“IFRS adoption and earnings predictability: evidence from listed banks in Nigeria”

The quality of financial report and the extent to which investors rely on them to forecast future earnings is dependent on the accounting standards employed. The impact of IFRS adoption on earnings predictability of listed banks in Nigeria was examined in this study considering a sample of 11 listed banks in Nigeria. Categorically, data were obtained from the financial statement 2013 to 2014 (post-adoption period) and 2010 to 2011 (pre-adoption period). The data obtained were analyzed using regression on the Statistical Package for Social Sciences (SPSS). The study found a decrease in the ability of current earnings to predict future earnings after the adoption period. Thus, IFRS adoption has a negative impact on earnings predictability. The study further suggested that regulatory bodies of the banking sector should enforce strict adherence to IFRS procedures and principles, as well as put in place measures that will improve investors’ protection.


Introduction
The extent to which past earnings are used to explain current and future earnings is an important measure of earnings predictability and this can lead to more accurate valuation, as it enables investors to more accurately anticipate expected future cash flows (Valury and Jenkins, 2006; Schiemann and Guenther, 2013; Uwuigbe, Francis, Uwuigbe and Moyosore, 2016).The extents to which investors depend on disclosed or reported financial information to predict future cash flows is a function of the quality of information contained in those financial statements (Kantudu and Tanko, 2008).The quality of financial statements is also dependent on the accounting standards employed in their preparation (Yahaya and Adenola, 2011).The accounting standards issued by the authoritative body in different country influences the content of the financial statement to a large extent, thereby making it difficult to compare financial reports of companies prepared with different accounting standard in different countries.To enhance comparability, the need for a uniform set of international accounting standards is considered imperative.According to Arum (2013), the issues affiliated with financial statement comparability will be reduced by the embracing of a single set of international accounting standards.This will in the long run help to improve the quality of accounting information disclosed.
The adoption of International Financial Reporting Standard in accounting has been widely discussed in the accounting practice over the years, as it affects countries around the world.There has been a constant increase in the number of countries integrating the International Financial Reporting Standard (IFRS) as the main reporting standard around the world since 2001 (AICPA, 2012).The rate of adoption of companies increased more when mandatory adoption of IFRS by publicly listed firms was introduced by the European Union (EU), Australia, South Africa, and Turkey.Currently, over 120 countries are requesting the adoption of IFRS by their publicly listed firms (IAS PLUS, 2012; PWC, 2012).
Nigeria is one of those countries that have replaced their local GAAP with IFRS.The adoption of IFRS in Nigeria was considered imperative to enhance comparability as companies previously relying on the local capital market for fund search for debt and equity capital within and outside their shores based on the increasing rate of globalization of the world trade.Furthermore, Ernst and Young (2014) avowed that the financial crises that occurred within the period of 2007-2010 gave birth to political and economic pressure on countries to merge their accounting standards with more acceptable international accounting standards.As a major step to meet international standard, road map was disclosed by the Nigerian Government to provide a full transition to IFRS in January 2012 (Umoru and Ismail, 2012; Nassar, Uwuigbe, Uwuigbe and Abuwa, 2014).
The convergence of local accounting standards to IFRS has engendered major deliberation.Proponents claimed that compelling organizations in Nigeria to disclose financial information under the IFRS have made the acquisition of international capital easier, advanced market enlargement, improves comparability of financial reports across countries thereby reducing the learning cost faced by financial analyst in analyzing the financial statement prepared in different standards and will engender considerable competence to companies reporting beyond different jurisdictions (Okere, 2009;Essien-Akpan, 2011;Ahmed, 2011;Okundi, 2013;Madawaki, 2014).Nevertheless, opponents suggest that the adoption of IFRS by firms in Nigeria will be expensive and due to the difference in the application of IFRS across countries, it may be difficult to realize the benefits of comparability (Martins, 2011;Monisola, 2013).Moreover, factors such as economic, political, regional, quality of legal enforcement and the difference in interpretation, as well as firm-specific characteristics are considered less important when comparing the implementation effects of IFRS adoption among countries (Augustine and Eguasa, 2014; Uwalomwa, Francis, Uwuigbe and Ataiwrehe, 2016).
Although, there are plethora of prior research that has been carried out on the impact of IFRS adoption on firm's earnings predictability in developed economies (Atwood, Michael, James and Linda, 2011; Abdelmohsen, Gehan, 2014; Bahloul and Ben, 2014; Priscilla, Georgious, Konstantinous and Nikolaos, 2015).However, the same is not true for developing economies (e.g., like Nigeria), where there is a dearth of literature as it relates to the subject.Hence, the need for this study arises.This study basically attempts to contribute to the debate on IFRS adoption by examining whether earnings is predictable under IFRS compared to previous Nigeria GAAP and whether there is a difference in current accounting earnings and future cash flows in IFRS versus previous Nigeria GAAP.Extant studies such as Ashbaugh and Pincus (2001); Barth, Landsman and Lang (2008); Jeanjean and Stolowy (2008) suggest that there is a difference in earnings management and analyst forecast error under IFRS compared to domestic accounting standard, thus, motivating the need for this study to focus on the earnings predictability of firms under the IFRS framework.The current earnings of firms are reflected in their future operating cash flows (CFOt+1).Earnings predictability can not only provide a better understanding of expected future cash flows (CFOt+1), but also determines future investment and lending activities (Hasan, Park and Wu, 2012).Despite the importance of earning quality and predictability to banks, this has been barely considered in the empirical literature to the best of the researcher's knowledge.To this end, this study aims to examine the impact of IFRS implementation on earnings predictability of listed banks in Nigeria Stock Exchange (NSE).In addition, it attempts to examine whether IFRS adoption improves future cash flows of banks compared to the previous Nigerian GAAP.
To gain more insight into this study, the remainder of this paper has been organized in the following sequence.Section 1 presents the literature review and hypotheses formulation, section 2 presents the research method adopted.While section 3 presents the results from the analysis in tabular form and the discussion of findings.The final section presents the conclusion followed by the recommendation and suggested area of future research.
1. Literature review 1.1.IFRS adoption.The increasing rate of globalization and the integration of financial markets has prompted the need for a uniform financial language (IFRS), as it makes financial reporting and decisionmaking more efficient.The drive towards a high quality accounting standard arose as a result of the formation of International Accounting Standard Committee (IASC) in 1973 by 16 professional bodies from different countries, which includes Canada, United States of American, United Kingdom, Germany, Australia, Japan, Mexico and the Netherlands (Garuba and Donwa, 2011).International Financial Reporting Standard (IFRS), as the term implies, is a principle based standard that has been accepted globally and has been adopted by many countries (Wilson, Ioraver, Adaeze and Iheanyi, 2013).
With the road map provided by the Nigerian government and the subsequent adoption of IFRS on January 1, 2012 and January 2014 by listed firms on the Nigeria Stock Exchange (NSE) and small and medium scale enterprise (SMEs) as part of the plan to meet international standard (Umoru and Ismail, 2010), several studies have examined the benefit of IFRS adoption and concluded that its adoption will be of great benefit to countries that adopts it, as it will bring about uniformity in financial reporting (Uwuigbe, Olubukunola and Okorie, 2015; Najeb, 2014).According to Wilson et al. ( 2013), IFRS adoption offers companies (especially multinationals or prospective ones) the facilities and opportunity to demonstrate to the international market that their financial statements are IFRS compliant.Furthermore, in the findings of Ugbede, Mohd and Ahmad (2014), IFRS adoption engenders high quality of banks financial statement.Also, report of Asian and Dike (2015) released by KPMG in 2010 stated that Nigerian banks adopting IFRS will not only affect financial reporting, but also other areas of the business such as debt covenants, compensation and bonuses arrangements, legal contracts, among others.
Failure on the part of firms to fully implement IFRS requirements will lead to a lack of transparency and accountability, falsification of financial reporting, inconsistencies, which will, in turn, amount to insufficient financial reporting practice (Abata, 2015).

Earnings predictability.
To review the true value of a firm, there is a need for high-quality earnings, which serve as a signal of existing level of performance to market investors (Dechow et al., 2010).As a result, more attention is given to sustainability and stability of quality reporting and future cash flows enabling several users to select contract firms, make investment and valuation decisions about listed corporations (Mashayekhi and Baziz, 2010) and, in turn, estimate its future prospects.The use of accounting-based attributes such as predictability, accruals quality, persistence, earnings management and smoothness is widely supported (Srinidhi, Gul and Tsui, 2011).Earnings management can be described as the use of certain accounting methods, changing methods by the avenue of speeding up or delaying firm expenditures or revenue, or other techniques solely tailored to manipulate earnings (Uwuigbe, Uwuigbe and Okorie, 2015).Several features of a firm on the basis of operating environment and business model are relevant determinants of earnings predictability.Cash flow volatility, revenue, operating cycle duration, debt and firm size are related to earnings predictability (Valury and Jenkins, 2006).Several studies have shown that other factors such as audit quality, accounting standards, and governance activities also influence earnings behavior.Cash flow predictability is positively associated with form of ownership (public and private) and ownership types such as institutional, block and concentrated (Hai, 2014;Velury and Jenkins, 2006).This suggests that poor earnings quality does not necessarily arise from earnings manipulations, window dressing, and poor reporting, as mostly, low earnings quality reflects a possibility of declining firm-specific features or a scenario of risky business operations.

IFRS adoption and earnings predictability.
Earnings predictability is the measure of how current earnings can accurately forecast or predict future earnings of firms.There have been several arguments on the effects of IFRS adoption on earnings quality (earnings predictability) of firms.Some arguments suggest significant changes in earnings quality ( 2011).Arguments suggesting that the IFRS adoption yields significant improvements in earnings quality often rely on the premise that IFRS are arguably more principles-based than domestic accounting standards.Thus, accounting numbers under IFRS seem to better reflect a firm's underlying economics than those under domestic standards (Hai, 2014).For example, IFRS allows fair value accounting in order to best reflect the underlying economics.However, the prediction depends on whether the assets are sellable or have a market (Linsmeier, 2013).Atwood et al (2011) in a related study avowed and observed that IFRS and non-IFRS regimes show no difference in earnings and cash flow predictability across firms reporting under IFRS and U.S. GAAP.Rather, future cash flows were less related to losses reported under non-U.S.DAS than future profits.IFRS adoption is associated with lower cash flow predictability (Mollah et al., 2015).However, Hai (2014) in a study of Canadian firms, observed that earnings quality on the average improved following the adoption of IFRS suggesting that IFRS has a positive impact on earnings quality thereby increasing the ability of current earnings to predict futures earnings and cash flow.Meanwhile, Muhammad, Reza, Mohammad and Tony (2015) stated that "mandatory IFRS adoption improved earnings quality by decreasing abnormal accruals and earnings conservatism".But value relevance of earnings prepared using IFRS is not higher for large companies (Hayfa, Nadia and Serra, 2013) and earnings quality remains the same after IFRS adoption, as there is no incremental difference in discretionary accruals (Priscilla, 2012).This lacuna in the literature the effects of IFRS adoption on earnings quality (earnings predictability) of firms provides a basis for this study.

Development of hypotheses.
Based on the lacuna in extant literature, the following hypotheses stated in the null form will be tested in this study.
 H 1 : IFRS adoption does not improve earnings predictability compared to previous Nigeria GAAP. H 2 : IFRS adoption does not improve cash flows predictability compared to previous Nigeria GAAP.

Research method
The where Earnings i,t is net income before extraordinary items of firm i in year t divided by total assets.A higher β 1 indicates a more predictable earnings stream, while values close to zero reflect transitory earnings.The adjusted R 2 from the regression is interpreted as earnings predictability.Large (small) values of predictability suggest more (less) predictable earnings.
The second measure of predictability is the cash flow predictability, as useful performance measures for predicting future cash flows are considered more desirable by investors (FASB, 2002; Barton, Hansen and Pownall, 2010).In support of this view, Dichev, Graham, Harvey and Rajgopal (2013) suggest that earnings backed by cash flows are of higher quality.This study estimates the ability of earnings to predict future cash flow as the adjusted R 2 from the regression below: where CFO is cash flow from operation and earnings are net income.Both are scaled by coexisting total assets.

Control variables.
In accordance with the study of Norrman and Vaigur (2013) and Herly (2015), it is required that other variables influencing earnings predictability should be controlled.This study controlled for firm size, and leverage of the firm.According to Hassan and Bello (2013), larger firm earnings allow the ability to predict future earnings and cash based on the exceptional supervision of management due to their strong internal control system, governance system and external attention from a financial analyst.The natural logarithm function of the total asset is adopted in this study to evaluate the size of the firm in accordance with the study of Alves (2014).
Leverage represents the debt riskiness of a firm, as it reduces the ability to predict future earnings.Indebted firms have the ability to influence cash flow predictability resulting on high cash flow predictability (Valury & Jenkins, 2006).In line with the study of Uwuigbe, Uwuigbe and Okorie (2015), the total liabilities divided by total assets represents leverage, it is expected that as leverage increases, the predictive attribute of earnings reduces.

Model specification.
In order to test the overall effects of IFRS adoption on the predictability of each earnings component, the study interacts each component of earnings with an indicator variable (POST), which takes a value of 1 if the observation is in the post-adoption period, and 0 otherwise.If the coefficients of the interaction terms are positive, then, the persistence of earnings components improves following the adoption.On the other hand, if those coefficients are negative, then, the persistence of earnings components declines in the post-adoption period.Specifically, the study adopts the following regression specifications, is consistent with the study of Hai (2014) to the test of overall effects of IFRS adoption: To test the effect of IFRS adoption on earnings predictability, the earnings predictability measure is regressed on a set of related variables and control variables to which it runs the panel regression within the sample period.The Pearson correlation matrix is used to determine whether multicollinearity exists among the independent variables.Thereafter, the regression analysis was conducted on the separate models adopted in the study to show the extent and significance of the association.
where IFRSA = variable takes 1 for post-adoption period and 0 for pre adoption period; FSIZE = natural logarithm of bank total assets (measured in millions of naira); LEV = debt of a firm measured as the total liabilities divided by total assets; Ε = error term.
The coefficient of α2 shows the relationship between IFRS adoption and predictive ability of earnings.A positive (negative) estimate indicates that IFRS adoption is associated with lower (higher) earnings predictability.Meanwhile, α3 and α4 are likely to have negative coefficients for which larger banks and higher leverage indicates negative effects on future earnings.Similarly, the study following Oleyeke (2016), Hai (2014) and Dechow et al.
(2010) included another model with which each one of the study variables is interacted with current earnings in order to assess their incremental effect on the link between current earnings and future cash flows.
The control variables are measured as discussed above.
Predictions are similar to those of model 4 for β 1 to β 5 .

Discussion of findings
The data included in the empirical analysis consisted of 44 firm-year observations of 11 banks between 2010 and 2014.Four (4) banks were excluded from the sample size based on the inability to obtain complete data.The data were obtained from audited financial statements that indicated the accounting standards used in its preparation, and the capital market.The data obtained were divided into two, two years before IFRS was adopted (2010 & 2011), and two years after IFRS was adopted (2013 & 2014) excluding 2012 being the year of transition.Source: SPSS computation (2016).
Note: CEARN = current earnings based on total assets; FEARN = future earnings based on total assets; FCFO = future cash flow from operations based on total assets; Control Variables: FSIZE = natural logarithm of total assets; LEV = total liabilities divided by total assets.
Table 1 above shows the summarized statistics of the variables used in this study by distinguishing between the test variables and control variables.Both have been computed separately for banks that reported under previous Nigerian GAAP (PNGAAP) and under IFRS.However, results from the descriptive statistics depict an increase in change in earnings (∆NI), change in cash flow from operations (∆CFO), current earnings (CEARN) and future earnings (FEARN).This is evident in the descriptive mean values of -0.0055 to 0.0121; -0.0865 to 0.0785; 0.0142 to 0.300 and from 0.0202 to 0.0358, respectively.On the other hand, there was a decrease in future cash flow from operations.This is evident in the mean values of 0.0107 to 0.0253.The control variables showed a decline after IFRS adoption in leverage from 0.8418 to 0.7150 and to an increase in firm size from 20.396 to 20.681.The result from the descriptive statistics with variables controlled sends a mixed signal as to whether after IFRS adoption increased earnings predictability or not.This could be as a result of bank's specific characteristics.A more detailed result showing the relationship between each variable, as well as the ability of earnings to predict future accounting values with all the variables controlled is presented in Table 2.  ).However, it is negatively correlated.This outcome implies that banks adoption of IFRS has a negative impact on the future cash flows from operations for the sampled banks in Nigeria.
Similarly, the correlation of current earnings (CEARN) and financial leverage depicts a negative association for both variables, and it is significant to show that as current earnings increases, leverage decreases, while there is a significant positive correlation between firm size and financial leverage.This outcome implies that the larger the bank the higher the leverage ratio.In addition, empirical analysis of the result as depicted in Table 3 indicates a reduction in the ability of current year earnings to predict future earnings and cash flows after IFRS adoption.More so, the adjusted R-square, which indicates the ability of current earnings (CEARN) to predict future earnings (FEARN) reduced from 0.537 to 0.107 in the post-adoption period.Similarly, the ability of current earnings to predict future cash flows reduced from 0.209 to 0.043 after IFRS adoption.This outcome as depicted in Table 3 invariably implies a decrease in the ability of current figures to predict future figures after IFRS adoption.This result also indicates that current earnings in Nigerian banks do not absolutely predict future figures after the adoption of IFRS than they do before the adoption of IFRS.
In summary, findings from this study show a decrease in the ability of current figures (current earnings) in predicting future figures (future earnings and future cash flows) after the adoption of IFRS by Nigerian banks despite the fact that variability in change in net income, as well as variability change in cash flows of banks, increased from negative earnings to positive earnings after the post-adoption period.This could be as a result of banks overreliance on fair value accounting.According to Hai (2014), earnings quality of firms with intense reliance on fair value accounting deteriorates after the adoption of IFRS.Findings here are in tandem with the findings of Abdelmohen and Gehan (2014) where they observed a decline in the predictability of accounting information after IFRS adoption.Also, this result corroborates the findings of Hai (2014) where it was observed that earnings quality of firms with intense reliance on fair value accounting deteriorated after IFRS adoption.However, this outcome could be due to the enforcement mechanisms in the country the work was done.On the other hand, this result contradicts the findings of Atwood et al. (2011), Bahloul and Ben (2014), Hai (2014), Burca and Mates (2015) that found an increase in earnings predictability and cash flow predictability after IFRS adoption suggesting that IFRS has a positive impact on earnings quality.This difference could be accounted for as a result of the difference in the country's legal framework and enforcement mechanisms put in place.

Conclusions and recommendations
The study examined whether the ability of current earnings in predicting future earnings of listed Nigerian banks in NSE improved after the adoption of IFRS.In line with previous studies (Atwood et al., 2011;Hai, 2014;Burca and Mates, 2015), earnings predictability was measured with the adjusted R2 from the regression analysis using current earnings to predict future earnings.The empirical findings show a decrease in earnings predictability after IFRS adoption, thus, indicating that current earnings less predict future earnings and cash flow after the adoption of IFRS.This study, therefore, recommends the need for regulatory bodies, as it relates to the Nigerian banking sector (that is, the Central Bank of Nigeria (CBN), security and exchange commission (SEC), Banks and other Financial Institutions (BOFIA) among others) to put in place measures that will enforce strict adherence to IFRS principles and procedures, as well as enhance investors' protection.

Limitation of study.
Considering only the banking sector in this study is a major limitation.
Hence, this study suggests that future research in this area could address this salient limitation by examining listed firms in other sectors of the Nigerian economy, especially the manufacturing and the agricultural sectors.
Recently, some scholars(Velury & Jenkins, 2006; Dechow et al., 2010; Atwood et al., 2011; Hai, 2014; Mollah, Farooqueb, Mobareka and Molyneuxc, 2015; Uwuigbe, Francis, Uwuigbe and Oyenike, 2016)represented and evaluated earnings quality by the ability of current earning to predict future earnings and future cash flow (earnings predictability and cash flow predictability).According toMollah et al. (2015), the ability of current earnings to predict future cash flows is regarded as cash flow predictability, thus, quality earnings should be highly linked with future cash flows(Atwood et al. 2010).Meanwhile, Oluoch and Gichaiya (2015) indicated higher persistent earnings for medium and larger banks than small banks.Similar results on other firms were reported by Krishnan and Parsons (2008), Hili and Affes (2012), Atwood et al. (2011) that current earnings is a good predictor of future earnings.A major advantage of earnings persistence is linked to the perceptive nature of earnings for valuation purposes.In spite of this, it is largely open to managerial manipulations(Dechow et al., 2010).

Table 1 .
Descriptive statistics of test, earnings predictability proxies and control variables used in the analysis

Table 2 .
Test of correlation matrix of the sampled variables

Table 3 .
Earnings and cash flow predictability