The Effect of Using a Lattice Model to Estimate Reported Option Values


  • Accepted by Christine Botosan. This paper has benefited from helpful comments by Terry Adamson, Linda Bamber, Brian Cadman, John Campbell, Michael Crawley, Dain Donelson, Thomas Hemmer, Leslie Hodder, Derek Johnston, Linda McDaniel, Paul Newman, Robert Resutek, Steve Utke, Mohan Venkatachalam, Laura Wang, Matthew Wieland, Sunny Yang, Yong Yu, Dave Ziebart, and participants of accounting research workshops at the University of Texas at Austin, the University of Georgia, the Chinese University of Hong Kong, the 2008 American Accounting Association annual meeting, the 19th Annual Financial Economics and Accounting Conference, and the 2009 mid-year meeting of the Financial Accounting Research Section of the American Accounting Association. The authors thank AON Consulting for providing a list of firms that began using a lattice model in recent years and Sarah Shonka and Stephanie Wei for their research assistance. Finally, we gratefully acknowledge the financial support of the Von Allmen School of Accountancy endowment and the Gatton College at the University of Kentucky, the McCombs School of Business at the University of Texas at Austin, and the Terry College of Business and J.M. Tull School of Accounting at the University of Georgia.


Statement of Financial Accounting Standards 123R suggests that lattice valuation models may improve the estimates of reported employee stock option values relative to the more commonly used Black–Scholes (BS) model. However, lattice model critics have expressed concerns that managers may use lattice models' flexibility to opportunistically understate option values. In this study, we investigate a sample of firms that recently adopted a lattice model to value employee stock options to provide evidence on this issue by identifying the determinants of lattice model adoption and examining the effect of lattice model use on reported option values. We report three main results. First, we find that firms are more likely to adopt a lattice model when it is more likely to produce lower values than the BS model and when managers have incentives to lower stock option expense. Second, we find that firms adopting a lattice model increase understatement of reported option values more than firms that continue to use the BS model and that the incremental understatement is due to use of the lattice model. Third, we conduct several tests to examine whether the valuation effect of lattice model use is consistent with efforts to correct for documented shortcomings in the BS model and find no evidence that this is the case. Taken together, the evidence in this study suggests that firms adopt and implement lattice models primarily to lower reported option values.