The objective of our work is to highlight the determinants, magnitude and impact of actual earnings manipulation by means of the gains from asset securitization made between a parent company or sponsor and its unconsolidated financing subsidiary or SPE .
Several motivations have guided our research. First, the corporate scandals at the beginning of this millennium on the U.S. financial market as the case of Enron and elsewhere on the world financial centers, have shown the inability of participants in these markets to identify discretionary accounting practices operated by leaders. The Enron case is not unique. Moreover, several other firms have been accused of adopting aggressive accounting practices such that (WorldCom, Xeros, the Italian dairy group Parmalat etc..). More recently, the financial crisis of subprime mortgages in 2007 triggered and exacerbated by the collapse of the financial group Lehman Brothers has put into question again the excessive use of securitization and off balance sheet financing. The common feature of these court cases is the fact that firms have adopted aggressive accounting policies, which for a long time, allowed them to achieve various objectives as political and contractual (maximizing executive pay or respect restrictions imposed by debt contracts) or those relating to financial markets (alignment to financial analysts' forecasts and obtaining forms of external financing at lower cost).
Second, and according to Kothari (2001) questions addressed in the literature on management accounting results showed a profound change. Moreover, long dominated by political considerations as the contractual maximizing executive pay, these issues are transferred to the motivations related to capital markets since empirical evidence showed that the participants in these Markets underestimate on average the impact of these manipulations on the results published (eg Sloan 1996, Xie 2001, Bradshaw et al 2001).
Third, there is some empirical evidence that financial market participants are able to integrate the effect of discretionary accounting manipulations based on the traditional concept of discretionary accruals in their forecasts of future earnings (Sloan 1996, Ahmed et al 2000 , Bauman 2003). This encourages managers to adopt new management practices of accounting results difficult to detect by stakeholders in the financial market.
Finally, the promulgation of reference for the translation of complex financial accounting terms as the United States (FAS 125, FAS No. 140, SOX, FAS No. 156) that the international (IAS 32, IAS 39, IFRS 7) has necessitated the proliferation of research to assess the impact of the adoption of the users of financial information to these new standards.
To better understand the phenomenon of manipulation based on actual use of a technique such as financial asset securitization, we have divided our work into three main objectives:
Get your grade
or your money back
using our Essay Writing Service!
First, we examined the determinants of the value gain resulting from securitization transactions. In this regard, three main types of factors have guided our work: the factors financial s (Feng et al 2006), economic factors (Mills and Newberry 2005), factors related to the governance of firms (Dechow et al 2008) and other control variables suggested by the literature and financial accounting.
Then, and according to the work of Barton (2001), Pincus and Rajgopal (2002) on the use of derivatives, Roychowdhry (2006) and Zang (2007) on the handling of real variables affecting operating cash flows, we examined the extent of manipulation based on actual gains from securitization transactions in relation to the traditional concept of discretionary accruals. A confrontation between the gains from securitization transactions with discretionary accruals estimated using the modified Jones model (1995) allows us to know if the two management tools results are used as two alternative or complementary instruments.
Finally, and by analogy with the work of Bradshaw et al (2001), Ahmed et al (2001) we examined the association between income excluding the impact of securitization and the resulting gain with the view the listener and stock returns.
To meet these objectives, we have taken as a sample set of U.S. listed companies from different industrial sectors pourlesquelles data on securitization of their receivables are disclosed in their annual reports (Dechow et al 2008, Karaouglu 2005).
In a first step, we conducted a series of univariate tests (descriptive statistics and comparison of means tests for independent samples). The objective of these tests is to understand the differences that may be between firms that engage in securitization transactions and other firms.
In a second step, and having regard to any limitations of OLS estimators, we examined the determinants of the value of securitization gain by basing ourselves on the selection model of Heckman (1979) estimated in two steps.
In a third step, and by analogy to the studies of Barton (2001), Pincus and Rajgopal (2002) on the one hand and Roychowdhry (2006) and Zang (2007) on the other hand, we estimated a system of equations Simultaneous between discretionary accruals and gains from securitization transactions using the SURE method of Zellner (1962). Other control variables or instrumental variables, suggested by the literature, were included in each equation. The objective is to verify whether managers use these tools to manipulate accounting results as substitutes or if both means are manipulated simultaneously.
Finally, and referring to studies by Bradshaw et al (2001), Ahmed et al (2003) about the behavior of financial market participants to the discretionary accruals, we examined the impact of gains from securitization transactions on the probability to issue a qualified opinion by the auditor in a binary logistic model. Finally, we examined the association between stock returns as dependent variable based on the gain from securitization operations and results excluding the effect of any securitization transaction using data from Panel.
Descriptive statistics show that, like the discretionary accruals, gains from securitization transactions have an average close to zero and are thus less variable as well as the amount of securitization that income arising. The comparison of means tests show that companies that securitize their loans differ from other companies in the amount of their accounts receivables, their current ratio, size, return on equity, the funds available or free cash flow, the percentage of shares held by the CEO.
The model estimation selection of Heckman (1979) in two stages shows that the amount of the securitization gain is determined by financial considerations (debt ratio), economic (free cash flow and interest coverage ratio), governance (percentage of shares owned by CEO and proportion of independent directors present in the board) and other control variables (return on equity and industry).
The estimation of simultaneous equations method Seemingly Unrelated Regresssion (SURE) of Zellner (1962) shows that firms use securitization gains and manipulate discretionary accruals as two interchangeable tools. Furthermore, analysis of the correlation between the gains from securitization transactions with the results excluding the effect of these operations (earnings before securitization) shows that the two variables are negatively related. This implies that these gains are used for smoothing the theoretical result and before securitization.
Finally, examination of investor behavior combined with the role of auditors, showed that these two actors act independently of the negative consequences that may result from such gains on the quality of accounting profit. Indeed, the probability to issue a qualified opinion by the auditor appears to be independent of the amount of gain from the securitization transactions. Furthermore, and contrary to our predictions, stock returns are negatively related to income excluding the impact of these operations (earnings before securitization). This implies that the auditor's opinion as well as the generation process of stock prices move independently of the value of the gain from securitization transactions.
Like any research effort, our work raises a number of limitations: first, the lack of disclosure that characterizes securitization transactions and relationships between a company and its subsidiaries not consolidated funding has limited the amount of information to securitization transactions (date, estimation of the different risks of default, prepayment and interest rate cuts etc..). Then, without a proxy to measure the discretionary portion of the securitization gain forced us to consider this gain as totally discretionary (see Dechow et al 2008). Added to that measurement error models estimates of discretionary accruals. Finally, it would be useful to use abnormal stock returns and the error in analysts' forecasts to better study the behavior of investors and financial analysts deal with such creative accounting.
Several implications follow from the results shown in this research. First, manipulations based on actual gains from securitization transactions may lead to an inadequate allocation of resources. Thus, excessive reliance on securitization transactions and recording frequent gains that result may divert funds on financial markets to investment projects and companies with financial reporting and biased accounting information of low quality. This again puts into question the traditional concept of efficient capital markets on which was built much of modern finance. Second, the lack of disclosure that characterizes these operations (see for example Dechow et al 2008) may bias the quality of financial reporting firms.
For this, it will be imperative for organizations of accounting regulation and financial market authorities to harmonize the rules and issues of using these financial engineering practices based on off-balance sheet financing to ensure garentie additional users of accounting. Moreover, and like models for measuring discretionary accruals, academic research should be interested in developing a proxy manipulations based on off-balance sheet financial engineering techniques or creative accounting to assaying the quality of financial reporting, key operating markets.
Always on Time
Marked to Standard