432 International Organization resultant "crowding out"of private investment dampens the expansion. However,if capital moves freely across borders,bonds floated to finance increased government spending are bought by international investors,and there is no effect on interest rates,which are set globally.Relaxing the fixed exchange rate constraint has different effects with fiscal policy than with monetary policy.If the exchange rate varies,as foreigners buy more govern- ment bonds the resultant capital inflow causes a currency appreciation that tends to reduce domestic demand for domestically produced goods and thus to dampen the fiscal expansion.8 The general point is that in a world of fully mobile capital,national policy cannot affect the national interest rate;it can,however,affect the exchange rate.The above discussion of open-economy macroeconomics is meant simply to highlight this result. It may seem unimportant that the world has changed from one in which national macroeconomic policy operated primarily via interest rates to one in which policy operates primarily via exchange rates,but several points to defend the significance of this observation can be made.First,the distributional effects of interest rate changes are different from those of exchange rate changes.If monetary expansion in a stylized world before capital mobility (BCM)meant lower interest rates,then monetary expansion in a stylized world after capital mobility (ACM)means depreciation.To take one distributional example, lower interest rates are good for the residential construction industry,while depreciation is bad for it inasmuch as it tends to switch domestic demand away from nontradable goods and services.By the same token,manufacturers might have been more sympathetic to a tight money stance in the past,when the principal effect of this stance was to raise interest rates,than they are now, when the principal effect is a currency appreciation that tends to increase import penetration.This means that policy preferences of economic interest groups,and therefore political coalitions,are likely to differ between the BCM and ACM worlds.I later return to this point and discuss it in detail. Second,although by definition there is an international component to exchange rate changes,there is not necessarily an international component to 17.The point is not that foreigners buy all the government bonds but,rather,that the increased domestic demand for credit is met by an increased supply of credit as capital flows in,with the result that the price of credit remains unchanged.This of course assumes that the deficit country is not large enough to affect world interest rates,which may not always be the case.It also assumes that the government does not engage in monetary policies that accommodate the fiscal expansion. 18.For a good survey and evaluation,see Michael M.Hutchison and Charles A.Pigott,"Real and Financial Linkages in the Macroeconomic Response to Budget Deficits:An Empirical Investigation,"in Sven Arndt and J.David Richardson,eds.,Real-Financial Linkages Among Open Economies (Cambridge,Mass.:MIT Press,1987),pp.139-66. 19.More accurately,it does not affect the covered interest rate-that is,the interest rate minus (or plus)the market's expectation of currency movements.Obviously,if investors expect a currency to fall,they demand a higher interest rate for securities denominated in it,and vice versa.Covered interest parity appears to have held well from the mid-1970s onward among almost all major currencies.432 International Organization resultant "crowding out" of private investment dampens the expansion. However, if capital moves freely across borders, bonds floated to finance increased government spending are bought by international investors, and there is no effect on interest rates, which are set globally.17 Relaxing the fixed exchange rate constraint has different effects with fiscal policy than with monetary policy. If the exchange rate varies, as foreigners buy more govern￾ment bonds the resultant capital inflow causes a currency appreciation that tends to reduce domestic demand for domestically produced goods and thus to dampen the fiscal expan~ion.'~ The general point is that in a world of fully mobile capital, national policy cannot affect the national interest rate;19 it can, however, affect the exchange rate. The above discussion of open-economy macroeconomics is meant simply to highlight this result. It may seem unimportant that the world has changed from one in which national macroeconomic policy operated primarily via interest rates to one in which policy operates primarily via exchange rates, but several points to defend the significance of this observation can be made. First, the distributional effects of interest rate changes are different from those of exchange rate changes. If monetary expansion in a stylized world before capital mobility (BCM) meant lower interest rates, then monetary expansion in a stylized world after capital mobility (ACM) means depreciation. To take one distributional example, lower interest rates are good for the residential construction industry, while depreciation is bad for it inasmuch as it tends to switch domestic demand away from nontradable goods and services. By the same token, manufacturers might have been more sympathetic to a tight money stance in the past, when the principal effect of this stance was to raise interest rates, than they are now, when the principal effect is a currency appreciation that tends to increase import penetration. This means that policy preferences of economic interest groups, and therefore political coalitions, are likely to differ between the BCM and ACM worlds. I later return to this point and discuss it in detail. Second, although by definition there is an international component to exchange rate changes, there is not necessarily an international component to 17. The point is not that foreigners buy all the government bonds but, rather, that the increased domestic demand for credit is met by an increased supply of credit as capital flows in, with the result that the price of credit remains unchanged. This of course assumes that the deficit country is not large enough to affect world interest rates, which may not always be the case. It also assumes that the government does not engage in monetary policies that accommodate the fiscal expansion. 18. For a good survey and evaluation, see Michael M. Hutchison and Charles A. Pigott, "Real and Financial Linkages in the Macroeconomic Response to Budget Deficits: An Empirical Investigation," in Sven Arndt and J. David Richardson, eds., Real-Financial Linkages Among Open Economies (Cambridge, Mass.: MIT Press, 1987), pp. 139-66. 19. More accurately, it does not affect the covered interest rate-that is, the interest rate minus (or plus) the market's expectation of currency movements. Obviously, if investors expect a currency to fall, they demand a higher interest rate for securities denominated in it, and vice versa. Covered interest parity appears to have held well from the mid-1970s onward among almost all major currencies