4.4 Robustness check

To provide a robustness check on the results obtained, we have compare the output generated when running our returns and survivorship analyses using the Modified Jones Discretionary Accruals (MJDA) method. As shown in Table 7, there is no difference between the two results.

From Table 7, we can deduce that our results in evaluating predictors of future performance are robust in being replicated across the MJDA and JDA procedures. Each regression model provides the same significance and similar explanatory power through their R-squared. All variables across the two methods possess the same signage of coefficients and remain within the statistical significance zone. In testing the robustness of our results in finding predictors of the survivorship of tech-firms, we find similar result using the JDA as presented in Table 8. All models remain significant just as in the MJDA results and the variables hold the same meaning within the outputs. These confirm that our results are consistent with the initial findings. We further perform our robustness check on the results from Tables 7 and 8 where we used our overall data including ETHICS.


The table shows the valuation and accounting variables' effect on the technology firms' returns from 2008 to 2010. The dependent variables are dummy variables indicating 1 if the technology firm fails in 6 months (6m), 1 year (1y), and 3 years (3y), and 5 year (5y) and 0 otherwise. Then we use the earnings to price ratio (EP), book to market ratio (BM), price to sales ratio (PS), market capitalization (MktCap), EBITDA, sales growth (Salesg), Modified Jones for discretionary accruals (MJDA), research and development (R&D), and firm's ethical behaviour score (ETHICS) as our independent variables. The values in parentheses are the <sup>t</sup>-values to the corresponding coefficients. \*

Significance at 10% level.

\*\*Significance at 5% level.

\*\*\*Significance at 1% level.

#### Table 7.

Full logistic analysis of failed NASDAQ firms using modified Jones model for discretionary accruals (MJDA).

(2008, 2009, and 2010) and run linear and logistic regressions as in our Sections 4.2 and 4.3. The results for the future returns and future survivorships are shown in

Full multivariate analysis of periodic returns of NASDAQ technology firms—using modified Jones model for

In Table 5, we do find highly similar relationship of EP, BM, MJDA and R&D with the future returns as in our Table 3. However, using the overall sample in Table 5 shows more significant effects of these variables on the future returns. Then we find that the ETHICS has positive effect on the future returns in technology firms. In other words, the ethical behaviour of the technology firms tend to increase the future returns. Thus, the opportunistic behaviour of the technology

The similar relationship of EP, BM, PS, MktCap, Salesg, MJDA, and R&D on future survivorships are also found between Tables 4 and 6 while the latter one using overall data tend to show more significant relationships. However, in this case, we do not find highly significant effect of ETHICS on the future survivorships of the technology firms. Therefore, the opportunistic behaviour of the technology

Tables 5 and 6, respectively.

discretionary accruals (MJDA).

firms is likely to decrease their future returns.

values in parentheses are the <sup>t</sup>-values to the corresponding coefficients. \*

MODEL : ratþ<sup>p</sup> ¼ a þ b1EP þ b2BM þ b3PS þ b4MktCap þ b5EBITDA

�147.8\*\* (�2.05)

R&D 8.64 (0.84) 90.11\*\*\* (3.9) 205.73\*\*\* (5.27) 231.66\*\*\*

3m 6m 1y 3y 5y

�971.81\*\*\* (�7.98)

BM �1.63 (�1.5) 15.35\*\*\* (6.26) 23.33\*\*\* (5.64) 25.51\*\*\* (3.31) 21.78\* (1.79)

Accounting and Finance - New Perspectives on Banking, Financial Statements and Reporting

MktCap 0.55 (0.55) �0.71 (�0.32) �2.08 (�0.55) �4.01 (�0.57) 13.44 (1.21)

Salesg 1.18 (0.83) 0.52 (0.16) �2.97 (�0.55) �7.71 (�0.76) �20.39 (�1.28) MJDA 0.2 (1.07) 1.3\*\*\* (3.1) 1.42\*\* (2.01) 4.21\*\*\* (3.19) 0.32 (0.15)

ETHICS 2.64 (0.45) 27.78\*\* (2.09) 179.05\*\*\* (7.97) 52.32 (1.25) 211.71\*\*\* (3.2) F-test 1.49 13.43\*\*\* 21.88\*\*\* 6.69\*\*\* 4.01\*\*\*

N 866 866 866 866 866 The table shows the valuation and accounting variables' effect on the technology firms' returns from 2008 to 2010. The dependent variables are 3 months (3m), 6 months (6m), 1 year (1y), 3 years (3y), and 5 years (5y) returns of the technology firms. Then we use the earnings to price ratio (EP), book to market ratio (BM), price to sales ratio (PS), market capitalization (MktCap), EBITDA, sales growth (Salesg), modified Jones for discretionary accruals (MJDA), research and development (R&D), and firm's ethical behaviour score (ETHICS) as our independent variables. The

0.01 0.11 0.18 0.06 0.03

�263.65 (�1.16)

(3.19)

49.28\*\* (2.03) 174.69\*\*\* (4.26) 179.28\*\* (2.35) 327.49\*\*\* (2.72)

�0.58 (�0.7) 1.58 (1.12) �0.07 (�0.03) �3.83 (�0.92)

�0.16 (�1.29) �0.16 (�0.77) �0.2 (�0.52) 0.12 (0.21)

�1137.59\*\*\* (�3.18)

200.81\* (1.75)

þb6Salesg þ b7MJDA þ b8R&D þ b9ETHICS

(�0.18)

(�1.58)

(�0.73)

(�1.07)

Const �5.69

EP �17.02

PS �0.27

EBITDA �0.06

Adjusted R2

Significance at 10% level. \*\*Significance at 5% level. \*\*\*Significance at 1% level.

Table 6.

72


MODEL : ratþ<sup>p</sup> ¼ a þ b1EP þ b2BM þ b3PS þ b4MktCap þ b5EBITDA þ b6Salesg þ b<sup>7</sup> JDA þ b8R&D:

The key implications for investors, equity holders and creditors derived from our results and analyses are threefold. Firstly, the alignment of our results and underlying expectation that certain variables should demonstrate significant explanatory power at various times of crisis illustrates the strong relevance of contingency theory in evaluating the phenomenon of tech firms during periods of financial turmoil. Secondly, through our analysis of results, we reiterate that traditional relationships of accounting valuation and earnings management measures may not always hold especially during crisis periods. Thirdly, from a general perspective, the results indicate that the variables employed in this study demonstrate greater predictive power in determining the phenomena of future tech-firm failure

The Roles of Accounting Valuations and Earnings Management in the Survivorship of Technology…

than performance.

DOI: http://dx.doi.org/10.5772/intechopen.85395

Author details

75

Oliver Neoh and Matthias Nnadi\*

provided the original work is properly cited.

School of Management, Cranfield University, Bedfordshire, UK

\*Address all correspondence to: matthias.nnadi@cranfield.ac.uk

© 2019 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium,

The table shows the valuation and accounting variables' effect on the technology firms' returns from 2008 to 2010. The dependent variables are 3 months (3m), 6 months (6m), 1 year (1y), and 3 years (3y) returns of the technology firms. Then we use the earnings to price ratio (EP), book to market ratio (BM), price to sales ratio (PS), market capitalization (MktCap), EBITDA, sales growth (Salesg), Jones for discretionary accruals (JDA), and research and development (R&D) as our independent variables. The values in parentheses are the <sup>t</sup>-values to the corresponding coefficients. \* Significance at 10% level.

\*\*Significance at 5% level.

\*\*\*Significance at 1% level.

#### Table 8.

Multivariate analysis on periodic returns of NASDAQ technology firms—using Jones model for discretionary accruals (JDA).
