Construct Instrumental Variables Research Articles

Wage Shocks and Consumption Variability in Mexico During the 1990s

This paper presents evidence on the relationship between economic shocks to relative male wages and changes in household consumption in Mexico during the 1990s, which is a period characterized by high volatility. In addition to performing this type of analysis for Mexico for the first time, the paper makes two main contributions. The first is the use of alternative data sources to construct instrumental variables for wages. The second is to examine differences across four consumption categories: non-durable goods, durable goods, education and health. Our results for non-durable goods consumption reject the hypothesis that Mexican households are able to insure idiosyncratic risk. For the comparisons across consumption categories, the conclusion is that households in Mexico tend to react to temporary shocks by contracting the consumption of goods that represent longer-run investment in human capital, which makes them more vulnerable in the future.

The effect of marginal tax rates on income: a panel study of ‘bracket creep’

This paper uses a panel of individual tax returns and the ‘bracket creep’ as source of tax rate variation to construct instrumental variables estimates of the sensitivity of income to changes in tax rates. From 1979 to 1981, the US income tax schedule was fixed in nominal terms while inflation was high (around 10%). This produced a real change in tax rate schedules. Taxpayers near the top-end of a tax bracket were more likely to creep to a higher bracket and thus experience a rise in marginal rates the following year than the other taxpayers. Compensated elasticities can be estimated by comparing the differences in changes in income between taxpayers close to the top-end of a tax bracket to the other taxpayers. These estimates, based on comparisons between very similar groups, are robust to underlying changes in the income distribution, such as a rise in inequality. The elasticities of taxable income and adjusted gross income are around 0.4 and significant but the elasticities of wage income are in general insignificant and close to zero.

Consistent parameter estimation of systems disturbed by correlated noise

The paper presents a new bias-compensating least-squares (BCLS) method which can eliminate the noise induced estimation bias even without modelling the noise. By employing a composited signal in the parameter estimation, two estimates for the same part of the system parameter are obtained so that the noise induced bias can be extracted from the difference between these two estimates. With the bias removed, the consistent parameter estimate is obtained. It is shown in the paper that the BCLS method is a member of the class of weighted-instrumental-variable methods. Therefore, the BCLS method also provides a particular way to construct instrumental variables for the identification of a general system.