According to Bogstrand & Larsson (2012), accounting information (Book value per share, Earnings per share, Firm size, Leverage, Cash flow and Current ratio) is said to be value relevant if it can disclose pertinent information that influences the value of a firm (share price). Shilpa, Nisha, & Soral (2016) see value relevance as the ability of disclosed information of an entity’s financial statement to capture and summarize its firm’s value.

Lako (2018) posits that accounting information is said to be value relevant if the fundamental aspect-the ability of the financial statement to capture the intrinsic value of the firm; the predictor aspect-the ability of the financial statement to predict the future book value, earnings, cash flows, dividends, etc.; the information aspect—this is the statistical relationship between accounting numbers and share prices and the measurement aspect—the ability of accounting numbers to sufficiently capture and summarize firm value. Thus, the higher the value relevance, the higher the information that can be relied upon to make investment decisions and the closer the relationship between the financial statement and the share price of the firm. (Figure 1)

The decision usefulness theory was used to underpin this study. This theory can be traced to the American Accounting Association (AAA) in 1966. Accordingly, Dandago & Hassan (2013) opine that the decision usefulness theory is an approach in the preparation of financial statements meant to satisfy the primary need of the investors and the creditors but do not consider other stakeholder's needs. However, Shagari & Dandago (2013) noted that the primary users of the financial statement are the investors who are chiefly concerned with the risk and return of their investment and as such assess the effectiveness and efficiency of the management of the entity.

Hitz (2007) noted that decision usefulness theory has two groups. They are the decision model group and the decision-maker group. The decision model group refers to the stewards/managers that are the preparer of the financial statements and sees users of such information as secondary, whereas the decision-maker group represents the investors and other stakeholders and they focus on what they need from financial information and considers it as information useful.

Mardini (2012) posit that IFRS convergence has made financial reports of firms to be much more information useful, enhanced comparability and also increased information disclosure. This theory emphasizes that financial statements need to be useful so that they can be able to predict future performance. IFRS adoption has made it easy for comparability and increased information disclosure of entities and hence enhances the quality of decisions for all stakeholders. Thus, this theory can help to explain the level of IFRS compliance and the value relevance of accounting numbers.

Lako (2018); Sutopo, Sebastian Kot, Adiati, & Ardila (2018) have used the decision the decision theory to underpin their work.

Empirical evidence all over the globe on accounting information and value relevance nexus are mixed and thus conflicting. Some empirics report that accounting information in relation to the variables of interest is more value relevant in pre-IFRS than in post-IFRS. Hence, the study hypothesizes as follows:

H0₁: BVPS is not more value relevant in pre-IFRS periods than in post-IFIRS periods in Nigeria and South Africa.

Studies that reaffirmed that BVPS was more value relevant in pre-IFRS periods than in post-IFIRS periods include panel data studies by Olayinka & Ogundele (2017) while studying 52 Listed firms in the Financial sector in the Nigerian Stock exchange; Okafor, Ogbuehi, & Anene (2017) studied 12 Listed Consumer firms in the Nigerian Stock Exchange; and Prihatni, Subroto, Saraswayi, & Purnomosidi (2018) while studying all Listed firms in Indonesian Stock Exchange discovered that BVPS was more value relevant in pre-IFRS periods than in post-IFIRS periods. At the same time, others report that accounting information is more value relevant in post-IFRS than in pre-IFRS. For example, Juniarti, Novitasar,i & Tjamdinala (2018); Alade, Olwenhy, & Oluch (2017); Ali, Mahar, & Abdel-fetta (2018); Ateyebi, Salande, & Onyilokwu (2018); Ouki (2018) in separate studies discovered that BVPS was more value relevant in post-IFRS periods than in pre-IFIRS periods. However, Oraby (2017) while studying 12 Listed Banks on Saudi Arabia Stock Exchange discovered that BPS was more value relevant in pre-IFRS periods than in post-IFIRS periods.

H0₂: EPS is not more value relevant in pre-IFRS periods than in Post- IFIRS periods in Nigeria and South Africa.

In relation to EPS and Share price, Juniarti, Novitasari, & Tjamdinata (2018) while studying 60 listed manufacturing firms on Indonesia Stock Exchange discovered that EPS was more value relevant in post-IFRS periods than in Pre- IFIRS periods. Also, Alade, Olweny, & Oluoch (2017) while studying 46 Listed Non-financial firms on Nigerian Stock Exchange discovered that EPS was more value relevant in post-IFRS periods than in pre-IFIRS periods. Meanwhile, Ali (2018) and Prihatni, Subroto, Saraswayi, & Purnosmosidi (2018) in different studies both discovered that EPS was more value relevant in post-IFRS periods than in pre-IFIRS periods.

H0₃: FSZ is not more value relevant in pre-IFRS periods than in Post- IFIRS periods in Nigeria and South Africa.

In relation to FSZ and Share price, Ali (2018) in a study of 21 Listed Insurance firms on Saudi Arabia Stock exchange found that FSZ was more value relevant in pre-IFRS periods than in post-IFIRS periods; Eragbhe & Omokhudu (2018) in a study of 19 Listed Nigerian Money Deposit Banks in the Nigerian Stock Exchange found that FSZ was more value relevant in pre-IFRS periods than in post-IFIRS periods.

H0₄: LEV is not more value relevant in pre-IFRS periods than in post-IFIRS periods in Nigeria and South Africa.

In relation to leverage (LEV) and Share price, Sawcen & Hakim (2014) in a study of 10,838 observations from listed firms in UAE, Balirain, Jordan, Kwait, Qatar, Turkey and South African countries found that LEV was more value relevant in pre-IFRS periods than in post-IFIRS periods.

H0₅: OCF is not more value relevant in pre-IFRS periods than in post-IFIRS periods in Nigeria and South Africa.

In relation to OCF and Share price, Prihatni, Subroto, Saraswayi, & Purnomosidi (2018) in a study on all listed firms in manufacturing and financial services on Indonesian Stock Exchange found that OCF was more value relevant in post-IFRS periods than in pre-IFRS periods. Again, Olayinka, Olojade, & Ogundele (2017) in a study on 52 Listed Financial and Consumer goods firms on the Nigeria Stock Exchange found that OCF was more value relevant in post-IFRS periods than in pre-IFRS periods. More so, Okafor, Ogbuehi, & Anene (2017) in a study on 12 Listed Consumer goods firms on Nigerian Stock Exchange found that OCF was more value relevant in post-IFRS periods than in pre-IFRS period.

H0₆: CUR is not more value relevant in pre-IFRS periods than in post-IFIRS periods.

In relation to CURR and Share price. Prihatni, Subroto, Saraswayi, & Purnomosidi (2018) in a study on all listed firms in manufacturing and Financial services on Indonesian Stock Exchange, CURR was more value relevant in pre-IFRS periods than in post-IFRS periods. This report is in line with the findings of Olayinka, Olojade, & Ogundele (2017), and Okafor, Ogbuehi, & Anene (2017).

Given that the study involved already has existing data, the ex-post facto (after-the-factor) research design was adjudged as the most appropriate for the study. Meanwhile, the study specifically focused on the listed non-financial firms in Nigeria and South Africa. The companies used for the study are listed in each country’s stock exchange market. The study covered a period of 10 years for each country although the dates may differ given the fact that both countries adopted IFRS on different dates. For Nigeria firms, it covered from 2007 to 2016. The period of 2007-2011 relates to pre-IFRS adoption and 2012-2016 relates to post-IFRS period. For South African firms, the study covered from 2000-2009. The period of 2000-2004 relates to pre-IFRS adoption and 2005-2009 relates to post-IFRS period.

In the case of Nigeria, the sample size is 87 out of the 134 non-financial firms quoted on the Nigerian stock exchange as of 31^st December 2016 were selected. Whereas in the case of South Africa, the sample size is 80 out of the 200 non-financial firms quoted on the Johannesburg stock exchange website as at 31^st December 2009 selected. The sample selection was based on the Judgmental (Purposive) sampling technique. It was deemed appropriate for the study because it gives researchers the opportunity to choose the accessible population (sample size) using bias-free criteria. Justifiably, the selection criteria are:

1) The sampled firms must have complete data;

2) All financial firms were removed sampling frame;

3) Delisted firms are excluded from the sampling frame.

In order to determine value relevance, we compared the difference between the coefficient of independent variables (BVPS, EPS, FSZ, LEV, OCF and CUR) in pre and post periods. Where the difference in coefficients is positive or otherwise, it suggests that the variable value relevance increases in post-IFRS periods or otherwise The overall Adj R² for both pre and post was also used to determine the collective value relevance of all the independent variables. The study chose to use these statistics so that it can be reasonably assured whether the value relevance of the coefficients of the variables and the Adj R² statistics gives the same result or otherwise. Another reason is that previous studies have used the Adj R² as the measure of value relevance which includes but is not limited to Adeyemo et al., (2017); and Shilpa et al., (2016). In addition, one study that has used coefficient of the independent variable to determine pre and post value relevance is Ugbede, Mohd, Ahmed & Timothy (2014).

This study is modeled after the Ohlson (1995) model as shown in Ugbede, Mohd, Ahmed & Timothy (2014). Thus, the model of this study is stated below:

ASP i t = β 0 + β 1 EPS i t + β 2 BPS i t + β 3 FSZ i t + β 4 OCF i t + β 5 LEV i t + β 6 CUR i t + ε i t

ASP_it = Average share price;

EPS_it = Earnings per Share;

BPS_it = Book Value per Share;

FSZ_it = Firm Size (Logged value of Total Asset);

OCF_it = Operating Cash-flow;

Lev_it = Leverage (total debts/equity);

ε_it = Error term.

Note: All the study variables were logged so as to avoid spuriosity of result.

Decision Rule

The Adj. R-squared value of pre- and post-IFRS is compared in order to ascertain if accounting information is value relevant. Consequently, where in the Adj. R-Squared value of post-IFRS is higher than that of pre-IFRS. It suggests that accounting information has increased after IFRS adoption and thus more value relevant than in pre-FIRS adoption or otherwise.

Before regressing the sourced data, the data were first subjected to a series of preliminary analysis vis-à-vis descriptive statistics and correlation analysis with a view to track the variable trend over time and to ascertain if there were unusual occurrences. Furthermore, it was targeted at examining the extent of linearity between the regressors and the regressed. The analysis is therefore presented thus:

Table 1 below shows the pre-IFRS and post-IFRS adoption periods, Average share price (ASP), Book value per share (BPS), Earnings per share (EPS), Firm size (FSZ), Leverage (LEV), Operating Cash-flow (OCF), and Current Ratio (CUR) had average values of 24.665%, 0.0654%, 80.8325%, 3.5159, 0.202%, 3.0552% and 1.6235% whereas the standard deviation was put at 51.766%, 0.2829% 175.87% and 175.87%, 0.2302%, 1.0855% and 2.0373%. The lowest figure recorded were 0.0770, 0.0004%, −5.4931%, 2576.0, 0%, 1.9012% and 0.0049%, whereas the highest figure were 445.66%, 3.8602%, 1293%, 2.9219, 3.1687%, 29.827% and 1.3755%. As can be observed here, their standard deviation values were far away from their mean values which suggest that ASP, BVPS, EPS, FSZ, LEV, OCF, and CUR throughout the pre-IFRS adoption periods strongly deviated from their mean values.

Conversely, in the case of the post-IFRS adoption periods, Average share price (ASP), Book value per share (BPS), Earnings per share (EPS), Firm size (FSZ), and Operating Cash-flow (OCF) also strongly deviated from their mean values

Source: Result extracted from GRETL output (2021).

since their standard deviation values were far away from their mean values. Meanwhile, both Leverage (LEV) and Current Ratio (CUR) standard deviation clustered around their mean values.

Table 2 below shows the pre-IFRS and post-IFRS adoption periods in South Africa. The share price (ASP), Book value per share (BPS), Earnings per share (EPS), Firm size (FSZ), Leverage (LEV), and Operating Cash-flow (OCF) had average values of: 12.371%, 0.0834%, 0.1428, 4.9414, 0.2970 and 1.0374 whereas their standard deviation values stood at: 18.169%, 0.14312%, 0.3167, 1.0320, 0.4523, and 5.7135 Their minimum value were 0.01%, 0%, −0.2020, 643.00, 0%, and −7.4867 whereas their highest values were 147%, 0.9412%, 3.9500%, 7.3346, 6.8500, and 9.3550. The below result suggests that ASP, BPS, EPPS, FSZ, LEV, and OCF throughout the pre-IFRS adoption periods were highly volatile since their average figures deviate significantly from the mean. Comparably, from the above statistical figures, it can be noted that standard deviation of CUR was not far from the mean implying that the deviation from its mean value was not very significant throughout the pre-IFRS adoption periods.

Conversely, in the case of the post-IFRS adoption periods, Average share price (ASP), Book value per share (BPS), Earnings per share (EPS), Firm size (FSZ), and Operating Cash-flow (OCF) also strongly deviated from their mean values since their standard deviation values were far away from their mean values.

Source: Result extracted from GRETL output 2021.

The correlation analysis in Table 3 below reveals that BPS, EPS, FSZ, and OCF are positively correlated with value relevance proxy (ASP) throughout the pre- and post-IFRS-adoption periods in Nigeria. This clearly indicates that the more BPS, EPS, FSZ, and OCF increase, the more the value relevance. However, LEV and CUR are negatively correlated with value relevance proxy (ASP) throughout both periods. This implies that if LEV and CUR are reduced, accounting information will be more relevant. Overall, the correlation analysis clearly reveals that none of the independent variables exhibited high correlation. This suggests the possibility of the absence of a multi-collinearity problem.

The correlation analysis in Table 4 below clearly reveals that BPS, EPS, FS, LEV, and OCF are positively correlated with value relevance proxy (ASP) throughout the pre-IFRS and post-IFRS adoption periods in South Africa. This clearly indicates that the more BPS, EPS, FSZ, LEV, and OCF increase, the more the value relevance. However, CUR is negatively correlated with value relevance proxy (ASP) throughout the pre-IFRS adoption periods in South Africa. This implies that if CUR is reduced, accounting information will be more relevant. Overall, the correlation analysis clearly reveals that none of the independent variables exhibited high correlation. This suggests the possibility of the absence of a multi-collinearity problem.

Source: Result extracted from GRETL output 2021.

Source: Result extracted from GRETL output (2021).

This section covered the Panel Diagnostic Test, Coefficient and P-values, and Adjusted R-Squared values.