Studi ini bertujuan mempelajari tentang hubungan konservatisma laporan keuangan dengan earning response coefficient dengan menggunakan variabel akrual diskresioner, konservatisme laporan keuangan, dan earning response coefficient. Populasi dalam penelitian ini adalah seluruh perusahaan Bank yang terdaftar di Bursa Efek Indonesia ( BEl) periode 2005-2009. Pemilihan sampel penelitian berdasarkan metode purposive sampling terdapat 13 perusahaan. Penelitian ini menggunakan run test, chi square, dan compare means independent-samples t-test. Pada hipotesis pertama tidak terdapat hubungan yang signifikan antara akrual diskresioner dengan konservatisma laporan keuangan, dengan pengukuran symmetric measures yang ditunjukkan oleh Phi, cramer 's V, dan contingency coefficient sebesar 0,998 dengan probabilitas 0,319. Sedangkan untuk hiotesis kedua dengan menggunakan analisis Independent sample test yang menghasitkan tingkat konsevatisma laporan keuangan tidak dibedakan antara konservatisma yang sifatnya persisten atau permanen.

lbis research examines analysis of studi about the impact companies applying conservatism and more optimistic accounting of earnings response coefficient (ERC). The variabel on this research are discretionary accrual, conservatism level, and earning response coefficient (ERC). The population consists of bank and other then bank companies listed on the Indonesia Stock Exchange during period of 2005-2009. The sample of this research is only bank companies listed on the Indonesia Stock Exchange during period of 2005-2009. Sample members of 13 companies are selected using purposive sampling method. This research used analysis of run test, chi square, and compare means independent-samples t-test The result shown that on first hypothesis companies which no significantly correlation between accrual discretionary and conservative accounting. The correlation value both of variables using symmetric measure by Phi value, Cramer's V, and contingency coefficient value is 0.998 with probability value is 0.319. This probability is indicated that correlation both accrual discretionary and conservative accounting variables is weak. Second hypothesis shown conservative accounting would have not differenciated between permanent and persistency conservative, then earning response coefficient on both variables is no difference. Applying conservative accounting would result in fluctuated earnings and lower earnings predictability which is less beneficial in predicting future earnings, thus ERC would be low.

Kata Kunci : Akrual diskresioner,Konservatisma laporan keuangan,Earning response coefficient