Does EVA® beat earnings? Evidence on associations with stock returns and firm values Presented by: Dewi Kartika Sari, Dri Asmawanti.

Presentasi berjudul: "Does EVA® beat earnings? Evidence on associations with stock returns and firm values Presented by: Dewi Kartika Sari, Dri Asmawanti."— Transcript presentasi:

1 Does EVA® beat earnings? Evidence on associations with stock returns and firm values Presented by: Dewi Kartika Sari, Dri Asmawanti

2 Page  2 Agenda of Presentation  Introduction and Motivation  Components of EVA, Hypotheses and Statistical Tests  Sample Selection, Variable Definitions and Descriptive Statistics  Empirical Results  Sensitivity Analyses and Extensions  Summary and Potential Limitations

3 Page  3 INTRODUCTION AND MOTIVATION

4 Page  4 Research Background Hamilton (1777) Marshall (1890) To create wealth, a firm must earn more than its cost of debt and equity capital Used as an internal measure of business- unit performance (Solomons, 1965) Used as an external measure for financial reporting (Anthony, 1973, 1982a, b) Residual Income Stern Stewart & Company  EVA ® “Earnings, EPS, & earnings growth are misleading measures of corporate performance”, “The best practical periodic performance measure is EVA”, Stewart (1991) “EVA is almost 50% better than its closest accounting- based competitor in explaining changes in shareholder wealth”, Stewart 1994. Economic Value Added (EVA ®)

5 Page  5 Research Motivation The claims cited before Interest in EVA in the business press Increasing use of EVA by firms Increasing interest in EVA among academics Potential interest in EVA among accounting policy makers

6 Page  6 Research Questions Do EVA and/or RI dominate currently mandated performance measures, earnings and operating cash flow, in explaining contemporaneous annual stock returns? Q1 Do components unique to EVA and/or RI help explain contemporaneous stock returns beyond that explained by CFO and earnings? Q2

7 Page  7 Research Contribution This study provides evidence that researchers hope will be useful to policy makers who may be interested in EVA or RI as replacements (or complements) to earnings and CFO as key measures of firm performance

8 Page  8 COMPONENTS OF EVA, HYPOTHESES AND STATISTICAL TESTS

9 Page  9 Linkages between operating cash flow, earnings, residual income and EVA

10 Page  10  Linkages between operating cash flow, earnings, residual income and EVA

11 Page  11  Hypotheses Dengan mengasumsikan bahwa pasar ekuitas adalah efisien semi kuat, penelitian ini menggunakan return saham untuk membandingkan kandungan informasi atau value relevance dari CFO, EBEI, RI, dan EVA. Mengikuti Biddle et. al. (1995) penelitian ini membuat perbedaan antara kandungan informasi relatif dan inkremental Meskipun klaim oleh Stern Stewart dan yang lainnya bahwa EVA dan RI lebih value-relevance bagi partisipan pasar daripada EBEI dan CFO, penelitian ini mengambil posisi netral dan melakukan pengujian dua arah atas hipotesis nol bahwa CFO, EBEI, RI dan EVA mempunyai kandungan informasi relatif yang sama: H R : Kandungan informasi ukuran X1 sama dengan X2 Dimana X 1 dan X 2 menyatakan kombinasi pairwise sekumpulan ukuran kinerja yaitu CFO, EBEI, RI, dan EVA. Penolakan H R dipandang sebagai bukti bahwa terdapat perbedaan signifikan dalam kandungan informasi relatif

12 Page  12  Hypotheses Penelitian ini menguji value relevance incremental atas komponen EVA yang diringkas pada Figure 1 dengan menguji hipotesis nol bahwa komponen individu EVA tidak memberikan kandungan informasi incremental diluar komponen lain yang juga terdiri CFO dan EBEI H I : Komponen X 1 tidak memberikan kandungan informasi diluar yang diberikan oleh komponen X 2 – X 5 Dimana X 1 – X 5 adalah komponen EVA (CFO, Akrual, ATInt, Capchg, dan AcctAdj). Penolakan Hi dipandang sebagai bukti adanya kandungan informasi tambahan (inkremental)

13 Page  13  Statistical Tests Pendekatan standar untuk menilai kandungan informasi adalah menguji signifikansi statistik koefisien slope, b1, dengan regresi OLS berikut ini: Karena sedikit diketahui tentang proksi yang sesuai untuk harapan pasar untuk ukuran kinerja selain earnings, penelitian ini menggunakan pendekatan dari Biddle and Seow (1991) dan Biddle et.al. (1995) yang mengestimasikan harapan pasar bersama-sama dengan koefisien slope, yang akan menghasilkan persamaan akhir yang menunjukkan hubungan abnormal return dan lagged ukuran kinerja akuntansi, berikut ini:

14 Page  14  Statistical Tests - Test for Relative Information Content = Untuk menilai kandungan informasi relatif, penelitian ini menggunakan pengujian statistik dari Biddle et.al (1995) yang membuat pengujian hipotesis nol atas tidak adanya perbedaan pada kemampuan dua kumpulan variabel independen yang bersaing untuk menjelaskan variasi dalam variabel dependen. Menggunakan pengujian ini, penelitian ini membuat enam perbandingan regresi pairwise diantara ukuran kinerja akuntansi CFO, EBEI, RI, dan EVA seperti pada persamaan diatas  Statistical Tests - Test for Incremental Information Content Mengikuti metodologi standar (misal Bowen et.al. 1987) kandungan informasi incremental dinilai dengan menguji signifikansi statistik koefisien slope regresi. Khususnya pada persamaan diatas, digeneralisir pada dua ukuran kinerja akuntansi, X dan Y, kandungan informasi incremental dinilai menggunakan uji t pada koefisien individu dan uji F atas hipotesis nol gabungan:

15 Page  15 SAMPLE SELECTION, VARIABLE DEFINITIONS AND DESCRIPTIVE STATISTICS

16 Page  16  Sample Selection Data yang digunakan dalam studi ini dibeli langsung dari Stern Stewart & Co. Data ini meliputi 11 tahun observasi untuk EVA, capital, dan cost of capital perusahaan dengan tahun fiscal Juni 1983 sampai May 1994. Sampel terakhir yang digunakan adalah 6174 tahun pengamatan untuk 773 perusahaan  Dependent Variabel Variabel dependen penelitian ini, market adjusted returns, biasanya digunakan dalam studi kandungan informasi untuk mengukur unexpected returns (misal Biddle et.al 1995 dan Bowen et.al. 1989)

17 Page  17  Independent Variables and Descriptive Data-Relative Informaton Content Tests Empat ukuran kinerja akuntansi dalam pengujian kandungan informasi relatif, CFO, EBEI, RI dan EVA didefinisikan sebagai berikut: CFO: Cashflow dari operasi diperoleh dari laporan cash flow atau laporan perubahan posisi keuangan, tergantung pada tahun pengamatan. EBEI : Laba bersih sebelum extraordinary items. RI : Residual income setara dengan laba ditambah biaya bunga setelah pajak dikurangi biaya semua modal (RI = EBEI + ATInt – CapChg). EVA : Economic value added yang diperoleh dari 1000 database Stern Stewart.

18 Page  18  Independent Variables and Descriptive Data-Incremental Informaton Content Tests Variabel independen dalam pengujian kandungan informasi incremental adalah lima komponen EVA yang diringkas dalam Figure 1: CFO, akrual operasi, biaya bunga setelah pajak, biaya modal, dan penyesuaian akuntansi: Akrual : Akrual operasi didefinisikan sebagai laba dikurangi cash flow operasi (Akrual = EBEI – CFO). Akrual dapat positif atau negatif, tapi lebih mungkin menjadi negatif (merefleksikan biaya non kas seperti depresiasi dan amortisasi). ATInt: Biaya bunga setelah pajak dihitung sebagai 1 dikurangi tingkat pajak perusahaan dikalikan dengan biaya bunga. ATInt adalah non-negatif.  CapChg : Biaya modal didefinisikan sebagai rata-rata tertimbang biaya hutang dan modal ekuitas dikalikan modal awal tahun. Data ini diperoleh dari Stern Stewart. CapChg adalah positif.  AcctAdj : Penyesuaian akuntansi yang merefleksikan penyesuaian tahunan bersih Stern Stewart atas laba dan modal, dan didefinisikan sebagai EVA dikurangi residual income. CapChg dapat positif atau negatif.

19 Page  19 19

20 Page  20 20

21 Page  21 EMPIRICAL RESULTS

22 Page  22 Empirical Result

23 Page  23 Cont’d

24 Page  24 Incremental Information Content Test Venn diagram to sumarize our findings on relative and incremental information content for the four information variables CPF, EBEI, RI and EVA

25 Page  25 SENSITIVITY ANALYSES AND EXTENSIONS

26 Page  26 Sensitivity Analyses and Extensions The sensitivity of the basic results reported above to alternative specifications. 1.Partitioning annual observations into five, non-overlapping, two years test periods 2.Evaluating subsets of firms that claim to use EVA for internal business decisions. 3.Changing the return internal from one-year to five-years 4.Changing the return internal from one-year returns to two- years returns

27 Page  27 Partitioning the sample into sub-periods  In pairwise comparisons of relative information content, adjusted R2 are largest for EBEI in every two-year period.  1984 – 1985 differences between EBEI, EVA and RI are not statistically significant at conventional levels.  EBEI does not outperform RI (P=0.072) but does outperform EVA (0.045)  In incremental content tests, CFO and accrual are significant in every two- year period. Among the remaining EVA components only 1 of 15 f- statistics is significant at the 5% level – acc and in the 1984-1985 sampel period.  Results for the period after release of the 1993 Fortune article again show strong support for the incremental information content of CFO and Accrual, but little evidence for the incremental significance of the remaining EVA components.

28 Page  28 Adopters of ‘EVA-Like’ performance measure

29 Page  29

30 Page  30 Five year return as the dependent variable  The returns internal from one year to five years  In addition, coz five-year data are less sensitive to the choice of expectations models, these tests help address the possibility that the weaker performance of EVA is due to a poorer expectations model.  Sice all ten years of data are used to examine the association between five-year returns and each performance measure, only one test period is reported, Table 6.

31 Page  31

32 Page  32

33 Page  33 Two-year Return as the Dependent Variable Extent th return internal from the one-year contemporaneous period used above to a two-year period that includes both the contemporaneous and subsequent year. Consistent with result table 2 and inconsistent with the conjecture that the market subsequently learns about the importance of EVA, EBEI has significantly higher association with two-year returns (R2 : 4,4 %). Than any of the other three information variables (R2 : range from 2 % to 2,3 %).

34 Page  34 Market value of the firms as the dependent variable

35 Page  35 Overall assessment of the Sensitivity Tests No evidence to support the Stern Stewart claim that EVA (or RI) outperform EBEI. In only one case does EVA /RI have a higher R2 than EBEI and this difference is not statistically significant. R2 is highest for EBEI in the remaining comparisons and EBEI significantly outperform EVA in several sensitivity tests at the 5% level. In terms of incremental information content, the analysies provide only limited evidence that components unique to EVA add to the information set that includes earning and only two of the F-statistics and non of the t statistics are significant at 5%.

36 Page  36 SUMMARY AND POTENTIAL LIMITATIONS

37 Page  37 Summary  There is little evidence to support the Stern Steward claim that EVA is superior to earnings in its association with stock returns or firm values.  In one case does EVA significantly out perform EBEI in tests of relative information content.  Earning outperform EVA  While the charge for capital and Stern Stewart’s adjustments for accounting distortions show some marginal evidence of being incrementally important, this difference does not appear to be aconomically significant.

38 Page  38 Summary Posible reasons : 1.Our design uses current realizations, not future flows, of each performance measure.. 2.Stern Stewart’s estimates of the charge for capital and accounting adjustments may contain measurement error relative to what the market is using to value firms. 3.There exists little or no ‘surprise value’ in components unique to EVA including the capital charge and accounting adjustment. 4.Data needed to compute EVA are not easily estimated and the market does not have these data during our test period. 5.In attempting to approximate economic profit, adjustments made by Stern Stewart may remove accrual that market participants use to infer firms’ future prospects. 6.In violation of our maintained hypothesis of semi-strong market efficiency, the market may have failed to recognize the reporting benefits of EVA through the periode we study, consistent with the notion of earnings myopia.

39 Page  39  Although for some firms EVA may be an effective tool for internal decision making, performance measurement and incentive compensation, it does not dominate earnings in its association with stock market returns for the sample firms and period study.  Our evidence suggests that earnings generally dominates EVA in value-relevance to market participants.  Future research : examing more closely which components of EVA and earnings contribute to, or subtract from, information content.

40 Page  40 - THANK YOU -