Conceptual conformity of deferred tax accruals: evidence from Germany and the United Kingdom

ABSTRACT

This article argues that it is more appropriate to model the relation between deferred tax accruals and subsequent tax payments using a first-difference model rather than the levels model employed in the extant literature. We then use the difference model to assess the conceptual conformity of deferred tax accruals by testing if changes therein predict changes in tax payments. Using a combined sample of British and German firms we report conceptual conformity for deferred tax liabilities but not for deferred tax assets. By allowing UK and German firms to have different relations between deferred tax accruals and tax payments, we investigate the impact of country-specific factors, including the taxation system, on this predictive ability. We find conceptual conformity for deferred tax liabilities in both countries but report conceptual conformity for deferred tax assets only for German firms. Finally, we find that the introduction of the Accounting Law Modernization Act (Bilanzrechtsmodernisierungsgesetz (BilMoG)) in Germany in 2010 has brought the relation between changes in deferred tax accruals and changes in future tax payments of German and UK firms much closer together. Currently, the only notable difference between UK and German firms related to deferred tax liabilities pertains to the lag between changes in deferred tax liabilities and changes in tax payments that we attribute to differences in the tax depreciation rates of each country.

KEYWORDS:

• IFRS
• conceptual framework
• deferred tax
• accounting harmonisation

Acknowledgment

The authors acknowledge the support of the Editor, Giovanna Michelon, the Associate Editor Amy Hageman, and the extremely helpful comments, insights, and encouragement of two anonymous Reviewers. All remaining errors and omissions remain ours.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 For 10% of firms in their sample the deferred tax expense is almost 60% of profit before tax.

2 Thus, an accounting accrual is recorded in the balance sheet even if the cash flow is tax first.

3 See Appendix B of Dechow and Dichev (2002) for examples.

4 Note that Dechow and Dichev’s (2002) empirical model explains changes in working capital accruals with cash flows before, contemporaneous with and after the accrual. This is consistent with our approach.

5 Laux’s assumption of additional assets being purchased over each of the first three years confounds the examination of the accrual reversal in his analysis. While it is important to allow for reversal, we have extended our example to longer periods and have allowed for growth in the asset base. We have simulated the regression of changes and levels in the subsequent taxes paid on changes and levels of accruals. The changes regression has a higher R² and more significant F-statistics than the levels regression.

6 http://taxsummaries.pwc.com/ID/Corporate-income-tax-(CIT)-due-dates (accessed: 17/03/2023).

7 Large German companies could voluntarily adopt IFRS from 1998. By 2003, 19 of the 30 firms on the German DAX had voluntarily adopted IFRS and a further nine used US GAAP (Costa et al., 2005). The motivation behind this was the need for large German firms to access capital in foreign markets.

8 Refinitiv also includes data on deferred tax liabilities (WC18183). We downloaded those data and found that they were exactly the same as those we obtained from Global Vantage.

9 The five-year period can also be justified on the basis that in Germany, deferred tax assets pertaining to losses can be recognised only to the extent that the tax loss carry forward can be utilised within five years.

10 Models (1) and (2) are estimated using OLS with standard errors clustered by unique firm. Continuous variables are winsorised at the 1st and 99th percentile.

11 The mean level of deferred tax assets and liabilities scaled by average total assets for UK (German) companies are 0.025 (0.034) and 0.025 (0.023), respectively. The corresponding figures from Chludek’s (2011) sample of German firms are 0.032 and 0.027.

12 While the values of ΔDTLIAB look similar in Table 4, this is the effect of the rounding process. The actual value of ΔDTLIAB for the UK sample is 0.00141 while it is 0.00075 for the German sample.

13 Notwithstanding the low correlations, a diagnostic test for multicollinearity through the estimation of the variance inflation factor (VIF) coefficients for all regressions was carried out. The VIF coefficients are always below the threshold of 10 (Kennedy, 2008), suggesting that multicollinearity does not affect our analyses.

14 This analysis is based on a random subsample of 93 firms from our UK sample and from Table 2 of Evers et al. (2014). We estimate that 72% of firms in our UK sample have deferred tax liabilities associated with tangible assets, while the corresponding percentage for German firms is only 12%, although this is the largest proportion of German companies having deferred tax accruals associated with any underlying asset/liability type in the German sample.

15 While German tax depreciation is based on accounting depreciation, special tax rules are then applied and given priority. Further, German tax depreciation rates vary from 2% to 33% but the figures in the UK range from 0% to 100% (EY, 2022).

16 The book value of the pension liability may increase due to a variety of factors such as increased entitlements, changes to plan terms, unwinding of the present value of the pension liability and changes to actuarial assumptions used in calculating the present value. An increase in the net pension liability may motivate firms to increase tax-deductible future contributions to the pension scheme to reduce the deficit. This could cause deferred tax assets from pension liabilities to predict future decreases in taxation payments for German firms.

17 The tables that are not tabulated here and are not reported in any Online Appendix are available from the authors upon request.

18 For example, a net operating loss or credit carry forward cannot be utilised unless there is future taxable income; thus the deferred tax asset does not have the opportunity to crystallise and cause a reduction in taxation cash flows. Current losses that create a deferred tax asset can only crystallise if there is future taxable income to offset the carried forward loss against.

19 Net operating losses and the resulting absence of a tax liability can impact on the relationship between both deferred tax assets and deferred tax liabilities and future taxation cash flows. An example of a deferred tax liability would be capitalised development expenditure that is the subject of accelerated deductions for tax purposes. If there is no future taxable income, there will be no relationship between deferred tax liabilities and future tax payments, as the tax payments will be bound at 0.

20 We are grateful to an anonymous reviewer for drawing our attention to this test.

21 Losses can also be carried back in Germany but not as easily as in the UK.