On Testing the Efficacy of Selective Credit Controls: Comment
评论了Hamburger和Zwick以及Maris对选择性信贷控制有效性的实证检验，指出他们的方法不适合检验该问题，并分析了Mishkin模型中债务与耐用消费品支出的关系。
In recent notes in this journal, Hamburger and Zwick [2] and Maris [3] presented empirical tests that were designed to determine the efficacy of selective credit controls on on consumer durables. The purpose of this comment is to argue that neither one of these tests is well suited to address the issue in question. Hamburger and Zwick estimate a version of the model of consumer durable expenditure developed by Mishkin [4] and argue that theirvresults suggest that i'selective credit policies are unlikely to have a significant effect on the composition of real expenditures [2, p. 107]. A close look at Mishkin's theoretical model indicates that the Hamburger and Zwick adaptation is inappropriate for testing the efficacy of selective credit policies. The empirical results obtained by Hamburger and Zwick are the only ones that are consistent with the theoretical model and should not be interpreted as evidence on selective credit policies. Mishkin specifies consumer durables expenditure as a function of transitory income, permanent income, the lagged stock of durable goods, total household debt, gross financial asset holdings of households,l and the user rental cost of consumer durable capital. The rationale for including debt and financial asset holdings as separate arguments in the expenditure function is that consumer durables are illiquid and the expected cost of holding such an illiquid asset is positively related to the probability of having to make a distress sale. Since a distress sale will take place whenever consumption plus debt service exceeds income plus financial assets, the probability of such a sale and the expected opportunity cost of holding consumer durables will be positively related to debt service and negatively related to financial asset holdings. A ceteris paribus rise in debt holdings will raise debt service and thereby the opportunity cost of holding consumer durables. This will tend to decrease expenditure on consumer durables. What should be stressed here is that the single linkage between household debt and expenditure is debt service.2 The Mishkin equation is reestimated by Hamburger and Zwick with two rather than one household debt explanatory variables. Total debt is decomposed into