Determinants of aggregate income-tax-evasion behavior: The case of the US

Banca Nazionale del Lavoro Quarterly Review, Sep 1998 by Cebula, Richard J

where terms in parentheses are t-values and ' is the first-differences operator. In equation (10), * indicates statistically significant at the 5% level and FS indicates statistically significant at the 1% level. The Fstatistic is significant at the 1% level. In estimate (10), three of the coefficients are statistically significant with the hypothesized signs at the 1% level, and one is significant with the expected sign at the 5% level.

In equation (10), the estimated coefficient on the AEPIT variable is positive and significant at the 1% level, while the coefficient on the AESST variable is positive and significant at the 5% level. Hence, it appears that the higher the average effective federal personal income tax rate, the larger the relative size of the underground economy. Furthermore, the higher the average effective social security tax rate, the larger the relative size of the underground economy. These findings are consistent with the study of data from audits of individual tax returns by Clotfelter (1983), who finds underreporting of income to be an increasing function of marginal income tax rates. These results are also consistent with the findings based on `official data' in Slemrod (1985) and Pommerehne and Weck-Hannemann (1989), as well as the findings based upon experimentation in Baldry (1987), Alm, Jackson and McKee (1992), and Benjamini and Maital (1985). Moreover, this finding is also consistent with the regression estimate in Feige (1994, p. 135, n. 19), where the size of the underground economy is regressed in levels against a lagged tax variable (and a lagged second variable, D, which corresponds to the variable DIS in the present study).

Regarding the corporation income tax rate variable, the coefficient fails to be either positive or statistically significant at even the 10% level. Thus, the evidence in estimation equation (10) implies that this variable does not have a significant effect on the relative magnitude of the underground economy, i.e., on aggregate income-tax-evasion behavior. This finding may to some extent reflect the fact that, in the US, many corporations (particularly the large ones) are publicly owned, so that they are subject to a variety of disclosure requirements and extensive public scrutiny. Such scrutiny would make underreporting income very difficult. In addition, officers in publicly-owned corporations tend to have incentives to report income fully in order to provide a record of good performance to their stockholders.

Whereas the coefficient on the AUDIT variable in equation (10) is negative (as expected), it is not significant at even the 10% level. By contrast, the coefficient on the PEN variable in equation (10) is negative and significant at the 1% level. Thus, as tax evasion theory predicts (Pierre, Possen and Slutsky 1994), the greater the penalty from underreporting AGI, as measured in this study by variable PEN, the smaller the relative size of the underground economy because the greater PEN, the greater the expected IRS penalty if unreported income is detected.