Essay: Effect of release of deferred tax valuation allowance on next period earnings

Section I: Introduction
This study examines the effects of a release of the deferred tax valuation allowance on next period earnings as well as its effect on regulatory scrutiny. Although prior literature by Dhaliwal et al. (2013) has examined the consequences of a release of the valuation allowance on the future earnings of loss firm-years, the literature does not provide evidence on the effect that a valuation allowance release has on profitable firm- years. The study of profitable firm-years is important as it provides insight into the information content of the valuation allowance account in years where pretax income is positive. Additionally, prior literature observes the effect that tax avoidance has on the chances of a firm receiving a Security and Exchange (SEC) issued tax comment letter and how it impacts audit fees, but does not study the effect the deferred tax valuation allowance has on these measures for regulatory scrutiny. I focus on the deferred tax valuation allowance because its measurement is based on private, forward-looking information held by management. Accounting Standards Codification (ASC) 740 mandates that a firm recognize a valuation allowance if it believes there is less than a 50% chance it will realize the benefits that arise from its deferred tax assets. ASC 740 emphasizes the use of private information by managers to establish the valuation allowance account and subsequent adjustments it.
Because the deferred tax valuation allowance contains private, forward-looking information, intuition would say it is a good indicator for future pretax profitability. Despite this, prior literature by Frank and Rego (2006) suggests that managers use the deferred tax valuation allowance and subsequent adjustments to the account to smooth earnings over time, reducing the information contained in the account for investors. I
want to study how pertinent the valuation allowance is to future earnings, specifically in the period following a release of the allowance. I hypothesize that a release of the valuation allowance in period t leads to a pretax earnings growth in period t+1, given prior literature on the information content in the valuation allowance being able to predict loss persistence. Building off this, I then study the regulatory scrutiny that a valuation allowance release garners. I predict that since a release could be deemed to be either an earnings management tool or a change in earnings forecast, external monitors would be more likely to observe the company releasing the VA closer, either by issuing an SEC tax comment letter or through increased audit fees.
Using observations from 1993 to 2014 gathered from the Compustat Annual database, I study the various determinants that lead to a release of the deferred tax valuation allowance, as it is important to identify the information that leads firms to release a portion of their valuation allowance. I begin my observations in the 1993 fiscal year as it was the first full fiscal year since the Financial Account Standards Board (FASB) instituted the guidelines for setting a deferred tax valuation allowance. This sample provides a comprehensive set of firm-years with full valuation allowances while also containing the information necessary to study earnings growth in the following year. This sample is further adapted in order to study the effects of a valuation allowance release on regulatory scrutiny.
I examine the effects of valuation allowance releases on future earnings by studying the effect that a release of the deferred tax valuation allowance in period t has on the firm’s pretax earnings in period t+1. The results in both the full sample and profitable sample suggest that the release of the deferred tax valuation allowance is associated with increased pretax earnings in the subsequent earnings period. This confirms the first hypothesis that postulated that a release of the valuation allowance in the current period would result in pretax earnings growth in the following period.
I then examine tax accounts and regulatory scrutiny by observing how a valuation allowance release affects the regulatory scrutiny a firm is subjected to. When specifying regulatory scrutiny as receiving an SEC tax comment letter subsequent earnings period, I find evidence from my data suggesting a positive association between the release of the valuation allowance and the receipt of a tax comment letter. However, I do not find any statistically significant results suggesting that a valuation allowance leads to higher estimated audit fees. Despite this, my findings do provide initial evidence that regulatory bodies consider releases of the valuation allowance when looking at firms.
Building off the insights from the earnings test, I then proceed to examine the effect that a valuation allowance release has on a company’s stock returns. The findings from this test, however, were inconclusive as only one of the six tests yielded a statistically significant (p <0.1) result suggesting a positive association between a valuation allowance release and an increase in stock returns. The statistically significant result suggests that investors do not price the release of a valuation allowance regardless of prior cumulative profits or the interaction of prior cumulative profits and valuation allowance release, because if they did, it would be reflected in current period returns. This finding reveals that profitable firm-years are more likely to exhibit higher returns when there is a release of the valuation allowance, all else being equal. A potential reason for this is that as long as firms are reporting a pretax profit, investors see past earnings history and focus on events that occur in the current period, such as a release of the valuation allowance.
This study informs investors seeking to better understand how credible the information signal in a valuation allowance release is, while contributing to prior literature on the predictive ability of the valuation allowance on earnings and on prior literature on tax accounting’s impacts on regulatory scrutiny. I contribute to prior literature on the valuation allowance and earnings by finding evidence that suggests a positive association between the release of the valuation allowance in the current period and pretax earnings growth in the following period. I contribute to literature on tax accounting and its impacts on regulatory scrutiny by finding evidence that a release of the deferred tax valuation allowance is positively associated with receiving an SEC-issued tax comment letter, indicating that there is reason to believe releases lead to increased scrutiny by regulating bodies.
The remainder of this paper proceeds as follows: Section II provides institutional background on the valuation allowance for deferred tax assets. Section III discusses prior literature relevant to this study and develops the research hypothesis. Section IV discusses the research design and methodology. Section V discusses the results of the study and additional analyses performed. Section VI concludes.
Section II: Institutional Background
ASC 740 outlines two main objectives relevant to accounting for income taxes: to recognize taxes payable or refundable for the year and to recognize deferred tax assets and liabilities. Specifically, the latter objective aims to recognize the expected future tax consequences of events that have been recorded in the financial statements and/or tax returns (Deloitte 2015). These deferred tax assets and liabilities represent anticipated increases or decreases in taxes payable or refundable in future years that result from temporary differences and carryforwards present at the end of the current year. What this study aims to focus on specifically on the measurement of deferred tax assets.
Under ASC 740, deferred tax assets are recognized for “future deductions and operating loss and tax credit carryforwards” (Deloitte 2015). An example giving rise to a deferred tax asset would be the incurrence of net operating losses that could be used to offset taxes payable in future years. When measuring deferred tax assets, the FASB mandates that deferred tax assets are to be measured using a tax rate convention (FASB 1992). When recording a deferred tax asset, the entry is typically a debit to the deferred tax asset account with the associated credit being a credit to income tax expense, ultimately reducing the current period’s overall income tax expense.
ASC 740 formally recognizes four sources of taxable income by which a firm can recognize the financial statement benefits of deferred tax assets: “taxable income in prior carryback years if carryback is permitted under the relevant tax law, future reversals of existing taxable temporary differences, tax-planning strategies and future taxable income exclusive of reversing temporary differences and carryforwards” (PwC 2015).
ASC 740 outlines the need to reduce the measurement of deferred tax assets not expected to be realized (Deloitte 2015) All positive and negative evidence available should be considered and weighed when determining whether or not a valuation allowance is needed. If ultimately some portion or all of the deferred tax asset is deemed to be more likely than not realizable, a valuation allowance must be established. Avoiding a valuation allowance is difficult when the following criteria are present: “cumulative losses in recent years, losses expected in early future years, a history of potential tax benefits expiring unused, and uncertainties whose unfavorable resolution would adversely affect future results” (PwC 2015).
Adjustments to the valuation allowance can also impact income tax expense. Additions to the deferred tax valuation allowance further increase income tax expense and decrease earnings. The opposite is true when the deferred tax valuation allowance is reduced. In this instance, the valuation allowance for deferred tax assets is debited, and income tax expense is credited. This reduces the valuation allowance, increasing net deferred tax assets, and decreases income tax expense, increasing earnings for the period.
Take, for example, a firm that incurs a net operating loss of $250 that with a 40% tax rate yields a tax benefit worth $100 that must be carried forward into future years. At year-end, the firm will book a $100 debit to the deferred tax asset account with an offsetting credit to income tax expense. If the company expects to fully realize the entire net operating loss associated with the deferred tax asset, then there will be no need to book a valuation allowance. However, if the company does not expect to fully realize the entire deferred tax asset, then the company must establish a valuation allowance to account for the portion it does not expect to realize. For example, if the aforementioned firm thought it would only realize $40 worth of its $100 in net operating losses, then it would book a debit to income tax expense for $40 and a credit of $40 to deferred tax valuation allowance. This $40 credit to deferred tax valuation allowance serves as a contra-asset to deferred tax assets, bringing down net deferred tax assets down to $60. In turn, the establishment of the valuation allowance would decrease earnings for the period by $40 since income tax expense would have increased by the same amount. Additionally, deferred tax liabilities reduce the need for the establishment of a deferred tax valuation allowance. In practice, deferred tax assets and liabilities are netted and recorded as “net deferred tax assets” on a company’s balance sheet. A company with a net deferred tax assets balance of zero either has deferred tax assets equal to deferred tax liabilities or has deferred tax assets, less a valuation allowance, equal to deferred tax liabilities. For example, a company with $100 in deferred tax assets, $60 in deferred tax liabilities, and a $40 deferred tax valuation allowance has net deferred tax assets equal to zero. Thus, the presence of the deferred tax liability in the previous example reduced the amount of valuation allowance the company had to record in order to zero out its net deferred tax assets. The deferred tax valuation allowance plays an important role in the earnings reported by firms with significant deferred tax balances. In the fourth quarter of 2015, American Airlines reversed $2.9 billion of its deferred tax valuation allowance, directly increasing earnings by that amount (Star Telegram 2016). The reversal was put in place because American Airlines’ forecasts indicate that it will more likely than not realize its deferred tax assets, specifically it’s net operating losses. Likewise, American’s competitor, United Airlines, also made adjustments to its deferred tax asset valuation allowance. The company released its valuation allowance in the third quarter of 2015, which yielded it a $3.2 billion benefit in its provision for income taxes due to new expectations of realizing its net operating losses due to positive income forecasts in future years (ATW 2016). Anticipated fundamental transactions such as initial public offerings, business combinations or major financing transactions can also result in the establishment of a deferred tax valuation allowance (PwC 2015). Such events can limit a firm or firm acquirer’s ability to fully realize a company’s net operating losses due to limitations in the Internal Revenue Code. These events are important to be cognizant of when observing a firm’s deferred tax valuation allowance and any adjustments made to the account.
Section III: Literature Review and Hypothesis Development
Dhaliwal et al. (2013) investigate whether or not investors understand the material decisions managers take when establishing and adjusting a valuation allowance. They find that investors understand the significance of a valuation allowance. Additionally, they find that in firms reporting losses, the valuation allowance represents a credible signal about future earnings. This paper highlights the significance of having a valuation allowance and the implications it has on future earnings. This builds upon research conducted by Jung and Pulliam (2006) that finds incremental information content within the valuation allowance in predicting future income and future cash flows. Kumar and Visvanathan (2003) find that investors use the valuation allowance to gauge management’s performance forecasts.
Rayburn (1986) finds that current accruals have information content and have a significant association with abnormal returns. In addition to this, Rayburn’s study finds that noncurrent accruals may also have a significant association with return, despite prior research being inconclusive on this topic. Rayburn ultimately finds that the tax accrual provides incremental information in regards to cash flow. Chaney and Jeter (1994) reach similar conclusions in their study, finding that the deferred tax component of earnings provides incremental information to the market.
Building off of the research conducted by Rayburn, Chaney and Jeter, studies began to focus on the pricing of the deferred tax accounts by the market. Studies by Amir et al. (1997), Ayers (1998), Amir and Sougiannis (1999), and Dhaliwal et al. (2000) find that the market does price deferred tax accounts, with the exception of the deferred tax valuation allowance. They find that the market prices the various components of the
deferred tax asset based on the likelihood of realization and estimated time until realization. Givoly and Hayn (1992) study the change in market pricing of deferred tax accounts based off changes in corporate tax rates, and find that the market prices deferred taxes as true assets and liabilities, specifically as future cash sources and cash outlays that get factored in to a firm’s valuation.
Investigations by Schrand and Wong (2003) and Frank and Rego (2006) both look at whether or not managers use the deferred tax valuation account to manage earnings. Schrand and Wong examine the valuation allowance for publicly held bank holding companies. They find that in addition to using the account to manage earnings, the account is also established according to the nature of the valuation allowance’s related deferred tax asset or assets. When using the account to manage earnings, Schrand and Wong find that managers use the valuation allowance to smooth earnings towards both the analyst consensus target and prior years’ earnings per share. Frank and Rego expand on these findings by studying a broader industry set and confirming that managers do in fact use the deferred tax valuation allowance to manage earnings toward the mean market analyst forecast. Further research by Christensen, Paik, and Stice (2008) studies whether or not managers use a deferred tax valuation allowance reversal to manage earnings. They find that firms to do not create a deferred tax asset valuation allowance in a big bath setting primarily to create a cookie jar reserve, as it is a highly visible method of managing earnings that investors would see through. Their results also suggest that managers use private information about future performance to set the valuation allowances at a justifiable amount. In addition to these findings, they also discover that a reversal of the deferred tax valuation allowance is not used as a method of turning a loss into a profit, but may be used to meet or beat the mean analyst forecast. This study further confirms that managers use the deferred tax valuation allowance account to manage earnings, further mitigating the information content contained in the valuation allowance.
Allen (2012) provides additional insight as to why a company may record a deferred tax valuation allowance. He finds that a large majority of his sample reports a full valuation allowance, which he finds to be a function of conservative accounting. In addition to this, Allen finds that in the instance of a change in ownership, not all net operating losses may be realized by the new owner due to legal provisions, thus meriting the need for a valuation allowance. What Allen’s study suggests is that there may be various reasons outside of projected earnings for management to establish a valuation allowance, specifically, reasons that may have been ignored in previous studies. Ultimately, statutory restrictions imposed on the realization of some deferred tax assets may mitigate the usefulness of the signals investors infer from the valuation allowance.
Based on theory and evidence stating that the deferred tax valuation allowance contains forward-looking information regarding future income and future cash flows and the literature that indicates the importance of the valuation allowance to investors, I predict that a release of the deferred tax valuation allowance is predictive of earnings growth in the following period.
H1: A release of the deferred tax valuation allowance in period t will indicate earnings growth in period t+1.
In regards to external scrutiny, Donahoe and Knechel (2014) observe the relation between tax aggressiveness and audit fees. They find a positive association between the two during firm-years observed between 2002 and 2010. Prior research by Heninger (2001) and Manry, Mock and Turner (2007) explain this as being due to the increased risk of misstatements, restatements and lawsuits in tax aggressive firms, with auditors adjusting for this risk in their pricing. Frank and Rego (2006) observe increased audit efforts with companies recording tax reserves, such as the valuation allowance, in order to confirm that those accounts are not being adjusted solely for the purpose of earnings management. More so, studies by Hanlon et al. (2012) find that large book-tax differences tend to increase audit fees. This applies to the deferred tax valuation account, as both establishing and making adjustments to it result in changes in book income, but not taxable income.
Kubick et al. (2016) details that the Securities and Exchange Commission issues comment letters when it is deemed that a registrant’s filing is “materially deficient or if the filing requires additional clarification.” Upon receipt of an SEC comment letter, a firm has ten days to respond providing additional information, disclosures and a guarantee to either adjust future filings or restate previous filings. Literature by Deloitte (2012), PwC (2014) and Kubick et al. (2016) suggests that tax-related comment letters are most often related to issues concerning permanently reinvested earnings, uncertain tax benefits and the deferred tax valuation allowance. These findings build off studies conducted by Boone et al. (2013), which find a positive correlation between firms that are subject to standards that require subjective accounting estimates, such as the valuation allowance, and receiving an SEC tax comment letter. Additionally, Cassell et al. (2013) determines firms with lower profitability and higher complexity (e.g. those that have established and hold valuation allowances) are more likely to receive tax comment letters. Similar to the findings of Hanlon et al. (2012) for audit fees, Kubick et al. (2016) find that higher permanent book-tax differences are positively associated with increased scrutiny by the SEC. Based on evidence from prior literature suggesting that large book-tax differences increase external scrutiny, I predict that a release of the valuation allowance will increase regulatory scrutiny given its potential material impact on earnings.
H2: A release of the deferred tax valuation allowance in period t will increase regulatory scrutiny either in that same period or in subsequent periods.
Section IV: Research Design
Sample
This study concerning pretax earnings growth, H1, uses Compustat Annual files
to construct the primary sample for the period from 1993 to 2014. The sample begins in the year 1993 as this was the first full fiscal year that occurred after FASB instituted Statement No. 109 in 1992, now known as ASC 740. The study’s observations end in 2014 due to data limitations. This yields 17,904 observable firm years. Consistent with prior literature by Dhaliwal et al. (2013), I remove firms operating in banking or regulated industries. Additionally, firm-years missing beginning and ending net deferred tax assets are removed as this information is necessary to determine whether or not a firm released part of its valuation allowance. Further, I remove firm-years with net deferred tax liabilities as it is not possible to discern the existence of a valuation allowance given the data derived from Compustat. By excluding firm-years missing next period earnings (2,069 observations), firm-years missing return information (8,561 observations), firms in banking or regulated industries (SIC codes 4900-4999 and 6000-69992; 1,384 observations), firm-years with a net deferred tax liability (328 observations) and firm- years missing control variables (120 observations), this yields a sample size of 5,442 firm-year observations.
In order to test H2, the sample used to test H1 is further adapted. To identify firms that received a tax-related comment letter, the Audit Analytics Comment Letter Database is used. In line with Kubick et al. (2016), the taxonomy provided by Audit Analytics is used to identify a tax-related comment letter in addition to a separate search of comment letter issues listed under the following words: “Tax”, “FAS 109”, “FIN 48” and “ASC 740.” Audit fee information is extracted from Audit Analytics. Firm-years missing information regarding audit fee and SEC tax comment letter information are excluded for the purposes of testing H2. Removal of 1,350 firm-years missing audit fee and/or tax comment letter information yields a sample size of 4,212.