argument is right, and I am not trying to assess it here, that the optimum level of environmental quality is more likely to be accurately determined locally, then, problems of implementation aside, more efficient charges could be set locally.

This paper is a review of an attempt by V. K. Chetty to nmeasure the relative amount of monetary services rendered by nearmoney assets such as time deposits at commercial banks, deposits at mutual savings banks, and savings and loan association shares. If such a measure could be found, then one could use it to construct a better money supply total; that is, one which would more accuratelv measure the total amount of monetary services available in the economy at any point of time. For example, if it could be established that a dollar of savings and loan shares (SL) rendered the same amount of monetary services as fifty cents of money (M) then by adding 50 percent of the value of outstanding SL to the conventional money supply, we would arrive at an adjusted money supply (Ma) which would take into account the monetary services of SL as well as M. The Ma would then measure the amount of M alone it would take to provide the same monetary services as the actual combination of M and SL in existence. Interest in constructing an Ma along these lines has been expressed by several writers.1 Their interest stems from the belief that there exists a stable relationship between the level of income, properly defined, and the desired level of monetary services. If this is the case, then a more accurate measure of the total amount of monetary services available in the economy would enable one to more accurately predict the level of income. Chetty's attempt to construct an Ma will be reviewed in two sections. In Section I, the theory that Chetty uses to construct Ma is examined, and it is shown that Chetty's theoretical presentation contains an error. Moreover, even if the theory is corrected along the lines suggested by Chetty in a footnote, it would still be impossible to estimate the monetary services of near-money assets since these cannot be separated from the other nonmonetary services rendered by these assets. In Section II, Chetty's empirical results are examined. Because Chetty cannot separate the monetary from the nonmonetary services of near-money assets, he is led to assign weights to near-money assets which are greater than one, implying that these assets yield a larger amount of monetary services per dollar than does money itself.

In their article in this Review, W. Lee Hansen, Burton Weisbrod, and William Scanlon (HWS) conclude that estimated payoff to more schooling is for their sample of men who were rejected for military service due to low scores on the Armed Forces Qualification Test (AFQT). It is shown in this comment that they obtained a downward biased estimate of the true profitability of schooling for their sample and that the correct rate of return may be at least 13 percent. While this is lower than the rate of return to males from high school, it is similar to the private return from college.' The HWS data are a sample taken in November 1963 of approximately 2,400 males, aged 17 to 25. The dependent variable (Y) is annual income in 1962 after deducting transfer payments. The explanatory variables include years of schooling (S), age (A), AFQT score (AFQT), and a dummy variable which takes the value of one if the individual received training outside of school (T). Their model is