Effects of adopting IFRS 10 and IFRS 11 on consolidated financial statements: an exploratory research

Purpose: This paper investigates how the adoption of IFRS 10 and IFRS 11 affected consolidated financial statements. Specifically, the paper explores whether entities adopted mandatorily or voluntarily both IFRS, whether expressly declared effects, whether considered those effects as material and whether those effects had impacts on selected items of financial statements and on selected financial ratios. Design/methodology/approach: The research is an exploratory study using public entities from France, Germany and the United Kingdom. The majority of the data are manually collected from financial statements. Findings: The results suggest that the adoption of the new IFRS 10 affected the composition of a large number of entity groups but that their financial information and economic-financial indicators do not present material changes. There is also evidence of a large and material impact on the changes in the classification and accounting for interests in arrangements under joint control through the new IFRS 11. The evidence thus suggests unequal effects of the adoption of IFRS 10 and IFRS 11 on the proportion of entities declaring materiality of effects, on the quantitative effects on selected items of financial statements, and on financial ratios. A comparison between the pre-adoption and post-adoption periods reveals that the majority of the effects are driven by the adoption of IFRS 11. Originality/value: This exploratory paper is the first presenting the effectiveness of adopting the most important standards under the “consolidation package” and opens an avenue for future research by academics, for future post-implementation reviews by IASB, and for analysis of peer reviews between accounting practitioners.

The literature surrounding the effects of the adoption of IFRS 10 and IFRS 11 is still scarce. The effects of the adoption of these standards have already been preliminarily assessed in France, Germany and the United Kingdom, but: i) with few entities per country and without disclosing the conclusions for each country (e.g. EFRAG,2012aEFRAG, , 2012bESMA, 2016); ii) for only one industry (e.g. Priscilla and Ariyanto,2014;Demerens et al., 2014;Leitner-Hanetseder and Stockinger, 2014); or iii) based on the annual reports for 2013 (e.g. Mazars, 2014), and the results are still foreseeable (in whole or in part). Moreover, the effects of adopting IFRS 10 in these countries have been evaluated only for some items in the consolidated accounts and not at all in financial ratios (e.g. EFRAG, 2012b;Mazars, 2014;Priscilla and Ariyanto, 2014). On the effects of adopting IFRS 11 in these countries, research is limited to the preadoption period, and focused only on the elimination of the accounting options in the reporting of interests in target agreements of joint control (e.g. Demerens et al.,2014;Leitner-Hanetseder and Stockinger, 2014). Thus, as far as is known, there is still no robust post-adoption analysis (at the level of the global sample, by country and by industry) of the effects of adopting these standards in these three countries.
The main objective of this research is to contribute to the debate surrounding the effects of adopting new standards exploring how the adoption of IFRS 10 and IFRS 11 affected the consolidated financial statements. More precisely, this papers intends: i) to discover whether the new concept of control introduced by IFRS 10 has significantly affected the scope of consolidation, ii) to ascertain whether the classification of, and the accounting for, joint controlled arrangements have significantly changed with the adoption of IFRS 11, iii) to determine if the effects of adopting IFRS 10 and IFRS 11 have significantly affected the amounts of items presented in consolidated financial statements and, then, financial ratios.
Our sample includes consolidated entities from three European countries: France, Germany, and the United Kingdom. These three countries were included in earlier research but mainly with expected (and not observed) effects and with a small number of entities. Our research questions have an exploratory nature. Our main results suggest that the adoption of the new IFRS 10 consolidation model affected the composition of a large number of entity groups but made no statistically material changes in their financial information and economicfinancial indicators. Furthermore, the study also identifies a considerable number of cases with changes in the classification and accounting for interests in arrangements under joint control due to the adoption of IFRS 11 and the effects tend to be statistically material, both in the selected items of consolidated financial statements and in the consequent entities' financial ratios. Altogether, the evidence suggests unequal effects of the adoption of IFRS 10 and IFRS 11 concerning the proportion of entities declaring materiality of those effects as well as the quantitative effects on selected items of financial statements and on financial ratios. Diversity also exists among countries and industries, and the comparison between the pre-adoption and the post-adoption period reveals that the majority of the effects result from IFRS 11 adoption.
This study contributes to the scarce literature on the effects of adopting IFRS 10 and IFRS 11 in several ways. First, as far as we know, it is the first large exploratory study (in the detail of the analysis on the effects of the adoption of the standards and in the sample size), following the standards adoption. Second, it is the first analysis of the real effects in France, Germany, and the United Kingdom through a robust sample covering all Industry Classification Benchmark (ICB) industries and analysing all possible years of adoption of the standard (which was considered a limitation of past studies). Third, the findings show evidence of the effects of the adoption of IFRS 10 and IFRS 11 in a considerable number of cases, but are immaterial in the former and material in the latter, with differences by country and by industry, thereby opening an avenue for future research around the effects of adopting these standards.
The structure of the paper is as follows. Section 2 presents the literature review and the research questions. In Section 3 the research design is detailed. Section 4 offers the results and the discussion. Directions for future research are highlighted in Section 5. Section 6 concludes.

Literature review and research questions 2.1 . Background of the new "Consolidation Package"
In May 2011 the IASB published a new and revised set of standards that is still currently known as "package of five" or "consolidation package" (e.g. BDO, 2013aBDO, , 2013bMazars, 2014;Priscilla and Ariyanto, 2014). The goal was to address the accounting for consolidation, involvements in joint arrangements, and disclosure of involvements with other entities. The "package of five" includes the following: i) IFRS 10, Consolidated Financial Statements (IASB, 2011a), ii) IFRS 11, Joint Arrangements (IASB, 2011b), iii) IFRS 12, Disclosure of Interests in Other Entities (IASB, 2011c), iv) IAS 27, Separate Financial Statements (IASB, 2011d) and v) IAS 28, Investments in Associates and Joint Ventures (IASB, 2011e). The IASB has several publications explaining the reasons for issuing new standards (IFRS 10, 11 and 12), to review standards (IAS 27 and 28), or to supersede standards (IAS 31). A detailed analysis 1 of all the process can be seen in IASB (2011f, 2011g).
All five standards have an effective date for annual periods beginning on or after 1 January 2013 in the original version of each standard. However, the effective date in the 1 The objective of this paper is not to summarize or to present critical perspectives about the new consolidation package, but an understanding of the process between 2003 (when the project was added to the IASB's agenda) until 2011 (when the project was completed), which can be important to regulators, investors, and academia. 5 European Union is after 1 January 2014, with earlier application permitted if all the standards are early applied (EFRAG, 2016).
The theoretical effects of the five can be summarized as follows. The IFRS 10 replaces the "consolidation guidance" that was included in the previous IAS 27 (Consolidated and Separate Financial Statements) (IASB, 2008) and in the previous SIC-12 (Consolidation -Special Purpose Entities) (SIC, 1998). It introduces a single consolidation model for all entities based on control, regardless of the nature of the investee (IASB, 2011a). In practice, the direct effects of adopting the IFRS 10 are analysed through changes in the scope of consolidation and resulting impacts in items of consolidated financial statements.
The IFRS 11 replaces the IAS 31, Interests in Joint Ventures (IASB, 2003a) (superseded). Methods and classifications of joint arrangements are changed, eliminating jointly controlled assets and highlighting the distinction between joint operations and joint ventures (IASB, 2011b). In practice, the direct effects of adopting the IFRS 11 are analysed through changes in the classification and changes in accounting methods, which also impact the items of consolidated financial statements.
The IFRS 12 brings enriched disclosures about both consolidated and unconsolidated entities in which an entity has an involvement. It does not change previous recognition or measurement criteria, but requires higher level of disclosure. Both IAS 27 and IAS 28 still exist but in a revised form to be in accordance with the new IFRS 10, IFRS 11, and IFRS 12.
Changes in recognition and measurement of accounting elements in the consolidated financial statement after adopting the new "consolidation package" arise basically from the adoption of IFRS 10 and of IFRS 11.

. Effects of adopting IFRS 10 and IFRS 11 and research questions
Generally, the IASB will assess the likely effects throughout the development of a new or amended Standard and any revision process should start two years after its effective date (IFRS Foundation, 2016), which is known as Post-implementation Review (PIR). The PIR is the basis for assessing if a standard is achieving its objectives or if it should be revised (Ewert and Wagenhofer, 2012).
The debate around the inclusion of academic research on the PIR process is ongoing, and its advantages (e.g. Barth, 2006Barth, , 2007Barth, , 2008Larson et al., 2011;Ewert and Wagenhofer, 2012;Trombetta et al., 2012) or weaknesses (e.g. Leisenring and Johnson, 1994;Ewert and Wagenhofer, 2012;Trombetta et al., 2012) have been discussed over time. In general, there 6 are a great many researchers defending the benefits for standards boards if they consider the outputs of academic research in the process of amending or creating new standards (e.g. Schipper, 1994;Barth, 2000Barth, , 2006Barth, , 2007Barth, , 2008Fülbier et al., 2009;Larson et al., 2011;Abela and Mora, 2012;Ewert and Wagenhofer, 2012;Trombetta et al., 2012). So, researchers have long been interested in the effects of adopting new accounting standards in several topics from different geographies and with different effects (e.g. Mulford and Quinn, 2008;Mitra and Hossain, 2009;Smith 2009a,2009b;Houmes and Boylan, 2010;Amlie, 2012;Fiechter, 2011;Lopes et al., 2013;Paananen et al., 2012;. The analysis of adopting IFRS 10 and IFRS 11 is an opportunity for research. A prevision for the likely effects of the adoption of the new IFRS 10 and IFRS 11 were anticipated by IASB (2011h, 2011i), based on estimates (ex ante). Effective (ex post) effects (Demerens et al., 2014) are also important, usually supported in academic research. Attention to these topics was called by IASB with the public announcement of the PIR, in which the academic research could be important to take into consideration (IFRS Foundation, 2016), inviting the academia to contribute (IASB, 2011h, 2011i). However, scarce literature is found. Gluzová (2015Gluzová ( , 2016Gluzová ( , 2017 attributes this scarcity to the recent adoption of the consolidation package (becoming mandatory only in 2014 with financial statements published in 2015). In fact, the number of studies that covered effective (ex post) effects of the adoption of both IFRS 10 and IFRS 11 is limited and do not provide a complete picture.
The main objective of this research is to identify genuine (ex post) effects on the financial statements adopting IFRS 10 and IFRS 11. This objective can be divided into more detailed purposes as follows: i) to discover whether the new concept of control introduced by IFRS 10 has significantly affected the scope of consolidation, ii) to ascertain whether the classification of, and the accounting for, joint controlled arrangements have significantly changed with the adoption of IFRS 11, and iii) to determine if the effects of adopting IFRS 10 and IFRS 11 have significantly affected the amounts of items presented in consolidated financial statements and, consequently, financial ratios.
The focus of this study is limited to three countries: France, Germany, and the United Kingdom. These three countries were analysed in earlier research, but with an extremely small number of entities (e.g. EFRAG 2012a(e.g. EFRAG ,2012bESMA, 2016), focused only on a given industry (e.g. Priscilla and Ariyanto, 2014), developed in pre-adoption periods (e.g. Demerens et al., 2014;Leitner-Hanetseder and Stockinger, 2014), or used financial statements from 2013 and, accordingly, some effects were based on estimates (e.g. Mazars, 2014). Consequently, we 7 believe there is still a gap to fill regarding the real effects of IFRS 10 and IFRS 11 that could be of interest to standards boards (e.g. due to the PIR), investors (e.g. to understand changes on the scope of consolidation based on a change in standards and not in investments), and to academia (e.g. due to the complexity and understanding of both standards in practice).
Previous research documented that the majority of entities adopted these standards only when they became mandatory (e.g. Mazars, 2014;Vašek and Gluzová, 2014;Gluzová, 2015Gluzová, , 2017. Vašek and Gluzová (2014) clearly state that the real effects of adopting the package of five can be observed only after 2015. So, our first research question is the following: Ex post effects of adopting the package of consolidation will then be explored. Past reports and research are mainly about ex ante effects (e.g. IASB, 2011h, 2011i;Büdy-Rózsa, 2012;EFRAG, 2012aEFRAG, , 2012bBDO, 2013a;Demerens et al., 2014;ESMA, 2014;Leitner-Hanetseder and Stockinger, 2014) and only very few have covered ex post effects (e.g. Priscilla and Ariyanto, 2014;ESMA, 2016;Gluzová, 2015Gluzová, , 2017. Concerning IFRS 10, previous studies on adopting the standard can be split into two groups, namely, those defending limited and immaterial effects of the adoption (e.g. IASB, 2011h;EFRAG, 2012aEFRAG, , 2012bMazars, 2014;Priscilla and Ariyanto, 2014;Vašek and Gluzová, 2014;Gluzová, 2015Gluzová, , 2017 and those defending material effects (e.g. Büdy-Rózsa, 2012;ESMA, 2014ESMA, , 2016. The IASB (2011h) foresaw that the majority of the entities would have no impacts, notwithstanding the possible new inclusion or exclusion of subsidiaries previously included in the pre-adoption period, especially in complex groups. Vašek and Gluzová (2014) revealed that 10 entities (out of 13) listed on the Prague Stock Exchange (Czeck Republic) did not register or foresee any effect, as IASB (2011h) also predicted. Therefore, our second research question, based on ex post effects, is to identify whether companies expressed the existence of effects of the adoption, namely: RQ2: Did the entities self-declare effects of adopting IFRS 10?
Not just the existence but also the type of effects is of interest. EFRAG (2012a) invited 27 companies from the European Union and Turkey to apply the new IFRS 10 in a sample of their investees in order to identify estimated benefits and challenges, and the majority did not foresee a material change in the items of financial statements. The EFRAG (2012b) also 8 performed a study focusing on expectations of adopting IFRS 10 but in special purpose entities and, in this case, 14 European entities confirmed the expectations of significant effects.
Minor effects were also highlighted by Priscilla and Ariyanto (2014). They sought to assess the real effects of adopting IFRS 10 (and not expectations) but their sample includes only 7 financial services entities listed in three countries (Australia, Austria, and the United Kingdom) from 2012 to 2014. All the entities had effects in the scope of consolidation, but the increase in the amount of total assets, total liabilities, shareholders' equity, and net income was less than 1% and not significant. Limited and not significant effects were also documented by Gluzová, who analysed the scope of consolidation and total assets of 40 entities listed in the Czech Republic and Poland (Gluzová, 2015) and 19 in the Czech Republic (Gluzová, 2017). Results were that only four (in 2015) and two (in 2017) groups presented a change in the number of consolidated subsidiaries, and minor differences in total consolidated assets. Similar findings are observed by Mazars (2014), who analysed a sample of 54 entities of CAC40 and Euro Stoxx50. Concerning IFRS 10, the results highlighted just one entity with changes on their list of subsidiaries and more than 75% of the entities with not significant effects (real or expected) on shareholder's equity and total assets (changes always lower than 1%). Financial services entities were not included in this study, which may explain the results.
The possibility that the adoption of IFRS 10 has had different effects, depending on the industries, has been assumed. In fact, some authors suggest that banks, insurers, and other financial institutions may have been more affected than non-financial entities (e.g. European Commission, 2012;EFRAG, 2012a;Mazars, 2014;Priscilla and Ariyanto, 2014;Vašek and Gluzová, 2014;Gluzová, 2015Gluzová, , 2017. also estimated (in an illustrative example) that the new standards could significantly affect the qualification of an investment as a subsidiary, joint arrangement or an associate, the scope of consolidation of parent companies and the liquidity, profitability and solvency of groups.
Consequently, we can split the effects of adopting IFRS into two direct categories: on the scope of consolidation and on the amounts of elements presented in consolidated financial statements, and we added two research questions: RQ3: Has the scope of consolidation of groups changed with the adoption of IFRS 10?

RQ4: Did the adoption of IFRS 10 have material effects on the amounts of elements presented in consolidated financial statements and in financial ratios?
A similar approach is now applied to IFRS 11. Previous studies on adopting the standard can also be split into two groups, namely, those claiming effects on few arrangements (e.g. IASB, 2011i; BDO, 2013a) and those suggesting material effects on the adoption of IFRS 11 (e.g. Büdy-Rózsa, 2012;EFRAG, 2012a;Demerens et al., 2014;ESMA, 2014ESMA, , 2016Leitner-Hanetseder and Stockinger, 2014;Mazars, 2014). Accordingly:

RQ5: Did the entities declare effects of adopting IFRS 11?
In analysing the expected effects of adopting IFRS 11, the IASB (2011i) predicted that many joint arrangements would not be affected, because: i) most were jointly controlled assets or operations in IAS 31 (2003)  In the process of endorsing the "consolidation package", EFRAG (2012a) also inquired into the benefits and challenges of the requirements of IFRS 11 amongst 26 European Union and Turkey entities, which applied them on a trial basis, and concluded about the existence of estimated effects more pronounced than other organizations and studies. Also, Mazars (2014) assessed more effects of the adoption of IFRS 11 than of the adoption of IFRS 10, surpassing the IASB (2011i) forecast on both. Mazars (2014)

Methodologic approach
Earlier research on the effects of the adoption of IFRS 10 and IFRS 11 is still limited.
We will thus use an explorative methodology (Ryan et al., 2002;Sekaran, 2003;Smith, 2014) in order to examine entities' real effects of adopting IFRS 10 and IFRS 11 on the scope of consolidation, classification of, and accounting methods for reporting joint arrangements, and material changes on items of financial statements and on selected financial ratios. The research questions are presented, but they should be understood as an initial step for future research.
Descriptive and inferential statistics are used. First, univariate descriptive analysis is used to characterize the sample and the type of adoption of the "consolidation package", to observe the number of entities that self-declared effects from moving to IFRS 10, and/or that self-declared effects from adopting IFRS 11. Second, inferential statistics are used to examine the association between some variables and the industry and country [independent test of Chisquared (χ 2 ) and Cramer's V]. We use non-parametric tests, as well as tests to differences of averages or medians for the same group of entities before and after the adoption 3 . In summary, RQ1, RQ2, RQ3, RQ5, and RQ6 required the use of the t-test for one sample, χ 2 and Cramer's V, while Wilcoxon test and t-test for paired samples were used in RQ4 and RQ7.The small number of studies and the scarce theory on the effects of IFRS 10 and IFRS 11 adoption gave us the opportunity to identify a pattern based on our sample and create space for future research to develop deductive hypotheses, as in an exploratory study.

Sample characteristics and data collection
We used a sample of European firms from France, Germany, and the United Kingdom.
We limited the sample to only three countries due to the data-collection cost (e.g. Wieczynska, 2016). We chose these countries since they i) are European Union countries and consolidated financial statements are mandatorily prepared under IFRS, ii) have developed and large stock exchanges (ESMA, 2016), iii) are in different legal regimes and levels of regulatory regime strength (e.g. La Porta et al., 1998;Nobes, 2008) and iv) have been included in earlier research on group entities and consolidation topics (e.g. Lourenço et al., 2012;Lopes et al., 2013).
The initial sample is retrieved from Datastream Worldscope® based on the indexes with the largest entities listed in Euronext Paris (France), in Frankfurt (Germany), and in London (United Kingdom) stock exchanges, namely, SBF 120, HDAX, and FTSE 100. The weight of entities from each country in the total sample is similar (France 120; Germany 110; U.K. 101).
We downloaded the consolidated financial statements prepared by the entities between 2011-2015. Then, we selected those prepared at the end of the fiscal year of the first effective adoption of IFRS 10 and IFRS 11 and matched them with the immediately preceding year (comparative). We hand-collected the data needed.
A content analysis was then made to Notes in order to assess: i) the confirmation that IFRS were used to prepare consolidated financial statements; ii) the date of adoption of the "consolidation package"; (iii) whether they had interests in joint ventures in the last year of use of IAS 31 (2003) and/or interests in joint arrangements in the year of initial application of 12 IFRS 11; iv) information disclosed by the entities on the effects of the adoption of IFRS 10 and IFRS 11; and v) the final balances of selected items at the end of the year before the IFRS 10 and/or IFRS 11, restated. We removed the observations with no sufficient information. The final sample includes 254 entities, totaling 508 reports analysed in the time window of 2011-2015. Table 1, Panel A summarizes the sample selection procedures, and Appendix 1 provides an illustration of the hand-collected data and the interpretation done. Furthermore, Chart 1 is helpful to follow the changes in the sample throughout the successive phases of our study.
( Industrials is the predominant sector in Germany (20 entities, 22.47%) and in France (19 entities, 21.59%) whereas in the U.K. it is the Financials (21 entities, 27.27%). Consumer goods take the second place in France and in the U.K. (18.18%) but in Germany the second most important is Technology (15.73%). Oil & Gas, Telecommunications, and Utilities have the lowest weight. A weak association between the variable "Industry" and the variable "Country" was obtained (χ 2 = 39.284 ** and Cramer's V = 0.278 ** ).

Results and discussion
Our first research question (RQ1) asks whether the entities applied the consolidation package voluntarily (before 2014) or mandatorily (in 2014). Diversity is visible. The first step in answering this RQ was to perform a content analysis to find the date of the adoption of the consolidation package (IFRS 10, 2011, §C2B) and to show if the adoption was taken as voluntary or mandatory. Table 2, Panel A, reports the output. The majority of the entities (188 out of 254, around 74%) waited for 2014, the year of the effective adoption of IFRS 10 and IFRS 11, to mandatorily apply the new version of both standards. The other 65 entities (around 26%) adopted in 2013 and only 1 in 2012. There is sufficient evidence to conclude that the 13 two populations differ significantly in terms of their decision to adopt the "package of five" before (voluntary) or after (mandatory) the effective date (t = 8.710 *** ). (Table 2) Table 2, Panel B, shows the number of entities that mandatorily and voluntarily adopted the consolidation package in each country. More than 80% of French and German entities adopted only in 2014, when mandatory, which is statistically significant when compared with the number of entities who voluntarily adopted in earlier periods (Germany: t = 10.729 *** ; France: t=8.694 *** ). The United Kingdom followed a different pattern with the majority (53%) adopting voluntarily, but no statistical significance in the different proportions of entities adopting voluntarily or mandatorily in this country (t = -0.567). Globally, there is a moderate association between the influence of the country over the adoption of IFRS 10 and IFRS 11 (χ 2 = 42.992 *** and Cramer's V = 0.411 *** ).  Our findings corroborate conclusions of earlier studies (e.g. Mazars, 2014;Vašek and Gluzová, 2014;Gluzová, 2015Gluzová, , 2017, namely, most entities adopted the "package of five" in 2014, when mandatory for European Union countries (Table 2, Panel A). However, our study extends and demonstrates that this behaviour is more pronounced in France and Germany, as opposed to entities from the United Kingdom, where the voluntary adoption stands out (Table   2, Panel B). Thus, it seems pertinent to assess in more detail what the real motivations were behind the choice between early adoption and mandatory adoption of the "package of five".
The second (RQ2) and fifth (RQ5) research questions will be presented simultaneously.
Both RQ question whether or not the entities declared real effects of adopting IFRS 10 and IFRS 11, respectively. Table 3 reveals the output 4 .

(Table 3)
Panel A (B) of Table 3 displays whether the adoption of IFRS 10 (11) was self-declared as having effects or not. Panel A shows that the majority (154 entities, 62.60%) declared the 14 existence of effects, while 92 entities (37.40%) reported no effects of IFRS 10 adoption. Panel B reveals that the majority (140 entities, 79.10%) expressed the existence of effects, while 37 entities (20.90%) revealed no effects from IFRS 11. The number and proportion of entities self-declaring the existence of effects from voluntary adoption of both standards is higher than from mandatory adoption (Panels C and D).
The cases of self-declared effects of the adoption of IFRS 10 and IFRS 11 significantly exceed the cases with no effects (t = 4.077 *** and t = 9.493 *** , respectively). The difference of adopting IFRS 10 is more pronounced in France (t = 2.472 ** ) and in the United Kingdom (t = 4.098 *** ), and both have also showed a greater tendency for the voluntary adoption of the "consolidation package" (especially in the U.K.). German entities follow a different pattern, since no significant differences between existence and non-existence of effects of adopting IFRS 10 are found (t = 0.851). The pattern differs in IFRS 11. The number of entities selfdeclaring effects far exceeds the number of entities not claiming effects in all three countries (Germany: t = 3.112 ** , France: t = 8.157 *** , U.K: t=5.667 *** ). However, results reveal that there is no association between the country and the existence of effects of both standards (IFRS 10: χ 2 = 5.020 and Cramer's V= 0.143; IFRS 11: χ 2 = 4.377 and Cramer's V= 0.158).
Table 4 also details the trend of self-declaring existence or non-existence of effects, but focusing on industry. The majority of Oil & Gas and Basic Materials entities reported no effect with the adoption of IFRS 10 (Panel A), as opposed to all the other industries. These two industries are included in the group with a higher tendency for voluntary adoption of the "consolidation package" (see Panel C, Table 2). On the contrary, Utilities and Financials are the two industries with a higher trend for stating the existence of effects of adopting IFRS 10.
In turn, Table 4, Panel B shows that the number of entities reporting the existence of effects of adopting IFRS 11 is always higher than the number of entities self-declaring no effect.
Furthermore, the pattern of adopting IFRS 11 is again different since all the entities in Telecommunications and Utilities declared the existence of effects, while Financials and Technology are the industries with lower percentage of effects declared (but above 66%).

(Table 4)
Earlier studies suggest that the effects of IFRS 10 adoption were more evident in Financial services (e.g. European Commission, 2012;EFRAG, 2012a;Mazars, 2014;Priscilla and Ariyanto, 2014;Vašek and Gluzová, 2014;Gluzová 2015Gluzová , 2017, but we add also Utilities. Furthermore, our study corroborates earlier studies concluding that entities within Industrials, Basic materials, Utilities and Financial services would declare effects of adopting IFRS 11 (e.g. Leitner-Hanetseder and Stockinger, 2014). We also find that the great majority of entities in Oil & Gas and all in Telecommunications' industries self-stated the existence of effects, as suggested by IASB (2011h), KPMG (2011), and BDO (2013a). Nonetheless, no association was found between the industry and the existence of effects of adopting both standards (IFRS 10: χ 2 = 9.139 and Cramer's V = 0.193; IFRS 11: χ 2 = 7.046 and Cramer's V = 0.200).
After identifying the number of entities self-declaring effects of the adoption (IFRS 10: 154 entities; IFRS 11: 140 entities), the third (RQ3) and sixth (RQ6) research questions seek to present a classification (supported in a content analysis to the Notes) of the direct categories of effects that are possible to identify after the adoption of IFRS 10 and IFRS 11. We classify two direct effects of the adoption of IFRS 10 and four of IFRS 11. This classification is our own (see the example in Appendix 1). The two effects from the comparison between the older and the newer version of IFRS 10 are: i) Effect #1: first time consolidation of a subsidiary not included in consolidation under pre-IFRS; ii) Effect #2: exclusion from consolidation of a subsidiary previously included under pre-IFRS 10. The four direct effects of IFRS 11 are: i) Effect #3: reclassification from jointly controlled operation and/or assets in IAS 31 to joint ventures under IFRS 11; ii) Effect #4: reclassification from jointly controlled entity using equity method in IAS 31 to joint operation under IFRS 11; iii) Effect #5: reclassification from jointly controlled entity using proportionate consolidation in IAS 31 to joint operation under IFRS 11; iv) Effect #6: change of methods from jointly controlled entity using proportionate consolidation in IAS 31 to joint venture using equity method under IFRS 11. However, the number of entities self-declaring the existence of effects is larger than the number of entities disclosing sufficient information that would help users identify the specific effect underlined. Results are presented in Table 5.

(Table 5)
Panel A of Table 5 reveals that only 28 out of the 154 entities (18.18%) reported that after adopting IFRS 10 they included as subsidiaries under control some entities that were not included in consolidation before IFRS 10 (Effect #1), while 13 entities (8.44%) reported loss of control of some subsidiaries after IFRS 10 adoption (Effect #2). A total of 35 entities (22.73%) had changes to the scope of consolidation due to at least one of the analysed effects 5 .
British entities seem to be the ones with more direct effects. Table 5 discloses that Effect #3, Effect #4 and Effect #5 were observed in 2 (15.38%), 11 (19.30%), and 18 (26.09%) entities, respectively. Effect #6 is the most reported, observed in 58 (84.06%) entities. In general, 75 entities (53.57%) revealed at least one of the effects. French entities seem to have more direct effects than the German or British entities.

Panel B of
Our results based on Tables 3 and 5 seem not to support the expectation of IASB (2011g) on limited effects of adopting IFRS 10 and on the lower number of joint arrangements affected by the adoption of IFRS 11 (IASB, 2011h). In fact, a statistically significant and higher number of entities expressly self-stated the existence of effects as compared with the nonexistence (Table 3). However, only a minor number of companies then disclosed clear information to analyse the type of effect on the scope of consolidation and on the classification of joint arrangements. An exception is related with the change from proportional consolidation to equity method in the reclassification to joint ventures (Effect #6), which is the main effect as expected in IASB (2011h), but we can add that the number of reclassifications from joint ventures to joint operations is higher than expected (Effects #4 and #5).
Materiality of the effects of adopting IFRS 10 and IFRS 11 are evaluated through the fourth (RQ4) and seventh (RQ7) research questions, respectively. Usually, companies restate financial information when the effects are considered material. Materiality is, however, subjective in nature, conditional on judgement by preparers of financial statements, who then detail in the Notes the comparatives and the restated information. We do not intend to judge the criteria used by companies in choosing what to restate and disclose. Instead, we assumed that an effect is material if that fact was clearly self-declared by the entity in the Notes of consolidated financial statements or if the entity presented those reports with comparative restatements 6 . So, the answers to RQ4 and to RQ7 are derived in two steps: first, identifying whether those entities declaring the existence of effects classified them as material; second, calculating quantitative effects on selected items of financial statements and on some financial ratios. The results of the first step are presented in Tables 6 and 7, while those of the second step are in Tables 8 to 12. The number of companies in first step is not exactly the same as the second since sometimes the information disclosed was not completely clear and could not be retrieved (reason why the number of companies self-declaring/suggesting the materiality of the effects does not match those disclosing the underlined quantitative impacts -Appendix 2).  A and B) shows that the number of companies declaring the effects of adopting IFRS 10 and IFRS 11 as "material" is lower than the number of companies declaring them as "non-material" (IFRS 10: 26 out of 154; IFRS 11: 56 out of 140). The t-test for the difference on the proportions is statistical significant (IFRS 10: t = -10.935 *** ; IFRS 11: t = -2.407 ** ). This pattern in the number of companies for the pooled sample is similar in each country. The t-test for the difference on proportions is statistically significant for IFRS 10 (Germany: t = -8.883 *** ; France: t = -5.738 *** ; U.K.: t = -5.200 *** ), but not for IFRS 11 (Germany: t = -1.505; France: t = -0.785; U.K.: t = -2.002). Also, material effects of adopting IFRS 10 were more declared in the U.K. and France, as compared to Germany, while German and French entities declared more effects of adopting IFRS 11 than British entities. However, findings do not suggest that the existence of material effects is influenced by country (IFRS 10: χ 2 = 2.145 and Cramer's V = 0.118; IFRS 11: χ 2 = 1.006 and Cramer's V = 0.085).

(Table 6)
Panels C and D reveal the trend of material effects between mandatory and voluntary adopters. The number and the proportion of entities declaring that the effects are not material is greater than those declaring they are material (in both groups).  (Panel B) showing that the number of entities stating material effects is greater than or equal to the number of entities stating no material effects in Utilities (58.33%), Basic materials (56.25%), and Oil & Gas (50.00%). There is a moderate influence of industry in the existence of material effects of adopting IFRS 10 (χ 2 = 27.072 ** ; Cramer's V = 0.419 ** ), but we cannot conclude that the materiality of adopting IFRS 11 is influenced by industry (χ 2 = 9.990; Cramer's V = 0.267).

(Table 7)
In the second step we assess the quantitative effects of the adoption of IFRS 10 and IFRS 11 to answer RQ4 and RQ7, respectively. We do so through the restated information that the entities made using the retrospective restatement on the year preceding the adoption of those standards. In fact, we hand-collected the adjustments (restatements) disclosed by entities self-declaring effects (an example of hand-collected data and the adjusted comparative figures from the old IFRS to the new IFRS is presented in Appendix 1). However, only 85 disclosed restated information for at least one of the accounting items collected and the number of entities were not equal for both standards. We excluded 9 entities reporting monetary effects aggregated for IFRS other than just IFRS 10/11. Only 76 entities are used in further analysis 7 . Table 8 discloses whether or not restatements due to the adoption of IFRS 10 and IFRS 11 in selected items on consolidated financial statements occurred. Column A (B) includes the 20 (51) entities under IFRS 10 (IFRS 11) for which separate data about the presentation of restated information were possible to collect. Column C includes all 76 entities for which either separated or aggregated information was hand-collected. Monetary effects arising from the specific adoption of IFRS 10 are mainly included in the restatement of total assets (90% of the entities), total liabilities (85%), and total cash and cash equivalents (50%). On the specific adoption of IFRS 11, the monetary effects are mostly included in the restatement of total assets (88.24%), total liabilities (86.27%) and revenues (76.47%), and this pattern is similar for the entities adopting at least one of the standards (Column C).

(Table 8)
Additional analysis is performed to see not just if the entity presents monetary effects of the transition but to analyse the magnitude (in euros) of those restatements, by comparing the pre-adoption amounts with the post-adoption amounts. Consequently, we hand-collected the amount of the monetary effects in the accounting items mentioned and disclosed by those 76 entities. The process used to collect the information assures that we capture exactly the effect of adopting IFRS 10 and/or IFRS 11 and that we can isolate it from any other effect (see Appendix 1). The data are presented in Table 9: i) the magnitude of the monetary effects of adopting IFRS 10 (Column A); ii) the magnitude of the monetary effects of adopting IFRS 11 (Column B); and iii) the magnitude of the combined effect of adopting at least one standard (Column C). We present the average and the median of post-adoption amount (as restated, in millions of euros) and of the pre-adoption amount (original, in millions of euros).

(Table 9)
Concerning the adoption of IFRS 10 exclusively, column A of Table 9 shows that the average and the median of the post-adoption amounts are higher than of the pre-adoption amounts for the most of the selected items of consolidated financial statements. Column B displays an opposite pattern followed in the exclusive adoption of IFRS 11, and Column C behaves similarly to column B. In all cases, the standard deviation shows a high level of dispersion in the comparison of the monetary effects (not tabulated).
Next, we perform tests to determine if the differences between post and pre-adoption amounts were statistically significant (regardless of having being considered as material by each entity). Tests to the normality of distributions advises the use of the Wilcoxon nonparametric statistics to compare groups on those 15 selected items. Results are presented in Table 10. Regarding IFRS 10 exclusively (Column A), post-adoption amounts are statistically significantly different (higher) from pre-adoption amounts only in total assets, total liabilities, and earnings before taxes. Concerning IFRS 11 only (Column B), and the effects of at least one of the standards (Column C), results show that post-adoption amounts are statistical significantly different from pre-adoption in the great majority of selected items.
Thus, the effects of adopting IFRS 10 and IFRS 11 have impacts on total assets and total liabilities, but the effects of adopting IFRS 11 are more visible and statistically different in other items of financial statements than those resulting from IFRS 10. (Table 10) The impact of adopting both standards can also be analysed in terms of the materiality effects on some economic and financial ratios. Based on earlier studies that also analysed the effects of adopting these accounting standards on financial indicators (e.g. IASB, 2011i; KPMG, 2011;Demerens et al., 2014;Leitner-Hanetseder and Stockinger, 2014) we selected 9 financial ratios and one indicator for the size of the entities, which is a control variable commonly used in earlier investigations (e.g. Smith, 2009a, 2009b;Houmes and Boylan, 2010;Paananen et al., 2012;Lopes et al., 2013;. Table 11 presents the results (definition of columns is similar to Table 9).
Column A of Table 11

20
Column C reflects the joint effect of adopting at least one standard. The average and median of the majority of the indicators are higher in the post-adoption period. (Table 11) Table 12 presents the results for the Wilcoxon nonparametric test. Concerning IFRS 10, Column A shows that, except for size, the difference between post and pre-adoption amounts are not statistically significant. Related with IFRS 11, Column B reveals that the post-adoption amounts are statistically significantly different from pre-adoption amounts in the great majority of ratios except for return on assets, current liquidity and cash ratio. Post-adoption amounts are statistically significantly higher than pre-adoption amounts in profit margin and financial autonomy, but lower in the return-on-equity, debt-to-equity, debt-to-assets, total asset turnover and size. Finally, Column C, with the joint effect of adopting at least one IFRS, shows that post-adoption amounts are significantly higher than pre-adoption in profit margin, financial autonomy and size, and lower in debt-to-equity, debt-to-assets and total turnover.

(Table 12)
Summarizing the answers to RQ4, only one quarter of the entities declaring effects of IFRS 10 disclosed the adjustments (restatements) to consolidated financial statements of the year previous to the adoption and only half considered it relevant to present the effects of IFRS 10 isolated from any other effect (not tabulated). A small number disclosed those adjustments in all of the selected items. Descriptive and inferential statistics enable us to conclude that the effects of adopting IFRS 10 on selected items of financial statements and financial ratios are generally reduced and not significant. In only four variables (total assets, liabilities, earnings before taxes and size), the impact was significant. Thus, our results in fact appear to corroborate previous studies and reports foreseeing weak and immaterial effects of the initial application of IFRS 10 (e.g. EFRAG, 2012a, 2012b; Mazars, 2014;Priscilla and Ariyanto, 2014;Gluzová, 2015Gluzová, , 2017. The answers to RQ7 are considerably different. More than half of the entities declaring effects of IFRS 11 disclosed the adjustments (restatements) to consolidated financial statements of the year previous to the adoption, and more than 70% considered it relevant to present the effects of IFRS 11 isolated from any other effect (untabulated). The number of entities disclosing effects on selected items is usually higher than the number not disclosing.
Descriptive and inferential statistics enable us to conclude that the effects of adopting IFRS 21 11 in selected items of financial statements and financial ratios are generally stronger and more statistically significant than the effects of IFRS 10. Thus, the analysis of the effect of adopting IFRS 10 and IFRS 11 simultaneously demonstrates that the most important changes to selected items and financial ratios are probably related with effects of IFRS 11 (due to similarities of results compared with effects isolating IFRS 10 and 11). This seems to validate

Directions for future research
The results achieved with the present study suggest several avenues for future research 8 .
We especially wish to point out that IASB will soon begin the Post-implementation Review (PIR) of IFRS 10 and IFRS 11 (including IFRS 12). The PIR usually uses evidence to support the IASB's decisions, giving confidence in making the appropriate decision, from which evidence on diversity in the practice of implementation of new procedures is welcomed. Thus, we suggest a deep analysis on the identification of determinants of voluntary versus mandatory adoption of the "consolidation package".
Notwithstanding the relative small number of companies that adopted these standards before the mandatory date, analyses of firm and country 9 characteristics, as well as incentives and motivations for that behaviour, could be of interest for regulators, standard-setters, and market participants. Another direction would be the identification of the reasons why some firms declare that the effects of applying the new IFRS 10 and IFRS 11 are material, but then do not disclose the amounts separately from simultaneous adoption of other standards, or do not justify why the impacts are material. In a similar approach, investigating whether and why 8 Results could be expanded to include more firms from other countries where IFRS are mandatory since 2005. 9 It may be relevant to assess the characteristics of the country of origin of the entities (including the dynamic continental European countries versus Anglo-Saxon countries) and if these entities anticipated or not effects of the adoption of the five standards. For example, Maroun and Zijl (2016) and Zijl and Maroun (2017), interviewed preparers, audit partners, analysts, and academics in the area of Financial Accounting, examining how the IFRS 10 was interpreted and adopted. The first one supported their results in the institutional isomorphism model of DiMaggio and Powell (1983) to evaluate the influence of normative, coercive and mimetic pressures and the existence of a logic of resistance in the application of the standard. The second one related the Foucault's (1977) power and control theories in the compliance with IFRS 10. The perspective of pressures and resistances used in both studies may be relevant for explaining IFRS 10 adoption and extending then to IFRS 11. all the companies should have impacts in the same direction (e.g. including versus excluding subsidiaries, increasing versus decreasing in assets) would increase the attention to the impacts of future changes on the package of consolidation standards. Quantitative or qualitative research would be appropriate and would help i) standards-setters to identify the appropriate level of disclosures, and ii) market participants to interpret financial statements and restatements.

Conclusion
Given the small and still scarce analysis of the effects of adopting IFRS 10 and IFRS 11, this study is explorative. We illustrate and analyse the real effects on consolidated financial statements with the adoption of those standards, which can be an avenue for future research.
The focus of this study was France, Germany, and the United Kingdom, countries not yet widely analyzed regarding the real effects of the adoption of IFRS 10 and IFRS 11. These countries have many complex groups of entities listed in European stock exchanges applying IFRS, and are countries favourable to the analysis of the topic. An analysis by industry was also considered.
Our results are supported in hand collected information, examining the content of consolidated financial statements in the year of the adoption of the "consolidation package" and the immediate preceding year. Qualitative and quantitative results are thus based on actual disclosures by the entities. Our results on real effects support expectations of earlier studies, since the great majority of entities adopted the IFRS 10 and IFRS 11 only after becoming mandatory in European Union. However, we found the same trend between Germany and France (mandatory adoption) but distinct from the United Kingdom (voluntary adoption), where the proportion of entities with effects from adopting IFRS 10 is greater. Oil & Gas, Basic Materials, and Telecommunications are the industries with a greater tendency for early adoption, but Financial services is the industry most affected by this adoption.
Our findings suggest that the adoption of IFRS 10 affected more cases than the IASB (2011g) predicted, but that these effects assumed minor and not significant impacts. Among the entities adopting IFRS 10, evidence indicates that a significant majority expressly declared the existence of effects but only about 20% considered those effects as being material. In addition, the statistical significance of the effects of the adoption of IFRS 10 (comparing the pre-adoption with the post-adoption amount) was observed in only a minor number of accounting items and financial ratios.
Our research also shows the effects of adopting IFRS 11 on a larger number of arrangements than the IASB (2011h) had anticipated, and tends to be significant. Among the entities adopting IFRS 11, evidence indicates that a significant majority expressly declared the existence of effects and about 40% considered those effects as being material. The statistical significance of the effects of the adoption of IFRS 11 (comparing the pre-adoption with the post-adoption amount) was observed in the majority of the items and financial ratios.
This study corroborates the idea that the abolition of the proportional consolidation and the mandatory use of the equity method would have been the most notable effect of the IFRS 11. However, evidence of jointly controlled entities under IAS 31 reclassified as joint operations under IFRS 11 seems to be potentially higher than that forecasted by the IASB (2011h), which may have contributed to the effects of the adoption of IFRS 11. Our study shows more significant effects in French entities (whose preference for the use of proportional consolidation was known), and in Industrials, Basic Materials, Utilities, and Financial industries, which is in line with expectations of earlier studies.
Our study suffers from some limitations associated with content analysis. The main data are retrieved from information disclosed in consolidated financial statements and handcollected. This information was sometimes not easy to interpret and some assumptions were made. Prudence and conservatism were used, but the effect of the adoption could be different from the one we report. Our suggestions for future research may solve and mitigate some of these concerns.

Appendix 1
We use one of the French entities included in our sample to illustrate how the entities presented their qualitative and quantitative effects of the adoption of IFRS 10 and IFRS 11 and to clarify how we gather the data used in this study. We exemplify the adjusted comparative figures and effects hand-collected for some of the items of consolidated financial statements analysed in our study and demonstrate how we calculated the pre-adoption amounts used to statistically evaluate the material effects of the standards.
The entity "Société Générale", one of the largest financial services groups in Europe, is used to illustrate how the information was hand-collected from its annual financial report of 2014, since the entity adopted the Consolidation Package as of 1 st January 2014, when it became mandatory in the European Union. Figure 1 summarizes this illustration. Information provided by the company is posted, and the analysis made by the authors is presented.

(Figure 1)
Content analysis to the annual consolidated financial statements revealed that the entities do not always seem to be completely clear when reporting the effects of standards adoption, as we would like them to be (worries between what entities declared about the effects of a standard adoption, and how they disclosed those effects). We were not able to hand-collect the impacts from all the entities that self-declared material effects of IFRS 10 and/or IFRS 11 adoption, but we also did not ignore the effects disclosed by companies that considered the effects of the adoption as being non-material. The following cases exemplify:  Some entities declared the existence of material effects of IFRS 10 and/or IFRS 11 adoption, but presented the quantitative impacts together with the effects of changes to other standards or events not contemplated in this study  we were not able to isolate the effects of adopting IFRS 10 and/or IFRS 11  we did not collect the effects.
o Example: Vodafone Group declares: "The previously reported comparative periods have been restated in the consolidated financial statements for the amendments to IAS 19 and IFRS 11"  Effects of IAS 19 and IFRS 11 disclosed together  we classified the effects of adopting IFRS 11 as material but could not obtain the isolated impacts of its adoption.
 Some entities declared immaterial effects of IFRS 10 and/or IFRS 11 adoption, but disclosed those impacts  we considered the effects of the standards as non-material, but hand-collected the effects disclosed to evaluate its materiality from a statistical perspective, which contributes for the difference between the number of material effects and effects hand-collected.
o Example: Veolia Environment declares: "The procedures performed did not identify any material impact of the first-time application of this [IFRS 10] standard"  Effects were disclosed  we hand-collected.
 Some entities suggested a material effect from adopting IFRS 10 and/or IFRS 11, but disclosed the effect together, preventing us from isolating the individual effects of the adoption of each standard we consider the effects of IFRS 10 and/or 11 as material for the purpose of Table 7, but the effects are only presented in column C of Table 8, as no individual effect could be hand-collected for the purpose of columns A or B.
o Example: AXA reports: "The retrospective application of these standards and amendments by AXA resulted in: the change in the consolidation method of a limited number of investment funds and real estate companies (with the full consolidation of some entities previously accounted for under the equity method or not consolidated, and, in contrast, the deconsolidation of others); the change from the proportionate consolidation method to the equity method for joint ventures"  Despite the effects of adopting both IFRS 10 and IFRS 11, the entity presented the restatements in prior consolidated financial statements together  we included in aggregate. -1 -5 -3 -Entities not preparing consolidated financial statements -Entities not using IFRS to prepare their consolidated financial statements = Entities that effectively adopted the new "consolidation package" -3 -1 318 Less: -Entities not indicating the year of the adoption of the new "consolidation package" -Entities applying "consolidation package" prospectively -Entities with lack of consistency in the application of the new standards -10 -2 -8 -Entities created, ceased, or restructured during the sample period -Entities changing the end of reporting period during the sample period -Entities with lack of information         Conclusion: a) Post-adoption amounts are statistically significantly higher than pre-adoption amounts b) Post-adoption amounts are statistically significantly lower than pre-adoption amounts c) There are no statistically significantly differences between post and pre-adoption amounts  Return on Equity (ROE) = Net income/Shareholders' equity; Return on Assets (ROA) = Net income/Assets; Profit Margin = Net income/Revenues; Current liquidity ratio = current assets / current liability; Cash ratio = cash and cash equivalents / current liabilities; Financial autonomy = Shareholders' equity / Total assets; Debt to Equity = Total liabilities / Shareholders' equity; Debt to Assets = Total liabilities / Total assets; Total Asset Turnover = Revenues / Total assets; Size = Log(total assets) Conclusion: a) Post-adoption amounts are statistically significantly higher than pre-adoption amounts b) Post-adoption amounts are statistically significantly lower than pre-adoption amounts c) There are no statistically significantly differences between post and pre-adoption amounts How many selfdeclared the existence of effects (Table 3)  "Following the retrospective application of IFRS 10 Consolidated Financial Statements and IFRS 11 Joint Arrangements, the main changes to the consolidation scope are listed below:  Two securitization special purpose vehicles on behalf of third parties, Antalis SA and Barton Capital LLC, were fully consolidated;  77 entities meeting the definition of joint ventures, previously proportionately consolidated, were retrospectively accounted for using equity method;  2 mortgage financing entities meeting the definition of joint operations, previously proportionately consolidated, were retrospectively consolidated for the assets, liabilities, revenues and expenses relating to the Group's interest in those entities." o IFRS 11 Adoption -1) change from proportionate consolidation for equity method in entities previously classified as jointly controlled entities, but now reclassified as joint ventures; 2) change in the classification of entities previously considered as jointly controlled entities, but now classified as joint operations.


The entity does not indicate whether the effects presented were material. But restated consolidated financial statements of preceding year, both standards contributed to this restatement and revealed the impact of both standards  we considered these effects as material.