| Abstract: | Abstract: Using confidential data from US manufacturing firms' tax returns and Inland Revenue Service (IRS) audit adjustments, Mills (1998) tests, and finds support for, her hypothesis that IRS audit adjustments increase as the book-tax differences increase. We test Mills' hypothesis using confidential data obtained from the New Zealand Inland Revenue (hereafter Inland Revenue). Confidential data provide the key variable of interest, Inland Revenue's proposed audit adjustment, which is not available from public sources. These data provide the exact audit adjustment amounts, eliminating measurement errors inherent in proxy variables, and enable a temporal alignment of the book-tax differences with the Inland Revenue audit adjustments, thereby enhancing the internal validity of the relation between book-tax differences and Inland Revenue audit adjustments. Because the results of our study using New Zealand data, another time period, a more diverse set of firms, and a different institutional environment are consistent with those of Mills, we argue for the generalizability of Mills' hypothesis that proposed audit adjustments are positively related to the excess of book income over taxable income. |
