GOVERNMENT BONDS involves no shift in co or U2--must be the optimal pattern for an individual. It follows that constancy of the net pretax bequest for all members of generation I is the equilibrium solution. In this case,a marginal shift in B again has no effect on consumption patter The basic conclusion here is that the existence of taxes on intergenera tional transfers makes less likely an interior solution for these transfers but if these transfers are operative, even if at reduced levels, the margin effect of B on consumption plans-and, hence, on r--remains nil E. Bond Issue and Tax-Collection Costs Suppose now that the issue of government debt and the collection of taxes to finance this debt involve transaction costs. In particular, in the case where the principal is paid off by generation 2. sur that a net of B to generation I is now associated with a tax levy of(1+?)rB on young)generation 2 and a levy of(1 +y)B on(old) generation 2. That is, y amounts to a proportional transaction cost associated with government debt issue and tax collection. 9 For simplicity, suppose now that the inheritance tax rate is zero. Equation(10) again remains valid but equation(12) is now modified to +(1-r)-(1+y)B=4+(1-n)+(1-r)242,(15) Consider, again, the conjecture that, when B rises, co and, hence, (1-r)Ai -B remain fixed. From equation (15),y>0 implies a negative-wealth effect on generation 2, so that U2 would fall. Since this effect would be anticipated by generation l, it can be supposed in the normal case that A would actually rise by somewhat more than 1/(1-r)times B, so that co would fall. In general, y>0 implies that an increase in B amounts to an overall negative-wealth effect, which would choice of bequests by all other individuals; and (2)all individuals choose the lue for their bequests. It can also be shown that the solution that maintains the net bequest for all individuals is the unique equilibri the solution involves the assumption that each individual perceives the shift in the transfer term, tAi, associated with the average response of quests to the change in B. Alternatively, if indi change in wealth. The typical response iduals treated tAi as fixed, they would an increase in B as, effectively, a negat would be a reduction in c which would be associated with an increase in Ai by more than 1/(I-r) times the change in B. In the aggregate, there would be an increase in desired saving, A%+ A,, which would lead to a reduction in r and to an increase capital formation. In particular, if the shift in transfers associated with inheritance ta wealth rises with B e is associated with a decrease in other taxes, rathe there could be an offsetting reduction in transaction assumed to be positive, must be interpreted in this net