Analysts' Interpretation and Investors' Valuation of Tax Carryforwards
Tel Aviv University; London Business School
University of Illinois at Urbana-Champaign - Department of Accountancy
Contemporary Accounting Research, Spring 1999
This study examines the usefulness of forward-looking financial disclosures to two major users of financial statements: financial analysts and equity investors. Specifically, we examine how analysts incorporate deferred taxes from losses and credits carried forward in earnings forecasts. Then, conditional on analysts' earnings forecasts, we examine how investors incorporate deferred taxes from carryforwards into share prices. We focus on deferred taxes from carryforwards because in measuring and disclosing this information each period, management must use their private information about the firm's profitability prospects. Thus, accounting valuation and disclosure of tax carryforwards is another way for managers to provide earnings forecasts.
We identify two conflicting effects that influence the relation between share prices (or forecasted earnings) and deferred taxes from carryforwards. First, from a measurement perspective, deferred tax assets from carryforwards represent future tax savings; hence, they should be valued positively as assets by analysts and investors. In contrast, from an information perspective, the existence of tax carryforwards may signal a higher probability of future losses. This higher likelihood of losses would translate into higher perceived earnings volatility, and hence, may have a negative effect on expected earnings and share prices. Thus, our empirical tests are structured in a way that allows for both of these effects to affect analysts' earnings forecasts and equity values.
We estimate a recursive cross-section system of two regression models: an analysts earnings prediction model and an equity valuation model. The analysts prediction model relates the present value of expected abnormal earnings to current values of book value of equity, abnormal earnings, and deferred taxes from carryforwards. The valuation model relates share prices to book value of equity, the present value of expected abnormal earnings, and deferred taxes from carryforwards. We distinguish in our earnings prediction and equity valuation models between the measurement and information effects by allowing the regression coefficients in both models to vary by whether the firm has accumulated tax carryforwards or not. We predict positive coefficients on net asset values and earnings. However, we also predict that net assets and earnings would be valued less in companies that have deferred taxes from carryforwards than in companies without deferred taxes from carryforwards.
Results of estimating the analysts earnings prediction model reveal a positive association between current abnormal earnings and the present value of forecasted abnormal earnings. Interestingly, this association between current and future abnormal earnings is weaker for firms with deferred taxes from carryforwards than for firms without such deferred taxes. This result suggests that analysts consider earnings of firms with carryforwards to be less persistent due to the increased likelihood of future losses.
We also investigate whether analysts' forecast errors exhibit different patterns in firms with and without carryforwards. We find that analysts tend to be less precise and more optimistic (biased) in forecasting earnings of firms with carryforwards. However, the higher optimism and lower precision are most significant in fiscal 1992 and they become less visible in later years. Since fiscal 1992 was the first year companies adopted SFAS No. 109, we believe that the higher optimism and lower precision in analysts' earnings forecasts resulted from the introduction of a new and complex accounting rule for income taxes.
An analysis of investors' valuation indicates a strong positive relation between deferred taxes from carryforwards and share prices, suggesting that these carryforwards are valued as assets. In addition, investors value earnings and book values of equity of firms that have carryforwards less than in firms without carryforwards. Finally, we find that investors value the earnings and book values of firms with valuation allowances less than the earnings and book values of firms without a valuation allowance.
The combined results also suggest that analysts fail to consider at least some of the implications of carryforwards on expected cash flows. It is also possible that the analysts' forecast horizon of five years is too short to capture all the valuation implications of carryforwards.
JEL Classification: G12, G29, M41
Date posted: October 14, 1998