We investigate the use, determinants, and earnings effects of special purpose vehicles. Based on a proxy of SPV activity that can be applied to a broad cross-section of firms over time, we find a two-and-a-half fold monotonic increase in the percentage of firms using at least one SPV during the eight-year period from 1997 through 2004. Tobit regressions of the determinants of SPV use show that SPV activity increases with financial reporting incentives and economic and tax motivations, but strong corporate governance tends to mitigate their use. In addition, the evidence is consistent with SPVs arranged for financial reporting purposes being associated with earnings management, whereas the same does not appear to be the case for SPVs set up mainly for economic, tax, and other reasons.
Using a sample of 6,473 firms from 1997 to 2004, we obtain 22,604 firm-year observations for which the requisite data are available. SPV use appears highest among industry groups that tend to be leasing-activity intensive, such as trading, real estate, and construction, traditionally viewed as one of the main activities involving SPV use. We also find relatively high SPV use in banking and telecommunications, consistent with these industries providing new avenues for SPV use during the 1990s, such as the securitization of financial assets and broadband capacity.
With respect to investigating the determinants of SPV use, Tobit regression results show that SPV activity is increasing in financial reporting incentives and economic motivations, but strong corporate governance mitigates SPV use. Specifically, we find that SPV use is positively related to: (1) leverage, (2) CEO bonus compensation, (3) availability of funds, and (4) demand for tax benefits, but decreasing in board independence and independent directors’ stockholdings. These results are robust to a variety of sensitivity tests, including the use of other model specifications besides Tobit (e.g., logit and OLS), different sample selection criteria, and alternative definitions of the dependent and independent variables. In terms of economic magnitude, inter-quartile increases in leverage, availability of funds, intangible assets, and board independence result in changes in expected SPVs of 1.31, -0.36, 1.05, and -1.04, respectively. These effects are quite large given that more than 70 percent of our sample observations have zero or one SPV. Also, SPV use is increasing in firm size, consistent with larger firms having greater technical expertise to handle the complexity of structured financing arrangements.
With respect to examining SPVs’ role in earnings management, we use the Tobit regression results to parse the number of SPVs for each firm-year into those predicted by financial reporting motivations, those predicted by economic considerations, and those predicted by other variables. We then investigate the relation between these predicted SPV components and two measures of earnings management – discretionary accruals and frequency of small profits or losses. We hypothesize that SPVs arranged for financial reporting reasons are likely to be positively associated with earnings management measures, whereas we do not expect a similar relation for SPVs arranged for other reasons. Our evidence based on both univariate and regression tests is consistent with this hypothesis. The economic magnitude of this association appears to be substantial. For example, when the number of predicted SPVs for financial reporting purposes increases by one, on average the probability that a firm reports a small gain instead of a small loss increases by 18 percent.
The full paper is available for download here.