Sylvia Auer, Tobias Bornemann, and Eva Eberhartinger

This paper analyzes the implications of using fair value accounting to define taxable income. Although the inclusion of fair values in book income provides value-relevant information to users of financial statements, taxable income typically relies on reliable and verifiable measurements such as historical cost and lower of cost or market. Taxable income, therefore, usually excludes unrealized gains and (partly) unrealized losses, resulting in costly book-tax differences. Several countries, such as Germany or Hong Kong, have adopted fair value tax regimes for financial institutions, including any fair value changes of held-for-trading assets in both taxable and book income to eliminate these costly book-tax differences. Fair value tax regimes, however, can have significant effects on banks’ liquidity when unrealized gains are fully subjected to tax and unrealized losses are fully tax deductible.

We exploit cross-country variation in fair value tax regimes and intertemporal variation in bank tax rates to analyze the effects of fair value taxes on banks’ investment portfolios and risk taking. Our results suggest that banks’ investment portfolios are sensitive to fair value taxes. First, held-for-trading securities of banks subject to fair value taxes decrease by about 1.9 percentage points of total assets for a one-standard-deviation increase in the tax rate relative to banks not subject to fair value taxes. Correspondingly, securities not subject to fair value taxation increase by about 2.2 percentage points of total assets for a one-standard-deviation increase in the tax rate. Our results are more pronounced for three types of banks. First, savings and cooperative banks, as opposed to commercial and investment banks, are more sensitive to fair value taxation. Second, the negative association between the level of held-for-trading securities and fair value taxes is pronounced for banks that report under local GAAP for tax purposes as opposed to IFRS. Third, the results are pronounced for banks that are comparatively short in liquidity, suggesting that fair value taxes are relatively costly for banks with higher liquidity needs.

We also find that banks’ risk taking, measured as the three (five) year standard deviation of banks’ pretax return on assets and their pretax distance to insolvency, is positively associated with the tax rate when subject to fair value taxes. This effect can only be observed in savings and cooperative banks, which have lower risk tolerances than commercial and investment banks. Further, the effect is more pronounced for banks that report under local GAAP instead of IFRS. Our results provide some evidence that banks’ risk taking is sensitive to including unrealized gains and losses in taxable income and is in line with prior evidence in nonfinancial settings, albeit subject to banks’ business model.

Chelsea Rae Austin, Donna D. Bobek, Marcus M. Doxey, and Shane R. Stinson

This study investigates how the timing of tax payments on retirement savings affects spending decisions during retirement. Our findings are based on an experiment where, across four periods, participants spend funds either from a deferred-tax account or a currently taxed account. A deferred-tax account is like a traditional IRA, where taxes are paid when funds are withdrawn from the account. A currently taxed account is like a Roth IRA, where taxes were paid when the funds were contributed to the retirement account and withdrawals are tax-free. We find that deferred-tax account holders consume their savings faster than currently taxed account holders with equal after-tax spending power. Further, when the nominal balances of their accounts are equivalent, deferred-tax and currently taxed account holders spend nominally equivalent amounts on goods, despite deferred-tax account holders also needing to pay taxes on withdrawals. This finding suggests deferred-tax account holders under-adjust for taxes and, therefore, consume their wealth at a faster rate. These findings have important implications for financial advisors and tax policy makers. They also add to a growing body of research showing that tax timing affects individual decision making and that currently taxed, versus deferred-taxed, accounts often lead individuals to make better retirement planning and spending decisions.

Zhuoli Axelton, Kerry K. Inger, Mollie E. Mathis, and Abbie E. Sadler

Despite prior research largely identifying benefits of auditor-provided tax services (APTS), there is also evidence investors penalize companies for purchasing these services, suggesting there are also costs associated with APTS. We examine whether regulator scrutiny is an additional cost of APTS. Regulatory scrutiny is costly because firms must spend time and resources responding to regulator concerns and defending their disclosures or tax positions. Regulators have publicly questioned whether external auditors can maintain their independence while providing tax services to audit clients. The European Union banned the provision of tax advisory services for audit clients, and this ban played a major role in the ultimate failure of EY’s proposed split of audit and advisory services, suggesting the fees are materially important to the firm. Regulators’ perceptions of independence issues may increase their scrutiny of firms engaging in APTS.

We examine two types of regulator scrutiny: tax disclosure scrutiny (i.e., tax-related SEC comment letters) and tax authority scrutiny (i.e., relative greater releases of uncertain tax benefit reserves due to settlements than due to lapses in statutes of limitations). We use two measures of APTS (i.e., continuous and indicator) and run our analyses on three samples; an unmatched sample, a propensity score matching (PSM) sample, and a sample using a Heckman two-stage model (i.e., inverse Mills ratio (IMR)). We find evidence that APTS is positively associated with both tax disclosure scrutiny and tax authority scrutiny. In additional analysis, we find that the association between APTS and tax disclosure scrutiny is concentrated in firms where client importance to the auditor is high.

We make three primary contributions to the literature. First, we add to the literature on the provision of nonaudit services (NAS) by offering evidence that one cost of APTS is increased regulator scrutiny. Second, we contribute to the literature examining tax-related SEC comment letters by demonstrating that in addition to tax avoidance, purchasing APTS is also a significant determinant of tax-related SEC comment letters. Third, we contribute to the literature on tax authority scrutiny by documenting that engaging in APS is associated with more scrutiny from tax authorities. Finally, our study has practical implications for various stakeholders. As firms continue to grapple with where APTS fits in the larger picture of audit and advisory practices, our results suggest that regulators do notice the materiality of these fees.

Anne C. Ehinger, Joshua A. Lee, Bridget Stomberg, and Erin Towery

Although voluntary tax disclosures can benefit investors by providing more detailed information, they can also inform tax authorities about companies’ tax avoidance strategies. This study explores how IRS enforcement affects voluntary tax disclosures during quarterly earnings announcements and conference calls. We begin by providing new descriptive evidence on the tax topics discussed in voluntary tax disclosures. The most common topics relate to the effect of tax expense on reported income, tax rate projections, and comparisons of current period taxes to prior periods.

Next, we estimate the amount of income tax disclosures in quarterly earnings announcements and conference calls as a function of IRS enforcement and other factors. Consistent with managers having concerns about the potential costs of voluntarily disclosing tax information to tax authorities, we estimate a negative relation between IRS enforcement and voluntary tax disclosures that is concentrated in the subset of firms with high levels of tax avoidance. Finally, we investigate the potential benefits and costs of voluntary tax disclosure. We find that voluntary tax disclosures benefit analysts, but at the cost of more attention from the IRS and larger income tax settlements during income tax audits.

In sum, this study provides some of the first evidence on the income tax issues managers and analysts view as important. We also provide evidence that IRS enforcement is associated with less voluntary disclosure, which, in turn, is associated with increased analyst forecast errors. These results suggest that although IRS enforcement improves corporate disclosure in some contexts, it can also deter managers from voluntarily disclosing information to financial market participants.

Qintao Fan, David A. Guenther, and Kaishu Wu

We introduce corporate tax expense into the cash-flow-based capital asset pricing model of Lambert, Leuz, and Verrecchia (2007) and investigate the relation between tax expense and cost of equity capital. In our model, tax expense has both a variable component (a tax rate) and a fixed component that can be negative (a tax credit or subsidy). We show that a reduction in the fixed tax component will always decrease the cost of capital. However, a reduction in the tax rate, such as shifting the firm’s income into a low-tax country or state, will increase (decrease) the cost of equity capital when the fixed tax component is negative (positive).

We conduct empirical tests by first generating firm-specific estimates of the fixed tax component and tax rate using time-series data. We find that cost of equity capital is positively related to the fixed tax component, but the relationship between cost of capital and the tax rate is negative (insignificant) when the fixed tax component is negative (positive). We then conduct a cross-sectional test by separating our sample firms into terciles-based cash effective tax rates (Cash ETRs). We find a significant negative (positive) relation between cost of capital and Cash ETRs for the lowest (highest) Cash ETR tercile, for which the average fixed tax component is negative (positive).

Corporate income tax is an important and relatively exogenous cash outflow. Unlike other types of costs that firms incur, the fixed component of tax expense can be negative for many firms. Our findings help explain the previously inconclusive results on the relationship between tax avoidance and cost of equity capital and provide insights into how the cost of capital depends on primitive distribution properties of cash flows.

Cristi A. Gleason, Tyler S. Menzer, and Jaron H. Wilde

The Tax Cuts and Jobs Act (TCJA) of 2017 dramatically reshaped the tax landscape for companies and reduced incentives for multinationals to leave foreign earnings abroad. Although advocates argued the TCJA would spur economic growth, critics objected to it based on potential adverse effects and the unusually rapid progression of the bill through Congress, especially for such landmark legislation (The New York Times 2017; Hunt 2017; Wang 2017). We examine one specific portion of the law, the use of prospective financial reporting dates for the repatriation toll tax, which allowed some companies—such as Apple and Microsoft, which have non-December year-ends—a substantially longer time period to react to the provisions in the law and reduce the toll tax.

Consistent with firms with non-December fiscal year-ends having up to 11 months to reduce the “cash” balances in their foreign subsidiaries, we find that such firms exhibit significant decreases in TCJA “cash” balances. Using a variety of estimates, we find an average firm tax savings of between $18.3 and $28.7 million. We also find that the amount of time was important for firms’ tax savings. Firms with the longest amount of time saved between $45.1 and $69.7 million. Overall, our estimates suggest that this additional time allowed firms to save about $8.5 billion in taxes simply by reducing their cash balances. Our evidence is consistent with the idea that firms were able to exploit the future measurement dates of the TCJA provision to generate significant tax savings and should encourage policy makers to take the tradeoffs between prospective and retrospective measurement dates into consideration when drafting new laws.

2025